Optical line sensor, image processing method, image processing system, and three-dimensional tomographic image forming apparatus

Abstract

To provide a compact optical line sensor having a deep depth of field, an image processing method using the same, an image processing system, and a three-dimensional tomographic image forming apparatus.SOLUTION: A plurality of light-receiving lenses 11 are arranged along a main-scanning direction. A plurality of light-receiving elements are arranged in line along the main-scanning direction to receive the light passing through the light-receiving lenses 11. The light-receiving lenses 11 are separated from each other so as not to overlap fields of view of the light-receiving lenses 11. The light-receiving lenses 11 constitute a telecentric optical system 300 together with an aperture 21. The light-receiving elements arranged in an array constitute a light-receiving element array 120 that forms at least one reading line. In front of light-receiving surfaces of the light-receiving elements, an image intensifier 200 including a photoelectric surface 201, a microchannel plate 202, and a phosphor surface 203 is provided to face them.SELECTED DRAWING: Figure 16A

Description

[Technical field]

[0001] The present invention relates to an optical line sensor that detects scratches and defects on the surface of thin objects to be inspected, such as printed matter or film, and scratches and defects inside transparent films, as well as an image processing method, image processing system, and 3D tomographic image forming device that use the same. [Background technology]

[0002] Contact type optical sensors (hereafter referred to as CIS) that have been used in inspection machines for determining the authenticity of banknotes, etc., and in flatbed scanners such as commercial copy machines and home printer scanners, have been considered for application to so-called surface inspection machines for checking the quality of printed matter, inspecting the surfaces of thin and wide film products in the manufacturing process, and inspecting labels attached to various types of beverage containers, food containers, cans, etc., and some of these have already been commercialized.

[0003] However, the working distance (hereinafter referred to as WD) of CIS that uses SELFOC lenses ("SELFOC" is a registered trademark; the same applies below) is still short, and in order to avoid contact when used in processes, a CIS with a long WD is desired. In addition, while a relatively shallow depth of field was sufficient for the inspection of paper sheets such as banknotes, a CIS with a deep depth of field is also strongly desired in the manufacturing process of the above-mentioned inspection objects because of the large fluctuations in the optical axis direction of the inspection objects.

[0004] A typical CIS with a deep depth of field is a telecentric optical system using a mirror optical system, as shown in Patent Documents 1 to 5. From the above patent documents, it can be seen that the optical system is very complicated. It is very difficult to manufacture the optical system and use it as a product. That is, during manufacturing, the process becomes complicated, and manufacturing stability and cost increase become problems. Even after commercialization, there remain problems such as deviation of the optical axis due to the complex optical system caused by environmental changes and changes over time, and the performance is more likely to deteriorate than in conventional CIS with a simple structure.

[0005] Therefore, it is conceivable to improve the WD and depth of field by using a refractive lens made of glass or resin, instead of using the telecentric reflective optical system. As shown in Patent Document 6 and Patent Document 7, a certain degree of solution has been proposed for the refractive optical system. For example, Patent Document 6 attempts to realize an optical system with a deep depth of field by arranging one telecentric refractive optical system at a distance from a line sensor arranged in a staggered pattern, and arranging the lenses of the refractive optical system at a distance to form an array. In addition, Patent Document 7 considers a method of preventing crosstalk between lenses by providing a partition plate between the separated lenses. In Patent Document 6 and Patent Document 7, the depth of field can be improved and crosstalk between lenses can be prevented, but a normal telecentric refractive optical system is large and it is difficult to make it compact. In addition, the partition plate shown in Patent Document 7 causes missing pixels during reading, resulting in incomplete reading. Furthermore, no solution is shown for the shading that a single lens has in principle, which occurs when the lenses are spaced apart. There is no mention of a method of suppressing so-called ripples in the reading line direction. Moreover, the refractive optical system has not yet been realized.

[0006] Furthermore, inspection machines using camera lenses, such as line cameras, which are different from the above-mentioned method, are large in size, and many of them are required to handle wide inspection objects at manufacturing sites. As a result, the entire device becomes very large and the cost is enormous, making it difficult to install them at each process in a factory.

[0007] In order to solve the above problems, what is needed is an optical line sensor that is small and inexpensive enough to be introduced into each process in a factory, uses a new refractive lens with a long WD and a deep depth of field, and is made up of an illumination system that uses a new method of suppressing ripples, which are optical unevenness on the light-receiving sensor caused by shading from each lens. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] JP 2018-019334 A [Patent Document 2] JP 2018-152713 A [Patent Document 3] JP 2009-244500 A [Patent Document 4] JP 2018-019334 A [Patent Document 5] JP 2018-022948 A [Patent Document 6] JP 2009-246623 A [Patent Document 7] Japanese Patent Application Publication No. 5-14600 Summary of the Invention [Problem to be solved by the invention]

[0009] Therefore, the present invention aims to provide an optical line sensor with a deep depth of field and a compact size, as well as an image processing method, an image processing system, and a three-dimensional tomographic image forming apparatus using the same. Another aim of the present invention is to provide an optical line sensor that can improve the accuracy and precision of detecting foreign objects and defects, as well as an image processing method, an image processing system, and a three-dimensional tomographic image forming apparatus using the same. [Means for solving the problem]

[0010] The optical line sensor according to the present invention is an optical line sensor that reads an inspection object transported in a sub-scanning direction with a reading line extending in a main scanning direction, and includes a plurality of light receiving lenses arranged along the main scanning direction, and a plurality of light receiving elements arranged in a line along the main scanning direction and receiving light transmitted through the plurality of light receiving lenses. The plurality of light receiving lenses are arranged at a distance from each other so as not to overlap with each field of view of the light receiving lenses. The plurality of light receiving lenses, together with an aperture, form a telecentric optical system. The plurality of light receiving elements are arranged in an array to form a light receiving element array that forms at least one row of the reading line. An image intensifier including a photocathode, a microchannel plate, and a phosphor surface is arranged in front of the light receiving surfaces of the plurality of light receiving elements so as to face each other. Effect of the Invention

[0011] According to the present invention, by using a telecentric optical system, it is possible to realize a compact optical line sensor with a deep depth of field. Furthermore, by combining the telecentric optical system with an image intensifier, the accuracy and precision of detecting foreign objects and defects is improved. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a cross-sectional view of a typical CIS. [Diagram 2] FIG. 2 is an exploded perspective view of a line-shaped illumination optical system for a CIS. [Diagram 3] FIG. 13 is a schematic diagram of a light receiving system in which the fields of view of the light receiving lenses do not overlap when each light receiving lens is arranged to function as a monocular lens. [Figure 4] 13 is a schematic diagram showing another example of a light receiving system in which a plurality of light receiving element arrays are arranged. FIG. [Diagram 5] FIG. 13 is a schematic diagram showing yet another example of a light receiving system in which a plurality of light receiving element arrays are arranged. [Figure 6] 1 is a graph showing the relationship between NA and the diffraction limit for each wavelength. [Figure 7A] 1 is a graph showing the relationship between the effective diameter of SELFOC lens A and the diameter of the circle of confusion. [Figure 7B] 1 is a graph showing the relationship between the effective diameter of SELFOC lens B and the diameter of the circle of confusion. [Figure 7C] 1 is a graph showing the relationship between the effective diameter of SELFOC lens C and the diameter of the circle of confusion. [Figure 8A] 1 is a graph showing the relationship between the effective diameter of the rod lens A and the diameter of the circle of confusion. [Figure 8B] 13 is a graph showing the relationship between the effective diameter of rod lens B and the diameter of the circle of confusion. [Figure 8C] 1 is a graph showing the relationship between the effective diameter of the rod lens C and the diameter of the circle of confusion. [Figure 8D] 1 is a graph showing the relationship between the effective diameter of a rod lens D and the diameter of the circle of confusion. [Figure 9A] FIG. 1 is a schematic diagram showing an example of an arrangement method when RGB-LEDs or RGB-LDs (laser diodes; semiconductor lasers) are used as light sources. [Figure 9B] 11 is a side view showing a specific example of an arrangement method when an LD is used as a light source. [Figure 10A] FIG. 13 is a schematic diagram showing the positional relationship between a light source and a light-receiving lens when the light-receiving element array has two rows. [Figure 10B] 1 is a diagram showing the relationship between the light amount distribution of a light source (on an inspection surface) in the main scanning direction, the light amount distribution on a light receiving element surface, and the shading of a light receiving lens. [Figure 10C] This is a graph of the light receiving system MTF for each lens diameter (√A=0.077). [Figure 10D] This is a graph of the light receiving system MTF for each lens diameter (√A=0.154). [Figure 10E] 13 is a graph showing the MTF of a light receiving optical system having a refractive index distribution constant √A of √A=0.1027. [Figure 11] 11 is a graph showing MTF characteristics when defocused. [Figure 12A] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12B] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12C] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12D] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12E] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12F] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12G] FIG. 13 is a schematic diagram showing a modified example of the light receiving system. [Figure 12H] FIG. 2 is a schematic diagram for explaining the position of an optical axis in a light receiving system. [Figure 12I] FIG. 2 is a schematic diagram for explaining the position of an optical axis in a light receiving system. [Figure 13] 13 is a diagram showing the light intensity distribution on the light receiving surface of a light receiving element array when the light intensity distribution of the light source is flat when the lens end is a parallelogram. FIG. [Figure 14] 13 is a diagram showing details of the light intensity distribution in the vicinity of a connection portion between adjacent light receiving lenses. FIG. [Figure 15] In the example of FIG. 12E, the light intensity of the light receiving areas is shown, and the reason why no signal dropout occurs is illustrated diagrammatically. [Figure 16A] 4 is a cross-sectional view of a light receiving optical system (telecentric system) as viewed from the main scanning direction. [Figure 16B] FIG. 2 is a schematic diagram of a microchannel plate having a rectangular cross-sectional shape. [Figure 16C] This is a view of a light-receiving element array (3 lines) superimposed on a microchannel plate as viewed in the direction of the optical axis. [Figure 17] FIG. 2 is a schematic cross-sectional view of the image intensifier as viewed from the sub-scanning direction. [Figure 18] FIG. 2 is a schematic diagram of a pulse delay phase difference detection method. [Figure 19] FIG. 2 is a schematic diagram of a timing chart of a pulse delay phase difference detection method. [Figure 20A] Example of a delay ladder gate circuit (a schematic diagram of a circuit consisting of inductance and resistance with a switch added). (Example of a delay ladder gate circuit (a circuit consisting of inductance and resistance with a switch added) [Figure 20B] FIG. 20B is a diagram showing that the position of the SW in the upper part of FIG. 20A selects the length of the coaxial cable. [Figure 21] FIG. 1 is a schematic diagram of a strip line. [Figure 22A] FIG. 1 is a schematic diagram of a reflective optical system in which a λ / 4 wave plate is placed between a light receiving lens and an image intensifier. [Figure 22B] FIG. 1 is a schematic diagram of a reflective optical system, showing a case where a λ / 4 wave plate is placed between a receiving lens and an object to be inspected. [Figure 22C] FIG. 13 is a schematic diagram of another reflective optical system, showing an object-side telecentric optical system. [Figure 22D] FIG. 1 is a schematic diagram of another reflective optical system, showing a double-telecentric optical system. [Diagram 23] FIG. 1 is a schematic diagram showing a multi-stage telecentric light receiving optical system and an aperture diameter of an earlier stage being smaller than an aperture diameter of a later stage. [Figure 24A] FIG. 2 is a schematic diagram of an optical system arranged in a gantry. [Figure 24B] This is a schematic diagram of the optical system arranged inside the gantry, showing the case of the light receiving optical system in FIG. 25A. [Figure 25A] 1 is a schematic diagram showing the rotation of a pair of multi-row light receiving element arrays and a planar light source within a gantry. [Figure 25B] 1 is a schematic diagram illustrating a case where the main scanning direction of the light receiving element array is parallel to the direction in which the entire optical system rotates inside the gantry. [Figure 25C] 2 is a schematic diagram of a light receiving optical system configured with a light receiving element array and a light receiving lens. [Figure 26A] FIG. 2 is a schematic diagram showing a case where the lenses of a telecentric optical system are removed. [Figure 26B] FIG. 1 is a schematic diagram showing a case where an optical system with a long optical path is made compact by using a multi-reflection mirror. [Figure 27] FIG. 2 is a schematic diagram showing the relationship between the rotation angle pitch and the length of the light-receiving element array, and shows only some of the light-receiving elements and light sources. [Figure 28A]FIG. 1 is a schematic diagram showing a reflective optical system for reflective tomography. [Figure 28B] 1 is a schematic diagram showing a reflective optical system for reflective tomography in which the focal length of the light receiving lens is increased. [Figure 28C] 13 is a schematic diagram showing an optical system capable of receiving a more parallel light beam than other modified examples. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] 1. Overall configuration of optical line sensor A typical CIS is shown in FIG. 1, and a linear illumination optical system for a CIS is similarly shown in FIG. 2. FIG. 1 shows a cross-sectional view of the CIS near the center in the longitudinal direction. Meanwhile, FIG. 2 is an exploded perspective view. The X direction is the main scanning direction, and the Y direction is the sub-scanning direction. The Z direction is perpendicular to the X and Y directions. The linear light source unit 10 is an illumination optical system having a light amount distribution that is elongated in the main scanning direction.

[0014] In the CIS shown in FIG. 1, two housings 16 are arranged opposite to each other with a focal plane (inspection plane) 20 in between. A linear light source unit 10 is provided in each housing 16 for illuminating an object to be inspected on the focal plane 20. A light receiving lens 11 and a light receiving unit 12 are provided in one housing 16, and light from the illuminated object to be inspected is guided to the light receiving unit 12 by the light receiving lens 11. The light receiving lens 11 forms an image of the light from the object to be inspected on the light receiving unit 12. In the CIS shown in FIG. 1, one of the two light source units 10 is arranged on the light receiving unit 12 side and the other is arranged on the opposite side to the light receiving unit 12 side with respect to the focal plane 20.

[0015] The light receiving unit 12 is mounted on a substrate 13 fixed to one of the housings 16. The light passing through the light receiving lens 11 is received by the light receiving surface 12A of the light receiving unit 12, and a signal according to the amount of received light is output from the light receiving unit 12. As the inspection object is transported in one direction Y along the focal plane 20, the light from the inspection object is continuously received by the light receiving unit 12, and an image (such as a color image or a fluorescent image) of the inspection object is obtained based on the output signal from the light receiving unit 12. In this way, the inspection object transported in the sub-scanning direction (Y direction) is read by the light receiving unit 12 extending in the main scanning direction (X direction) at a reading line formed by the light receiving surface 12A of the light receiving unit 12.

[0016] Light B3 emitted from one light source unit 10 passes through a protective glass 14 fixed to a housing 16, is reflected by a reflecting member 17A provided on the inner surface of a protective glass 14A fixed to the other housing 16, and is guided to a focal plane 20. An ultraviolet light blocking filter (UV cut filter) 15 that blocks ultraviolet light from entering the light receiving unit 12 is provided at an arbitrary position between the focal plane 20 and the light receiving unit 12. In addition, a color filter 18 that passes visible light in a specific wavelength range is provided between the light receiving unit 12 and the ultraviolet light blocking filter 15. A substrate 5 for fixing a light source 103 (such as an ultraviolet light source or a visible light source) provided in the light source unit 10 is provided at a position facing the bottom surface of the light source unit 10 in one housing 16.

[0017] In the example shown in Fig. 1 and Fig. 2, the light source unit 10 includes a transparent light guide 101 extending along a longitudinal direction L, a light source 103 provided near one end face in the longitudinal direction L, and a cover member 102 for holding each side face of the light guide 101. The light emitted from the light source 103 enters the light guide 101, and while propagating through the light guide 101, is appropriately reflected by the light diffusion pattern P, and is emitted from the light emission surface in the direction of the arrow, becoming a line-shaped illumination light to illuminate the inspection object. The depth of field of such a CIS is shallow, and if the inspection object has a thickness, it is difficult to inspect the entire thickness direction, and since the WD is narrow, the inspection object itself is often not performed due to contact with the inspection object.

[0018] In the above-mentioned CIS, for example, a SELFOC (manufactured by Nippon Sheet Glass Co., Ltd.) lens array is used as the light receiving lens 11. The SELFOC lens array is an erect life-size lens array. In this lens array, cylindrical SELFOC lenses are stacked in a bale to form a multi-lens. The advantage of a multi-lens is that it is possible to make the so-called brightness of the lens brighter than that of a single lens. In other words, the F-number of a multi-lens made by arranging multiple single lenses is smaller than the F-number of a single lens. This is because the effective F-number becomes smaller at the point where the focal position of one lens at any position coincides with the focal positions of the lenses around the lens. In other words, in an erect lens system, the numerical aperture (hereinafter referred to as NA) is larger in an array than in a single lens. This property is the main reason why a SELFOC lens array is used in a CIS.

[0019] The advantages of the CIS as described above are disadvantageous in terms of the depth of field and the depth of focus. As is the case with monocular lenses, the larger the numerical aperture, the shallower the depth of field. For example, it is well known that in microscope objective lenses, the greater the magnification, i.e., the larger the NA, the shallower the depth of field. In addition, in camera lenses, the difference in the depth of field between distant and close views is clearly shown, and the aperture is adjusted to ensure the depth of field. In other words, the NA is changed to obtain the desired depth of field. In addition, in erect multi-lens lenses, such as SELFOC lenses, the optical axes of the lenses are different and intersect, so that compared to monocular lenses, the image is more likely to become blurred when the object to be inspected changes in the optical axis direction. The above is a major disadvantage of the multi-lens SELFOC lens array of the bale-stacking method. Therefore, we have examined how to deepen the depth of field of a compact optical line sensor, and the following examples are obtained. In the following embodiment, the light receiving lens 11 constitutes a telecentric optical system.

[0020] 2. Example of light receiving system The first method is to make the optical line sensor into an array structure that can be regarded as a single lens, as shown in Fig. 3. Fig. 3 is a schematic diagram of a light receiving system in which the fields of view of the light receiving lenses 11 do not overlap. In Fig. 3, the light receiving lenses 11 are arranged at intervals in the main scanning direction (X direction) and also at intervals in the sub-scanning direction (Y direction) so that the fields of view of the light receiving lenses 11 do not overlap, so that the light receiving lenses 11 are arranged in a staggered pattern.

[0021] That is, instead of the bale-stacking method, a plurality of light receiving lenses 11 are arranged along the main scanning direction (X direction) at a distance from each other. A plurality of light receiving lenses 11 arranged along the main scanning direction (X direction) are integrally held by a lens holder 110. A light receiving element array 120 is arranged at a position facing each light receiving lens 11 in the Z direction, the light receiving element array 120 being configured by arranging a plurality of light receiving elements (not shown) in a line along the main scanning direction (X direction). That is, a single light receiving element array 120 is configured by arranging a plurality of light receiving elements in an array along the main scanning direction (X direction). Each light receiving element receives light that has passed through each light receiving lens 11.

[0022] In this example, a light receiving element array 120 is arranged in correspondence with each light receiving lens 11. As a result, each light receiving element array 120 made of a short sensor is arranged in a staggered manner along the main scanning direction (X direction). A plurality of light receiving element arrays 120 arranged along the main scanning direction (X direction) form one reading line L, and in the example of FIG. 3, two reading lines L are formed. The lens holder 110 is not limited to a configuration in which it is provided in correspondence with each reading line L, and may be configured to integrally hold a plurality of light receiving lenses 11 corresponding to each reading line L with one lens holder.

[0023] As shown in FIG. 3, one light receiving lens 11 may be provided for each light receiving element array 120, and a plurality of light receiving lenses 11 may be arranged in number corresponding to the plurality of light receiving element arrays 120. The optical axis of the light transmitted through each light receiving lens 11 and guided to each light receiving element array 120 may pass through substantially the center of the main scanning direction (X direction) of each corresponding light receiving element array 120 on a one-to-one basis. In this method, a plurality of light receiving element arrays 120 are arranged in a plurality of columns in the sub-scanning direction (Y direction). That is, the plurality of light receiving element arrays 120 are arranged at intervals in the direction (Y direction) perpendicular to the arrangement direction (X direction) of the light receiving elements.

[0024] The width W1 of each light receiving lens 11 in the sub-scanning direction is smaller than the width W2 (lens diameter) in the main scanning direction. That is, each light receiving lens 11 has an elongated shape along the main scanning direction. The width W1 of each light receiving lens 11 in the sub-scanning direction corresponds to the field of view of each light receiving lens 11 in the sub-scanning direction. Also, the width W2 of each light receiving lens 11 in the main scanning direction corresponds to the field of view of each light receiving lens 11 in the main scanning direction. It is preferable that the width W1 of each light receiving lens 11 in the sub-scanning direction is set so as to satisfy 0.005 < N.A. < 0.03. In this example, each light receiving lens 11 has the same shape and is formed in a rectangular shape when viewed from the direction (Z direction) orthogonal to the main scanning direction and the sub-scanning direction. However, each light receiving lens 11 is not limited to a rectangle (rectangular shape), and may be an oval or an ellipse, or may have other shapes.

[0025] The multiple light receiving lenses 11 are arranged at intervals of less than the width W2 of the light receiving lens 11 in the main scanning direction. That is, it is preferable that the multiple light receiving lenses 11 are arranged at intervals of less than the field of view dimension (within the field of view range) of the light receiving lens 11 in the main scanning direction. As shown in the example of FIG. 3, the fields of view of each light receiving lens 11 may be overlapped in the sub-scanning direction. In this case, for the light receiving elements in the overlapping portion of the fields of view of the multiple light receiving lenses 11, the pixel output from the light receiving elements may be subtracted. For example, the image of one light receiving lens 11 (the amount of light received that has passed through one light receiving lens 11) may be excluded from the data output from the light receiving elements, or the pixel output from the light receiving elements may be set to approximately half the output value when the images are synthesized. By using multiple light receiving element rows (light receiving element array 120), the occurrence of pixel missing can be prevented more reliably than in the case of a single line of light receiving element rows.

[0026] Fig. 4 is a schematic diagram showing another example of a light receiving system in which multiple light receiving element arrays 120 are arranged. In the example of Fig. 4, each light receiving lens 11 does not correspond one-to-one to each light receiving element array 120, but multiple light receiving lenses 11 (two in this example) lined up in the main scanning direction correspond to one light receiving element array 120.

[0027] The multiple light receiving lenses 11 corresponding to one light receiving element array 120 are adjacent to each other in the main scanning direction. However, the multiple light receiving lenses 11 corresponding to one light receiving element array 120 may be spaced apart from each other, in which case they may be spaced apart by less than the width W2 in the main scanning direction of the light receiving lenses 11. A light blocking member may be provided between each of the light receiving lenses 11.

[0028] Fig. 5 is a schematic diagram showing yet another example of a light receiving system in which a plurality of light receiving element arrays 120 are arranged. In Fig. 5, a plurality of light receiving element arrays 120 (two in this example) each consisting of a long sensor of the same length (a length corresponding to the entire length in the main scanning direction) are arranged in parallel in the sub-scanning direction. As shown in Fig. 5, by associating one light receiving element array 120 with a plurality of light receiving lenses 11 arranged in the main scanning direction (X direction), a plurality of light receiving element arrays 120 may be arranged in the same number as the number of rows of light receiving lenses 11 in the sub-scanning direction (Y direction).

[0029] 3 to 5, since the width W1 of the light receiving lens 11 in the sub-scanning direction is smaller than the width W2 in the main scanning direction, the light receiving lens 11 can be arranged closer in the sub-scanning direction, and as a result, the optical line sensor can be made compact. As described above, the short light receiving element arrays 120 may be used in a staggered arrangement (see FIGS. 3 and 4), or two rows of light receiving element arrays 120 may be used spaced apart (see FIG. 5), but this is not limiting, and more light receiving element arrays 120 may be arranged spaced apart in the sub-scanning direction (Y direction).

[0030] 3.Long focal length of receiving lens Next, we will discuss the long focal length of the receiving lens. Conventional SELFOC lenses have placed emphasis on making the CIS compact and reducing costs, and lenses with shorter conjugate lengths have been required. However, this trend has contributed to reducing the allowable depth of field. Moreover, the lens diameter is becoming smaller and smaller. When making the receiving lens longer focal length, if a conventional receiving lens is used, the NA becomes extremely small. Therefore, the effect of diffraction becomes greater, and the blur due to the diffraction limit becomes the dominant factor in the deterioration of optical resolution, rather than the blur due to the geometric optical aberration of the receiving lens itself. Conventional CIS has a large NA, so it was possible to ignore the occurrence of image blur due to the diffraction limit. However, in order to increase the WD, it is necessary to extend the focal length of the receiving lens, which means that the NA becomes smaller. Therefore, with conventional lens diameters, the effect of diffraction increases as the focal length increases. In this embodiment, a method is proposed in which the lens diameter is increased to lengthen the WD, thereby preventing degradation of the optical resolution even when image blurring due to the diffraction limit is reduced.

[0031] Abbe's diffraction limit d is inversely proportional to the numerical aperture NA. Since the optical system is in air, the following formula 1 holds true using the wavelength λ in air. d=λ / NA (Equation 1) The relationship between the NA and the diffraction limit for each wavelength is shown in Figure 6. For light-receiving lens 11 with the same lens parameters, shortening the so-called pitch of light-receiving lens 11 itself extends the focal length and reduces the effects of aberration.

[0032] From the above, it is clear that in order to make the receiving lens 11 have a long focal length, the lens diameter must be made larger. If the NA is kept the same, the effects of diffraction can be made equivalent to that of a receiving lens 11 with a short focal length. However, increasing the lens diameter increases the geometric optical aberration. Therefore, for receiving lenses 11 with different lens parameters, it is necessary to consider the minimum circle of confusion diameter when the lens diameter is increased. The wavelength λ was set to λ = 630 nm, which is the largest diffraction limit diameter.

[0033] As a result of investigations by the inventors of the present application, it was found that it is sufficient to consider the relationship of the minimum circle of confusion to each focal length of a certain light receiving lens 11. For example, a case where the focal length f is 50 mm is shown in Figs. 7A to 7C. Here, when three types of SELFOC lenses (SELFOC lens A, SELFOC lens B, and SELFOC lens C) are used as the light receiving lens 11, Fig. 7A shows the relationship between the effective diameter and the circle of confusion diameter of SELFOC lens A, Fig. 7B shows the relationship between the effective diameter and the circle of confusion diameter of SELFOC lens B, and Fig. 7C shows the relationship between the effective diameter and the circle of confusion diameter of SELFOC lens C. In Figs. 7A to 7C, the solid line shows the total circle of confusion, the dashed line shows the circle of confusion due to diffraction, and the dashed-dotted line shows the circle of confusion due to geometrical optics.

[0034] 6 and 7A show the relationship between the minimum circle of confusion and the diffraction limit, i.e., the optical resolution for a certain lens diameter and focal length. Therefore, in the case of the light receiving lens 11 shown in Fig. 7A, the larger the effective diameter Φ, the smaller the circle of confusion diameter becomes, and it is sufficient for the effective diameter Φ to be 1.0 mm ≦ Φ ≦ 3.0 mm.

[0035] On the other hand, according to Figs. 6 and 7B, the geometrical optical circle of confusion is large and the degree of dependence on diffraction is small, so the circle of confusion diameter is smallest when Φ=1.0 mm. Moreover, even when Φ=1.0 mm, the circle of confusion diameter is nearly twice as large as that of the light receiving lens 11 shown in Fig. 7A. The light receiving lens 11 shown in Fig. 7A is a SELFOC lens with smaller aberration and larger effective diameter than the light receiving lens 11 shown in Fig. 7B, and it is clear that the light receiving lens 11 shown in Fig. 7A should be selected. Furthermore, the light receiving lens 11 in Fig. 7A, which has the same focal length, can have an NA at least three times larger than that of the light receiving lens 11 in Fig. 7B, i.e., the amount of light received is nine times or more, and therefore the output of the light receiving element is also nine times or more. Accordingly, the shot noise dependent on the amount of light received by the light receiving element is also reduced to one-third, so the light receiving lens 11 shown in Fig. 7A is preferable from the viewpoint of noise suppression. In addition, it can also be said that, if the same amount of noise is allowed, the light receiving lens 11 in FIG. 7A can improve the scanning speed by nine times compared to the light receiving lens 11 in FIG. 7B.

[0036] According to FIG. 7C, the SELFOC lens C, like the SELFOC lens A, also has little aberration and can have a large effective diameter.

[0037] Next, the parameters of SELFOC lenses A to C shown in Figures 7A to 7C are shown in Table 1 below. The most important parameter in Table 1 is the refractive index distribution constant. The light-receiving lens 11 with least aberration when the effective diameter is enlarged and the focal length is extended is SELFOC lens A, which has the smallest refractive index distribution constant, and the light-receiving lens 11 with the next least aberration is SELFOC lens C. It goes without saying that to aim for high-resolution, high-speed inspection, a light-receiving lens 11 with a large effective diameter, brightness, and little aberration is preferable. [Table 1]

[0038] Furthermore, when four types of plastic rod lenses (plastic refractive index distribution type lenses) are used as the light receiving lens 11, FIG. 8A shows the relationship between the effective diameter and the diameter of the circle of confusion of rod lens A, FIG. 8B shows the relationship between the effective diameter and the diameter of the circle of confusion of rod lens B, FIG. 8C shows the relationship between the effective diameter and the diameter of the circle of confusion of rod lens C, and FIG. 8D shows the relationship between the effective diameter and the diameter of the circle of confusion of rod lens D. In FIG. 8A to FIG. 8D, the solid line shows the total circle of confusion, the broken line shows the circle of confusion due to diffraction, and the dashed line shows the circle of confusion due to geometrical optics. In addition, the parameters of rod lenses A to D shown in FIG. 8A to FIG. 8D are shown in Table 2 below. It can be seen that the plastic rod lenses have the same tendency as the SELFOC lens. Considering the refractive index of the plastic rod lens and the refractive index of the glass lens, the on-axis refractive index is preferably about 1.45 to about 1.65. [Table 2]

[0039] From the above, it can be seen that the refractive index distribution constant is the dominant factor of aberration. In an ideal refractive index distribution lens, the more gradually the refractive index changes, the less the aberration. This is similar to how a sudden change in angle causes aberration even in a normal spherical lens. A sudden change in angle means an increase in the high-order nonlinear effect when Snell'Law is expanded into a polynomial. In other words, the aberration increases because the deviation from paraxial optics increases. The present inventors have found that in order to achieve a pixel resolution of 400 dpi or more when the focal length or WD is approximately 50 mm or more and the effective diameter Φ is approximately Φ≧1.0 mm, it is preferable to set the refractive index distribution constant to 0.12 or less.

[0040] 4. Modifications of the light receiving lens The light receiving lens 11 in the present invention is not limited to gradient index lenses such as SELFOC lenses and plastic rod lenses, but may be other lenses, for example, achromats, apochromats, etc., that have the same aberrations due to nonlinear effects as the gradient index lenses, i.e., have the same spherical aberration, coma aberration, and astigmatism, or a telecentric refractive optical system may be used in place of gradient index lenses such as SELFOC lenses and plastic rod lenses that have the same aberrations due to nonlinear effects and diffraction limits as the gradient index lenses, with similar arrangement and dimensions (aperture). This also applies to the light receiving lens 11 that forms an inverted image, which will be described later.

[0041] The optical system described above is a case where the erect lens is the center, but it may be an inverted optical system when the fields of view do not overlap. That is, a configuration in which multiple light receiving lenses 11 form an inverted image may be used. If the lens array is a two-row type, an inverted optical system can also be used. In the case of an inverted optical system, since the image is inverted symmetrical about the optical axis, the inverted image may be converted into an erect image by image processing when synthesizing the images. That is, the inverted images of the multiple light receiving lenses 11 may be inverted and converted into an erect image, and then image synthesis processing may be performed. In addition, during the operation, the necessity or non-necessity of the overlapping parts may be determined and corrected from the correction algorithm, and the image may be converted into an erect image based on the determined relationship between pixels. Alternatively, in an inspection in which an image is not constructed, since it is only necessary to detect scratches or defects, image synthesis and image processing are not necessary, and the detection parts on the inspection surface may overlap. In the case of overlapping, the position is corrected in advance using a correction chart.

[0042] Furthermore, in the case of an inverted refractive optical system, in the signal processing for each light receiving element, for example, data obtained from one of two rows of light receiving element arrays arranged in a staggered manner so as to be spaced apart in the sub-scanning direction may be obtained as a longer data, and data obtained from the other light receiving element array may be obtained as a shorter data, and the obtained images may be inverted to erect images before being combined into images. Alternatively, after converting the inverted image data of each light receiving element into an erect image, a correction coefficient may be multiplied or subtracted from the overlapping parts when combining the images.

[0043] Specifically, in the inverted refractive optical system, the multiple light receiving element arrays may be light receiving element arrays arranged in two reading lines, each of which is shorter than the reading line. Also, the light receiving element array arranged in one reading line and the light receiving element array arranged in the other reading line may be arranged in a staggered pattern alternately along the main scanning direction. This configuration is similar to the erect refractive optical system described in FIG. 3, so a detailed description will be omitted.

[0044] 3, the multiple light receiving elements are arranged in an array of two or more rows to form multiple light receiving element arrays 120. The multiple light receiving element arrays 120 are arranged spaced apart from each other by less than the width W2 in the main scanning direction of the light receiving lens 11 in a direction perpendicular to the reading line L. The multiple light receiving lenses 11 are arranged in a number corresponding to the multiple light receiving element arrays 120, and the optical axis of light transmitted through each light receiving lens 11 and guided to each light receiving element array 120 passes through approximately the center of each light receiving element array 120.

[0045] 5. Configuration of the illumination optical system In this embodiment, the focal length f of the light receiving lens 11 is f=50 mm, the NA is NA=0.01, 0.02, 0.025, 0.03, and the refractive index distribution constant √A is √A=0.077. As for the light source 103, since the WD is 10 times longer than that of the conventional CIS, the inspection surface illuminance needs to be 100 times or more in the same magnification system. Therefore, for example, a high-luminance white LED array is used as the light source 103. That is, the multiple light sources 103 may be configured to include white LEDs. When a semiconductor laser in the visible range is used as the light source 103, the emitted beam is expanded in the main scanning direction and collimated in the sub-scanning direction to reduce unevenness in the amount of light during irradiation.

[0046] FIG. 9A shows an example of an arrangement method when an RGB-LED or an RGB-LD (laser diode; semiconductor laser) is used as the light source 103. In this way, the multiple light sources 103 may be configured to include a red LED (R), a green LED (G), and a blue LED (B), or may be configured to include a laser diode. In FIG. 9A, the multiple light sources 103 are mounted on a light source board 134, and a heat sink 135 is attached to the light source board 134. The beam emitted from each light source 103 is collimated by an ellipsoidal condenser lens 104 having different lens powers in the main scanning direction and the sub-scanning direction, and is irradiated onto the inspection target. Here, an ellipsoidal condenser lens 104 is shown, but any lens may be used as long as its lens power is appropriately different in the main scanning direction and the sub-scanning direction. The power of a lens is the reciprocal of the focal length and is a measure of the refractive power of the lens.

[0047] Alternatively, as for the LD, if an edge-emitting LD in which the divergence angle of the emitted beam of the LD itself is different in the horizontal and vertical directions is used, a normal collimator lens may be used. FIG. 9B is a side view showing a specific example of an arrangement method when the LD is used as the light source 103. In this case, the LD with the larger divergence angle is arranged parallel to the main scanning direction. The beams emitted from the red LD 131, green LD 132, and blue LD 133 constituting the light source 103 are collimated by the condenser lenses 104 corresponding to each, and then narrowed down to the inspection surface by the cylindrical lens 105. In this way, each color of RGB is irradiated at approximately the same position in the sub-scanning direction, making it possible to reduce color unevenness in the sub-scanning direction. Meanwhile, the light irradiated by the light source 103 such as an LED or LD and diffusely reflected on the inspection surface is imaged on the light receiving element array by the light receiving lens system. In the life-size optical system, the light receiving element has an element size of 62 μm equivalent to 400 dpi to 42 μm equivalent to 600 dpi. When an element size of 600 dpi or more is used, the power of the illumination light can be increased accordingly. In this manner, the multiple light sources 103 may include light sources 131-133 having multiple different wavelengths, and the light sources 131-133 may be regarded as one unit, with multiple light sources 103 of one unit being arranged in the main scanning direction (X direction).

[0048] Furthermore, even if the radiance per unit area is the same, if the size of the light receiving element is different, it is synonymous with the area of ​​the light emitting part being different, so the amount of light received will decrease inversely proportional to the square of the light receiving element area. The illumination light amount is determined by taking the above into consideration. If the element size becomes smaller, the amount of light received will decrease inversely proportional to the element area for the same accumulation time. This is a physical property of semiconductor light receiving elements, and in order to maintain the S / N ratio, it is sufficient to increase the illumination light, increase the power density, and make the number of electrons generated per unit time the same. This allows the shot noise to be maintained at the same level as that of a larger size light receiving element even if the size of the light receiving element is reduced. In addition, the light received by the light receiving element depends not only on the illuminance of the inspection surface, but also on the range that the light receiving element is looking at when receiving light. Needless to say, if the solid angle of reception is different and the angle decreases, the amount of light received will also decrease, and if the inspection surface is a perfect diffusion surface, the amount of light received will comply with the so-called cosθ law. The light diffused and reflected from the inspection surface is captured by the light receiving lens and collected on the light receiving element, which outputs an output signal. The output signal from the light receiving element array is also split from serial to parallel to increase the processing speed and is transmitted to the image processing device.

[0049] FIG. 10A is a schematic diagram showing the positional relationship between the light source 103 and the light receiving lens 11, in the case where the light receiving element array 120 has two rows. FIG. 10A shows the arrangement of the light source 103 relative to the light receiving lenses 11 and the light receiving element array 120 arranged in two rows. In this embodiment, the light source 103 is arranged in the approximate center of the two rows of reading lines L. The light sources 103 are arranged parallel to the reading lines L. The optical axis of each light source 103 is arranged at an arbitrary position on a virtual plane that connects an intersection point with the optical axis of the light transmitted through each light receiving lens 11 and guided to each light receiving element array 120, and intersects with the optical axis of the light transmitted through each light receiving lens 11 and guided to each light receiving element array 120. In this embodiment, the focal length f of the light receiving lens 11 is f=50 mm, and the lens diameter Φ is Φ=4 mm (however, taking into account geometrical optical aberration, the effective diameter Φ' is set to Φ'=2.5 mm using an aperture). In this embodiment, light is irradiated from the light source 103 to the inspection surface at an angle of about 45 degrees. Therefore, the distance between the light source 103 and the inspection surface is set to 70 mm, taking into consideration the working distance of the light receiving lens 11. In this embodiment, an LD is used, but an LED of each wavelength may also be used.

[0050] The arrangement pitch of the light sources 103 is adjusted to the distance between the lenses, and the light sources 103 are arranged at a position that is approximately the center between adjacent light receiving lenses 11 and also approximately the center of two rows of reading lines L. This makes it possible to better prevent missing pixels on the inspection surface, and also to further reduce shading of the light receiving lenses 11 and unevenness in the amount of light from the light sources 103. The lens pitch between the light receiving lenses 11 in each row is set to 7 mm. In this case, the effective lens pitch is 3.5 mm.

[0051] Fig. 10B is a diagram showing the relationship between the light amount distribution of light source 103 (on the inspection surface) in the main scanning direction, the light amount distribution on the light receiving element surface, and the shading of light receiving lens 11. In the layout of Fig. 10A, the light amount distribution on light receiving element array 120 becomes flat as shown in Fig. 10B, making it possible to effectively use the dynamic range of the light receiving element.

[0052] 6. Ripple suppression method Next, shading caused by each light receiving lens 11 generates ripples in the direction of the reading line L, which in turn leads to a narrowing of the dynamic range of the light receiving element. A method for suppressing this will be described.

[0053] First, the illumination system irradiates the inspection object with illumination light having a negative intensity distribution in advance. For example, a method of arranging the light source 103 between adjacent light receiving lenses 11 does not provide a shading suppression method that is completely compatible with various light receiving lenses 11. That is, each light receiving lens 11 has its own shading, and if the light receiving lens 11 is different, an illumination optical system corresponding to the shading must be used. In this embodiment, a condenser lens 104 having a large power in the main scanning direction is arranged. The light beam transmitted through the condenser lens 104 is then guided to the inspection object by a converging lens such as a cylindrical lens 105, and a light intensity corresponding to the shading of the light receiving lens 11 is formed on the inspection object. Then, by appropriately changing the position of the condenser lens 104 in the optical axis direction according to the shading of the light receiving lens 11, for example, if the inspection object is a white reference plate for calibration, the light intensity distribution on the light receiving element can be smoothed. The condenser lens 104 may have a power in the sub-scanning direction. In short, it is sufficient to completely suppress the shading of the light receiving lens 11 and simultaneously realize an effective light intensity distribution that takes into account the focal length of the light receiving lens 11 with respect to the inspection object. In other words, it is preferable that the lens power in the main scanning direction is greater than the lens power in the sub-scanning direction. As the condenser lens 104, for example, a cylindrical lens, a lenticular lens, a Fresnel lens, or a combination of a prism sheet and a spherical lens, etc. are preferable. The positional relationship of the light source 103 with respect to the main scanning direction may be such that it is disposed at the middle position of each light receiving lens 11, or may be disposed on the optical axis of the light receiving lens 11. Furthermore, the light source 103 may be disposed at the outer periphery of the light receiving lens 11. Furthermore, a lens having a power distribution in the main scanning direction is more preferable.

[0054] In this embodiment, the condenser lens 104 that condenses light beams from the multiple light sources 103 is provided as a single lens body, and the power of the lens body in the main scanning direction is greater than the power in the sub-scanning direction. However, the condenser lens 104 may be configured to include a first condenser lens and a second condenser lens, rather than being a single lens body. In this case, the power of the first condenser lens in the main scanning direction may be greater than the power in the sub-scanning direction, and the power of the second condenser lens in the sub-scanning direction may be greater than the power in the main scanning direction. Also, the power in the main scanning direction may be adjustable by the first condenser lens and the second condenser lens.

[0055] In the condenser lens 104 having the first condenser lens and the second condenser lens as described above, the first condenser lens and the second condenser lens can be configured by a cylindrical lens, a lenticular lens, a Fresnel lens, or a prism row. For example, the first condenser lens may be a lenticular lens or a prism row. Furthermore, the second condenser lens may be a Fresnel lens or a cylindrical lens.

[0056] 7. Receiver lens parameters A graph of the light receiving system MTF for each lens diameter in this embodiment is shown in Figure 10C. As a comparative example, a case where the refractive index distribution constant √A=0.154 is shown in Figure 10D. The solid line, short dashed line, long dashed line, and dashed dotted line show the cases where the effective diameter Φ is Φ=1.0, 1.5, 2.0, 2.5, and 3.0 mm, respectively. The pixel size of the light receiving element is 42.3 μm in both the main scanning direction and the sub-scanning direction, which is equivalent to 600 dpi.

[0057] In this embodiment, when the refractive index distribution constant √A is √A=0.077 at the effective diameter Φ, the MTF characteristics are satisfied in the range from Φ1.0 mm to Φ3.0 mm if the performance is about 30% at 12 lines / mm equivalent to 600 dpi. In contrast, the comparative example does not satisfy the MTF characteristics in the entire range of the effective diameter Φ. This indicates that the aberration characteristics are better when √A is √A=0.077 than when √A is √A=0.154. In addition, for the focal length f=50 mm, the MTF of the light receiving optical system when the refractive index distribution constant √A is √A=0.1027 is obtained and shown in FIG. 10E. According to FIG. 10E, even when √A is √A=0.1027, the MTF characteristics are satisfied in the range from Φ1.0 mm to Φ3.0 mm at 12 lines / mm equivalent to a resolution of 600 dpi.

[0058] When the light receiving lens 11 is a gradient index lens, the lens is preferably made of glass or resin. In this case, the lens parameters of the light receiving lens 11 are preferably such that the axial refractive index N0 is 1.45≦N0≦1.65, the refractive index distribution constant √A is 0.05≦√A≦0.12, and the focal length f is 50 mm≦f≦150 mm.

[0059] The light receiving lens 11 may be an achromat or an apochromat that is a combination of multiple lenses. In this case, the multiple lenses may be a lens system that combines only convex lenses, or a lens system that combines convex and concave lenses. It is preferable that the focal length f of the multiple lenses is 50 mm≦f≦250 mm, and that the aperture Φ of the multiple lenses is 2 mm≦Φ≦20 mm.

[0060] Next, an example in which the light receiving lens system is a reduced optical system in order to improve the depth of field and the resolution will be shown. The configuration of the optical system is the same as that of FIG. 10A, and the magnification of the light receiving lens 11 is changed. That is, the relationship between the lateral magnification of the object point and the image point of the light receiving lens 11 is changed to 9:1 or 4:1. By doing so, the depth of field is increased to three times that of the life-size system when the lateral magnification is 9:1, and to twice that of the life-size system when the lateral magnification is 4:1. When the depth of field is tripled, the size of the light receiving element is set to 1 / 9 of the size of the light receiving element of the life-size system. When the depth of field is doubled, the size of the light receiving element is set to 1 / 4 of the size of the light receiving element of the life-size system. In this embodiment, the depth of field is doubled by using a light receiving element of 1 / 4 size.

[0061] The MTF characteristics of the reduction optical system in this embodiment when defocused are shown in the graph of Fig. 11 (solid line). As a comparative example, the MTF characteristics of a life-size system when defocused are shown in the graph of Fig. 11 (dashed line). From the comparison between the reduction system and the life-size system in the figure, it can be seen that the depth of field of the reduction optical system is about twice as deep as that of the life-size system. Furthermore, if the reduction ratio is S, the NA of the light receiving lens 11 increases by 1 / S, which is the reciprocal of the reduction ratio. Therefore, even if the lens diameter of the light receiving lens 11 is small, if the reduction ratio is increased according to the focal length, the effective NA can be maintained, and the blur due to the diffraction effect can also be kept constant.

[0062] In this embodiment, the focal length is f=50 mm and the lateral magnification ratio is 1 / 4, so in principle, the lens diameter Φ can be as small as Φ=0.25 mm. Since the same lens as in FIG. 10A is used, the lens diameter Φ is Φ=4 mm and the effective diameter Φ' is Φ'=2.5 mm. The NA is four times that of the 1x system in FIG. 10A. Therefore, the amount of received light is 16 times, which can offset the 1 / 16 reduction in the amount of received light caused by reducing the size of the light receiving element to 1 / 4.

[0063] 8.Telecentric optical system Next, the telecentric optical system applied to this embodiment will be described. By using a telecentric optical system, it is possible to achieve a configuration that is easy to assemble and has little change over time. In a telecentric optical system, a light-receiving lens 11 is arranged on either or both sides of an aperture (light-transmitting portion). In other words, the telecentric optical system is a double-sided telecentric optical system in which a light-receiving lens 11 is arranged on both the inspection object side and the light-receiving element array 120 side, or an object-side telecentric optical system in which a light-receiving lens 11 is arranged only on the inspection object side.

[0064] In a double-telecentric optical system, a thin light beam that spreads parallel to the optical axis from the object to be inspected passes through the front lens and becomes a parallel light beam. The parallel light beam passes through an aperture inserted after the front lens, and then passes through the rear lens to be focused on each light receiving element of the light receiving element array 120.

[0065] When the light receiving element array 120 is staggered in the sub-scanning direction as shown in FIG. 3 or FIG. 4, the memory capacity needs to be increased as the distance in the sub-scanning direction increases. The larger the memory capacity, the higher the cost. In this embodiment, by using a telecentric lens optical system, if the NA is about 0.03, the lens has a focal length of 50 mm, and the light receiving solid angle is a cone connecting Φ=3 mm on the main plane of the lens and the object point. Therefore, the shortest distance Ls of the light receiving element array 120 staggered in the sub-scanning direction is sufficient if Ls>3 mm. Therefore, if the circular outer periphery of a normal lens is made into a rectangular outer periphery by various processing or molding means, and the light receiving element array 120 is arranged with a narrower staggered interval as shown in FIG. 3 or FIG. 4, the light receiving solid angle can be satisfied, compactification is possible, and the memory capacity can be reduced at the same time. The NA of the light receiving lens 11 is preferably about 0.005 to 0.03 from the viewpoint of the amount of illumination light. As described above, the thickness of light receiving lens 11 in the sub-scanning direction is determined by the solid angle defined by the NA of the telecentric optical system, and therefore it can be made short in the sub-scanning direction and compact.

[0066] 9. Modifications of the light receiving system Next, modified examples of the light receiving system will be described with reference to Figs. 12A to 12G. In the above embodiment, a configuration in which a plurality of light receiving lenses 11 are arranged in two rows in the sub-scanning direction has been described, but Figs. 12A to 12G describe a configuration in which a plurality of light receiving lenses 11 are arranged in one row along the main scanning direction, specifically, a layout of one row of light receiving lenses 11 and a staggered light receiving element array 120. In Figs. 12A to 12G, a plurality of light receiving lenses 11 are arranged in a number corresponding to the number of light receiving element arrays 120. That is, each light receiving lens 11 corresponds one-to-one to each of the light receiving element arrays 120 arranged in a staggered pattern.

[0067] When two rows of reading lines L are formed by staggering the light receiving element array 120, if the light receiving lenses 11 are also staggered in two rows like the light receiving element array 120, it was necessary to provide a spacing in the sub-scanning direction of the light receiving element array 120 that is greater than the width of the light receiving lenses 11 in the sub-scanning direction in order to satisfy the required NA of the light receiving lens system. Therefore, as shown in Figures 12A to 12G, if the width of the rectangular lens of the light receiving lens 11 is set so as to satisfy the NA required for the light receiving system and the number of rows is reduced from two to one, it becomes possible to further narrow the spacing in the sub-scanning direction of the light receiving element array 120. In other words, the effect of speed fluctuation in the sub-scanning direction can be reduced, and at the same time, the cost of the light receiving lenses 11 can be reduced. Below, a method of arranging a plurality of light receiving lenses 11 in one row will be described.

[0068] 12A to 12G, the multiple light receiving lenses 11 have the same shape when viewed from a direction perpendicular to the main scanning direction. Specifically, the shapes of the light receiving lenses 11 are the same when viewed from the Z direction perpendicular to the main scanning direction (X direction) and the sub-scanning direction (Y direction). In FIGS. 12A to 12G, the light receiving lenses 11 are formed in a rectangular shape when viewed from a direction (Z direction) perpendicular to the main scanning direction and the sub-scanning direction. Note that "the same shape" includes substantially the same shapes, such as an inverted shape, and slight differences in shape are also included in "the same shape."

[0069] In Fig. 12A, the shape of each light receiving lens 11 when viewed in the Z direction is a parallelogram. That is, both end faces in the main scanning direction of each light receiving lens 11 are inclined with respect to the sub-scanning direction. The inclination angles of both end faces in the main scanning direction of each light receiving lens 11 are the same. When multiple light receiving lenses 11 are arranged side by side in the main scanning direction as in Fig. 12A, both end faces of each light receiving lens 11 face the end faces of adjacent light receiving lenses 11, so that the ends of adjacent light receiving lenses 11 in the main scanning direction are connected together without any gaps.

[0070] 12B, a rectangular notch is formed on both end faces in the main scanning direction of each light receiving lens 11 when viewed in the Z direction. When multiple light receiving lenses 11 are arranged side by side in the main scanning direction as shown in Fig. 12B, parts of adjacent light receiving lenses 11 fit into the notches formed on both end faces of each light receiving lens 11, so that the ends of adjacent light receiving lenses 11 in the main scanning direction are arranged in a state where they are connected without any gaps.

[0071] In FIG. 12C, the shape of each light receiving lens 11 as viewed in the Z direction is a trapezoid. That is, both end faces in the main scanning direction of each light receiving lens 11 are inclined with respect to the sub-scanning direction. The shape of each light receiving lens 11 as viewed in the Z direction is an inverted shape with respect to the main scanning direction with respect to the shape of adjacent light receiving lenses 11 as viewed in the Z direction. When multiple light receiving lenses 11 are arranged side by side in the main scanning direction as shown in FIG. 12C, the end faces of adjacent light receiving lenses 11 face both end faces of each light receiving lens 11, so that the ends of adjacent light receiving lenses 11 in the main scanning direction are connected to each other without any gaps. As a result, the multiple light receiving lenses 11 are arranged in a state in which adjacent light receiving lenses 11 are inverted and connected to each other.

[0072] 12D, a curved (semicircular or semi-elliptical) notch is formed on both end faces in the main scanning direction of each light receiving lens 11 when viewed in the Z direction. When multiple light receiving lenses 11 are arranged side by side in the main scanning direction as in Fig. 12D, parts of adjacent light receiving lenses 11 fit into the notches formed on both end faces of each light receiving lens 11, so that the ends of adjacent light receiving lenses 11 in the main scanning direction are connected together without any gaps.

[0073] Next, modified examples of Fig. 12E to Fig. 12G will be described. Fig. 12E is a modified example of Fig. 12C, which can minimize the occurrence of missing pixels in the above-mentioned light receiving element array 120. In Fig. 12E, each light receiving element array 120 is arranged on the longer side of each trapezoidal light receiving lens 11 from the center in the width direction. In addition, the length of each light receiving element array 120 in the main scanning direction is approximately equal to the length of each trapezoidal light receiving lens 11 in the main scanning direction at the position where the light receiving element array 120 is arranged. By doing so, the portions indicated by the dashed lines in Fig. 12E of the staggered light receiving element arrays 120 adjacent to each other overlap in the main scanning direction, and as described later, no missing portions of the output signal are generated as in the staggered light receiving element array 120 shown in Fig. 12C.

[0074] The above is not limited to the trapezoidal light receiving lens 11, and may be any shape that can be connected without gaps when each light receiving lens 11 is inverted in the sub-scanning direction to form a light receiving lens array. Examples are shown in Figs. 12F and 12G. Fig. 12F shows a case where a plurality of light receiving lenses 11 having rectangular notches are inverted and connected, and Fig. 12G shows a case where a plurality of light receiving lenses 11 having circular notches are inverted and connected. Thus, in Figs. 12E to 12G, a plurality of light receiving lenses 11 are arranged in a state where adjacent light receiving lenses 11 are inverted and connected. Generally, inversion connection is possible if the light receiving lenses 11 are line symmetric in the sub-scanning direction. In Figs. 12F and 12G, the dashed line parts overlap as in the trapezoidal inversion connection of Fig. 12E, and as will be described later, no signal missing parts are generated. The light receiving lens 11 in Fig. 12G has an edge part 112 for ensuring rigidity and avoiding cracks.

[0075] 12A to 12G, in a cross-sectional view in the optical axis direction of the light receiving lens 11, the lens end is set to have a cross-sectional shape such as a parallelogram, trapezoid, rectangular, or curved (circle, ellipse) cutout shape in the sub-scanning direction that satisfies the NA of the light receiving lens system, and the lens end is connected in one row to form a lens array extending in a straight line in the main scanning direction. In addition, two rows of staggered light receiving element arrays 120 are provided in the center of the lens array in the sub-scanning direction. Specifically, a plurality of light receiving element arrays 120 are disposed in the center of each of the plurality of light receiving lenses 11 in the main scanning direction, and are alternately disposed in a staggered manner along the main scanning direction. In this way, light emitted from the inspection object can be delivered to the two rows of staggered light receiving element arrays 120 by one lens array.

[0076] As shown in FIG. 12A, a shielding portion 111 is provided between the ends of the light receiving lenses 11 adjacent to each other in the main scanning direction. The shielding portion 111 is formed in a thin plate shape. The shielding portion 111 extends from between the ends of each light receiving lens 11 toward the light receiving element (light receiving element array 120) and protrudes from the light receiving lens 11 toward the inspection target. The shielding portion 111 has a width at least equal to or greater than the width W1 of the light receiving lens 11 in the sub-scanning direction. Although the shielding portion 111 is omitted in FIGS. 12B to 12G, the shielding portion 111 may be provided between the ends of each light receiving lens 11 in the same manner. In the embodiments of FIGS. 12F and 12G, the shielding portion 111 is a thin crank-shaped or wave-shaped shielding plate. In short, it is sufficient to prevent light from leaking out to the adjacent light receiving systems at the connecting portion.

[0077] 12B to 12D, a plurality of light sources 103 for illuminating an inspection target may be provided in one-to-one correspondence with a plurality of light receiving element arrays 120. Each of the light sources 103 may have an elongated shape extending linearly along the main scanning direction, and in this case, as shown in Fig. 12B to 12D, the light sources 103 may face each of the light receiving element arrays 120 in the Z direction. This allows the light sources 103 to be arranged in a staggered pattern, and allows linear light to be efficiently incident on each of the light receiving element arrays 120.

[0078] The light sources 103 have a light intensity distribution of illumination light of a plurality of lines in the sub-scanning direction corresponding to the plurality of light receiving element arrays 120. Although the light sources 103 are omitted in FIG. 12A and FIG. 12E to FIG. 12G, the light sources 103 may be provided in one-to-one correspondence with the plurality of light receiving element arrays 120. However, instead of arranging the light sources 103 in a staggered manner, a line illumination having a light intensity distribution including two intensity peaks in the sub-scanning direction may be used so that the light receiving element array 120 of each reading line L has a light intensity peak. In addition, since the intervals between the staggered light receiving element arrays 120 in the sub-scanning direction are narrow, a line illumination having one peak may be used. In addition, the light sources 103 may be located near the ends of the plurality of light receiving element arrays 120 in a cross section in the optical axis direction of the light source 103, and at approximately the center of the plurality of light receiving element arrays 120 facing each other in the sub-scanning direction.

[0079] Figures 12H and 12I are schematic diagrams for explaining the position of optical axis 113 in the light receiving system. In the following, the position of optical axis 113 in the light receiving system will be explained using Figures 12H and 12I, taking the configuration of Figure 12E as an example, but the position of optical axis 113 can be set in a similar manner in other configurations such as Figures 12A to 12D, 12F, and 12G.

[0080] 12H, optical axes 113 of light transmitted through each light receiving lens 11 and guided to each light receiving element array 120 pass through a position parallel to and spaced apart from the approximate center of each of the staggered light receiving element arrays 120 in the sub-scanning direction (Y direction). In this example, the optical axes 113 are aligned in a row along the main scanning direction (X direction), but the present invention is not limited to this configuration.

[0081] 12I, optical axis 113 of light transmitted through each light receiving lens 11 and guided to each light receiving element array 120 passes through approximately the center of each of the staggered light receiving element arrays 120. Therefore, each optical axis 113 is arranged in a staggered manner along two reading lines L, similar to each of the light receiving element arrays 120. In this manner, by aligning the optical axis 113 of each light receiving lens 11 with each light receiving element array 120 so that the optical axis 113 of each light receiving lens 11 is positioned approximately in the center of the width of the light receiving lens 11 in the sub-scanning direction, aberration can be reduced.

[0082] Fig. 13 shows the light intensity distribution on the light receiving surface of the light receiving element array 120 when the light intensity distribution of the light source 103 is flat when the lens end is a parallelogram. Fig. 13 shows an example of the light intensity distribution on the light receiving surface of each light receiving element array 120 when the configuration of Fig. 12A is adopted. For the sake of explanation, the lens array consisting of multiple light receiving lenses 11 is assumed to be a lens array consisting of three light receiving lenses 11.

[0083] The diagram on the left side of Fig. 13 shows the light intensity on each light receiving element array 120 of the entire lens array, and the two diagrams on the right side show enlarged views. Fig. 14 also shows details of the light intensity distribution near the joint between adjacent light receiving lenses 11. In this embodiment, in two adjacent light receiving element arrays 120 arranged in a staggered pattern, signal missing pixels of one light receiving element array 120 are interpolated with pixels of the other light receiving element array 120. In Fig. 14, the arrow marks indicate a method of interpolating signals of signal missing pixels.

[0084] In FIG. 14, a portion 121 surrounded by a dashed line is a light receiving element located at a connection portion between adjacent light receiving lenses 11, and a signal missing portion occurs when light does not reach the light receiving element. Therefore, in this embodiment, in the adjacent staggered light receiving element array 120, the signal missing portion is supplemented with the output signal of a light receiving element (portion 122 surrounded by a solid line) that faces the light receiving element of the portion 121 in the sub-scanning direction. In this way, a pixel interpolation portion is formed along the sub-scanning direction at the connection portion between adjacent light receiving lenses 11 (see FIG. 13). The signal missing portion may be corrected in advance using a reference medium for inspection. A white chart or a grid pattern for shading correction may be used as the reference medium.

[0085] The width of the light receiving lens 11 in the sub-scanning direction in FIG. 13 is 5 mm. Therefore, when the light receiving lenses 11 are arranged in two staggered rows, the minimum interval of the staggered light receiving element array 120 needs to be 5 mm or more. However, when a single row of lens array is used as described above, the interval becomes about 1 / 2 (just under 3.0 mm). In addition, if shading correction is performed while taking into consideration the dynamic range of the light receiving element array 120, it is possible to further narrow the interval. That is, the interval of the staggered light receiving element array 120 in the sub-scanning direction is narrowed, and at the same time, the number of light receiving lenses 11 can be halved. Therefore, there is an advantage in that it is less susceptible to the influence of fluctuations in the transport speed in the sub-scanning direction and at the same time, costs can be reduced.

[0086] A shielding plate with a thickness t=0.2 mm is placed in the shielding portion 111 in Fig. 13. If t=0.2 mm, there is no problem in terms of strength. The lens system is a double-telecentric system. Of course, it goes without saying that it may be a telecentric system on the object side.

[0087] In a specific image processing method or processing by the image processing system, when an output signal from one light receiving element (part 121 surrounded by a dashed line) is lower than an output signal from the other light receiving element (part 122 surrounded by a solid line) and does not reach a threshold value, the output signal from the other light receiving element is interpolated. The interpolated output signal is then combined with an output signal from a light receiving element at another position in the main scanning direction relative to the one light receiving element to generate an output signal for one row corresponding to the reading line L.

[0088] 14, the light receiving elements in portion 123 surrounded by a two-dot chain line face each other in the sub-scanning direction, but overlap each other because they are shifted from the joints between adjacent light receiving lenses 11. In such overlapping portions, there is no need for interpolation because there is no loss of output signals from the light receiving elements.

[0089] In the example of FIG. 12E, the light intensity of the light receiving region is shown, and the reason why no signal missing portion occurs is shown in FIG. 15. FIG. 15 shows the light intensity distribution at the end of the staggered light receiving element array 120 as in FIG. 14. The light receiving elements in the portion 124 surrounded by the two-dot chain line in FIG. 15 face each other in the sub-scanning direction, but they are all overlapped by being shifted from the connection portion between the adjacent light receiving lenses 11. As shown in FIG. 15, by arranging the light receiving element array 120 on the long side in the width direction of the trapezoidal light receiving lens 11, it is possible to obtain the signal required for one reading line L of the staggered light receiving element array 120, except for the signal missing portion in the light receiving portion.

[0090] A specific image processing method or processing by an image processing system involves selecting one of the output signals from two light receiving elements (area 124 surrounded by dashed lines) that are spaced apart in the sub-scanning direction at the same position in the main scanning direction where the output signal of one light receiving element overlaps with the output signal of the other light receiving element, and correcting the ratio of one output signal to the other output signal, which has been previously corrected using a reference medium, to combine the one output signal with the other output signal to obtain a single row of output signals corresponding to the reading line L.

[0091] 10. Example of light receiving optical system and detection circuit system FIG. 16A shows an embodiment in which an image intensifier 200 is arranged in front of the light receiving element array 120. FIG. 16B shows a more detailed schematic diagram of the image intensifier 200 arranged in front of the light receiving element array 120. The image intensifier 200 amplifies the intensity of an image formed by light, and is arranged opposite to the light receiving surface of a plurality of light receiving elements (light receiving element array 120). The image intensifier 200 includes a photocathode 201, a microchannel plate 202 (hereinafter referred to as MCP 202), a phosphor surface 203, and the like. An optical fiber plate 204 is arranged between the image intensifier 200 and the light receiving element array 120. The photocathode 201, the microchannel plate 202, and the phosphor surface 203 are arranged close to each other in a ceramic vacuum sealer 205.

[0092] The photocathode 201 converts light into electrons. The MCP 202 multiplies the electrons converted from light by the photocathode 201. The phosphor surface 203 converts the electrons multiplied by the MCP 202 into light. The optical fiber plate 204 guides the light converted from electrons by the phosphor surface 203 to each light receiving element of the light receiving element array 120. Specifically, the light imaged on the photocathode 201 is converted into electrons by the photocathode 201, and the electrons enter each channel of the MCP 202. The electrons that enter each channel of the MCP 202 are multiplied by the electron avalanche effect due to the high voltage gradient at both ends of the MCP 202, and then collide with the phosphor surface 203. The phosphor surface 203 outputs light according to the amount of electrons that collide with it, and the output light is guided to each light receiving element of the light receiving element array 120 via the optical fiber plate 204.

[0093] In FIG. 16B, the MCP 202 is rectangular in shape to match the optical line sensor (light receiving element array 120) including the image intensifier 200. The light receiving optical system in FIG. 16A is a telecentric optical system 300, which is a double-telecentric optical system. Specifically, the light receiving optical system in which the light receiving lenses 11 are arranged on the upstream side and downstream side of the aperture 21, respectively, guides light from the object to be inspected to the light receiving element array 120 via the image intensifier 200. That is, a plurality of light receiving lenses 11 constitute the telecentric optical system 300. However, the telecentric optical system 300 does not necessarily have to be a double-telecentric optical system, and an object-side telecentric optical system or another imaging lens system capable of corresponding the light receiving element array 120 to the observation point of the object to be inspected may be used. Also, FIG. 16C shows a case in which the light receiving element array 120 in FIG. 16B is a three-line sensor.

[0094] FIG. 17 shows a schematic cross-sectional view of the image intensifier 200. Extremely weak light is incident on the photocathode 201 of the image intensifier 200 from the left side of the paper in FIG. 17, and then photoelectrons are generated at the photocathode 201. The photoelectrons then enter the incident surface side of the MCP 202. Next, the photoelectrons that enter each tube partitioned by the secondary electron multiplier electrode surface of the MCP 202 head toward the exit of the MCP 202, where they are multiplied by the above-mentioned electron avalanche effect, increasing the number of electrons. The electrons that reach the exit of the MCP 202 enter the phosphor surface 203 and generate fluorescence. The fluorescence enters and propagates through each optical fiber 206 that constitutes the optical fiber plate 204, and is emitted from the exit of the optical fiber plate 204. The fluorescence emitted from each optical fiber 206 of the optical fiber plate 204 enters a pixel in the photodetector array 120 corresponding to each tube of the MCP 202 into which the extremely weak light was incident, making it possible to obtain an output several thousand to tens of thousands times greater than the output that can be obtained directly from the extremely weak light.

[0095] The above embodiment is particularly effective for inspection objects containing foreign matter or defects. That is, even if a medium contains foreign matter or defects and has low transmittance, imaging close to so-called photon counting is possible, making it easy to distinguish between foreign matter and defects. In addition, when the telecentric optical system 300 is adopted, the aperture diameter is narrowed down in consideration of the relationship between the diffraction limit and the NA of the light receiving optical system, and further the magnification (reduction ratio), so that only the geometrical optical linear light components can be selected, and the obtained image of the foreign matter or defect becomes clearer (has good contrast), and as a result, it is expected that the type, material, etc. of the foreign matter or defect can be specified.

[0096] By combining the telecentric optical system 300 and the image intensifier 200 and making the aperture size of the telecentric optical system 300 extremely small while balancing the diffraction limit and geometrical optics, the precision and accuracy of detecting foreign objects and defects is improved. According to the relationship between the diffraction limit and NA in FIG. 6 described above, in order to obtain a resolution of about 20 μm (equivalent to 1200 dpi) at a wavelength λ=850 nm, NA≒0.02 is required. Therefore, when a light receiving lens 11 with a long WD is used on the side of the inspection object, a lens with a short focal length may be used for the light receiving lens 11 on the side of the light receiving element array 120 to form a reduction optical system. For example, when a light receiving lens 11 with a focal length of f=100 mm is used on the side of the inspection object and the aperture size Da is Da=800 μm, the focal length of the light receiving lens 11 on the side of the light receiving element array 120 is f=10 mm, that is, a reduction optical system of 10:1 may be used.

[0097] In addition, as shown in Fig. 17, the MCP 202 has a bias angle between the direction perpendicular to the incident / exiting surface and the axis (center line) of each tube of the MCP 202. The bias angle is set to several degrees so that photoelectrons always enter the secondary electron multiplier electrode surface and do not pass directly through without hitting it, and is fixed so that the bias angle is maximum in the main scanning direction as shown in Fig. 17. For example, when the bias angle θb is θb = 8° with respect to the main scanning direction and the thickness of the MCP 202 is 0.5 mm, the axial shift Ls between the incident surface and the exit surface of each tube of the MCP 202 is Ls = 625 µm, so that an optical line sensor (photodetector array 120) longer than the axial shift amount in the main scanning direction is used, or an optical line sensor (photodetector array 120) slightly longer than the MCP 202 is previously arranged by shifting it in the main scanning direction by the bias angle.

[0098] In most cases, the object to be inspected has a transmittance distribution, and when the image intensifier 200 is used, if the transmittance of the object to be inspected is high, a measure is taken to alleviate the output saturation of the image intensifier 200 by the auto gate function of the image intensifier 200 itself. However, even if the image intensifier 200 is provided with measures against output saturation, it is assumed that the control range of the transmittance fluctuation may be exceeded. Therefore, it is possible to suppress saturation by controlling the amount of light on the light source 103 side at the same time as the output becomes saturated. Therefore, when the auto gate function of the image intensifier 200 works, a feedback signal is output to the light source 103 side, and there is a means for reducing the amount of light of the light source 103 based on the feedback signal, and multiple steps of the amount of light to be reduced are prepared, and the amount of light may be repeatedly increased or decreased each time feedback is received. In this way, the image intensifier 200 (MCP 202) may have an auto gate function, and the light source 103 side may have a function of adjusting the amount of light in conjunction with the auto gate function.

[0099] Next, a method of utilizing the difference in optical path difference, i.e., the difference in time of flight, as a means for removing scattered light will be described. Figure 18 is a diagram for explaining the time gate method used in the present invention. When a certain time has elapsed since the light receiving element started accumulation and the light source 103 is made to emit light, electrons fly accelerated by a high voltage in the MCP 202 in the image intensifier 200, and reach the phosphor surface 203 after repeatedly colliding with the wall surface of the MCP 202, so that the fluorescence emitted from the phosphor surface 203 is delayed for a very short time. If the fluorescent illumination light illuminates the light receiving element only for the accumulation time of the light receiving element and for an extremely short time before the time when the image intensifier 200 emits fluorescence and the accumulation time of the light receiving element ends, it becomes possible to receive only the early stage of fluorescence emission, i.e., the light that arrives early. The light that arrives early is a straight light component with a short optical path, while the scattered light component has a long optical path and reaches the light receiving element later than the straight light component, so as mentioned above, it is possible to prevent the scattered light component from reaching the light receiving element within the accumulation time. In this way, the scattered light component can be removed, enabling measurements with a good S / N ratio and good contrast, and ultimately making it possible to detect foreign objects and defects in transparent light scattering media.

[0100] 19 is a schematic diagram of a timing chart of a pulse delay phase difference detection method. The light receiving element starts accumulating photoelectrons generated by the fluorescence emitted from the image intensifier 200 with a predetermined delay from the start pulse. Next, the light source 103 is turned ON before the accumulation ends. At this time, the light source 103 is made to emit a pulsed light with a fixed delay from the start pulse using a delay circuit (equivalent circuit) 210 shown in FIG. 20A. The delay circuit (equivalent circuit) 210 is a delay ladder gate circuit having a plurality of LC circuits 211, and each LC circuit 211 is provided with a switch 212.

[0101] Next, the light source 103 is turned off in a very short time. Alternatively, the gate of the image intensifier 200 may be closed. However, since the fluorescent afterglow continues to illuminate the light receiving element, the accumulation time is ended so that the light receiving element does not obtain gain due to the fluorescent afterglow. The light emission timing of the light source 103, the timing of the end of the accumulation time, and the gate circuit of the image intensifier 200 can be adjusted (tuned) by increasing or decreasing the parameters and number of the LC circuit 211 by opening and closing the switch 212 of the delay circuit (equivalent circuit) 210 in Fig. 20A described above.

[0102] A representative example of the delay circuit 210 is a coaxial cable, which is a high-frequency transmission path (waveguide). In addition, there are also striplines and microstriplines, which are high-frequency transmission paths similar to coaxial cables. The above-mentioned striplines and microstriplines also operate on the same principle as coaxial cables, and the impedance does not change depending on the length, and only the delay time changes. A stripline is a circuit that replaces a coaxial cable, and is formed by cutting open both sides of the outer conductor of a coaxial cable to make two plates, and stretching the inner conductor into a foil shape. In recent years, it has been used in mobile phones and the like.

[0103] 21 is a schematic diagram of a stripline 220. The stripline 220 has a configuration in which conductors 222 are provided on the front and back surfaces of a plate-like dielectric 221, and a linear conductor 223 is provided inside the dielectric 221, and the conductor 222 is grounded. H is the height of the stripline 220, W is the width of the conductor 223, and t is the thickness of the conductor 223. The stripline 220 can achieve impedance characteristics equivalent to those of a coaxial cable over a short distance of about several meters, and can set the characteristic impedance Z0 to a constant Z0≈50Ω or Z0≈75Ω that is independent of the length.

[0104] For example, in FIG. 21, if H=0.2 mm, W=0.33 mm, t=0.04 mm, and the dielectric constant εr of the dielectric 221 is εr=4 (assuming it is an epoxy-based resin), the characteristic impedance Z0 is Z0≈50Ω. The delay time Tr is 103√εr / 0.3nsec / m, so Tr≈6.7nsec / m, and the delay time of the strip line 220 is Tr≈6.7psec / mm. In addition, the delay time of a coaxial cable is Tr≈5nsec / m, so the delay time is 5psec / mm. Therefore, a simple method is to create multiple strip lines with different switch positions on a board and switch multiple ladder stage circuits connected together with an FPGA gate to tune the delay time. Alternatively, there is a method of creating micro-coaxial cables of different lengths on a board and changing the length of the micro-coaxial cables as appropriate using a switch, or a method of using a microstrip line, etc. Coaxial cables with extremely thin diameters of Φ=0.2 to 0.3 mm are already commercially available and are used in the bent parts of mobile phones, notebook computers, etc. As described above, by adjusting the timing of the fluorescence emission time of the image intensifier 200 and the accumulation time of the light receiving element, it is possible to remove the scattered light component and obtain only the straight light component, thereby improving the detection accuracy of foreign bodies and defects inside the object to be inspected. As long as the delay is stable, other delay elements or delay circuits may be used, and the delay time can be set by the time constant of the circuit and the threshold value at which each element operates.

[0105] Near the end of the accumulation time of the light receiving element, the light source 103 is turned on / off and the gate of the image intensifier 200 is turned on / off while considering the response time of the light source 103 and the response time of the image intensifier 200, respectively, to shorten the light observation time, and it becomes possible to receive only the straight light component emitted from the object to be inspected that reaches the light receiving element the earliest. Since the response time of the image intensifier 200 is several tens of nsec, the timing of closing the shutter of the gate of the image intensifier 200 is adjusted while considering the response delay time of the image intensifier 200. The time adjustment is performed by using a coaxial cable circuit network with a changed length of the coaxial cable and selecting a specific circuit in the circuit network by the FPGA (switching the circuit). Alternatively, a stripline circuit network may be used instead of the coaxial cable circuit network. The circuit network may be a separate circuit for the light source 103 and the image intensifier 200, and the delay time according to each response time may be tuned. Incidentally, cameras with a 10psec delay gate function are commercially available as "ICCD cameras."

[0106] The light source 103 used in the above-mentioned pulse delay phase difference detection can be a picosecond pulse laser or a femtosecond pulse laser, and by synchronizing the repetitive pulse with the accumulation time, efficient foreign matter / defect detection becomes possible in a delay phase difference detection method that combines an image intensifier 200 and a pulse laser. Since the line rate of a line-type CMOS sensor is about 100 to 200 KHz, the repetition rate of a repetitive pulse laser can be realized without any problem.

[0107] By synchronizing the repetition accumulation time of the high repetition laser and the light receiving element, foreign objects and defects in a light scattering medium can be detected. In addition, by combining multiple femtosecond level pulse lasers, the position in the thickness direction of the object to be inspected appears as an optical path difference by a detector having multiple phase shifts as shown in FIG. 18, and by obtaining position information in the thickness direction, it becomes possible to grasp the internal structure of the object to be inspected itself. In addition to the high repetition laser, the gate opening and closing time and repetition frequency of the image intensifier 200 must also be considered, and there are commercially available products with a gate time of 3 nsec and a repetition frequency of about 300 KHz. In other words, there are commercially available products that have the same response as the high repetition laser. A small jitter is required to match the timing of each element, but in this embodiment, the fluorescence acquisition time is set to 1 nsec or less, and if the jitter is 10 psec (rms value) or less, it will be within the error range of 1%. The jitter of the reference clock is the most important, and at a clock frequency of 1 GHz, there have already been commercially available products that have achieved a jitter (rms value) of 1 psec or less.

[0108] The optimum value of each parameter can be determined by considering the repetition frequency of the light source 103 and the image intensifier 200, the timing (time interval) of the accumulation time of the light receiving element, and the transport speed. Furthermore, by performing a discrete Fourier transform on certain data in the main scanning direction and the sub-scanning direction of the output of the light receiving element with respect to the thickness direction of the inspection object, it is possible to obtain concentration information for any volume of the inspection object and obtain the three-dimensional structure of the inspection object. However, in this case, it is necessary to scan in the thickness (depth) direction.

[0109] The modified example will be described in detail below. The line rate of the light receiving element array 120 is 125 KHz. The repetition frequency of the gate opening and closing time of the image intensifier 200 is also set to 125 KHz in accordance with the line rate of the light receiving element array 120, and the gate time is set to 5 nsec. The accumulation time of the light receiving element array 120 is set to 5 μsec in consideration of the line rate. That is, the gate is opened only for the last 1 / 1000th of the accumulation time, and the light receiving element array 120 receives the fluorescence emitted from the image intensifier 200. For the high repetition laser, pulse width modulation (PWM) is performed, and the pulse width is set to 2 nsec, which is shorter than the gate operation time of the image intensifier 200. It is also possible to shift the phase of the operation time of the light source 103 and the image intensifier 200 to limit the light incidence on the photocathode 201 of the image intensifier 200. If the light source 103 emits light early and the gate operation start of the image intensifier 200 is delayed, the amount of light is further limited to the superimposed time. Also, if the duty ratio is reduced, it is possible to achieve high output in a short time. The repetition frequency is set to 125 KHz in accordance with the line rate of the light receiving element array 120.

[0110] In the image intensifier 200, it takes a certain amount of time for light to reach the photocathode 201 and exit the phosphor surface 203. That is, there is a delay compared to when the light reaches the light receiving element array 120 without the image intensifier 200. The delay time of the image intensifier 200 in this embodiment varies from one device to another, but is generally set to 50 to 60 nsec. Therefore, if the start of the accumulation time is delayed by 50 to 60 nsec, light can be received at the same time as when the image intensifier 200 is not present. Regarding the individual differences, a delay circuit network can be used for tuning to match the operation of each element.

[0111] That is, the gate of the image intensifier 200 is opened, and the accumulation start timing is delayed by 50 to 60 nsec until the light source 103 is illuminated. Furthermore, even if the delay time varies due to manufacturing variations in the image intensifier 200, the delay time by the delay circuit 210 can be dealt with by switching the circuit length (or the coaxial cable length). The length of the microstrip line or coaxial cable that is easy to manufacture may also be used. The delay time difference of 5 psec is 1 mm, but it is possible to realize a delay circuit 210 that can adjust the delay time with a resolution of 5 psec by increasing it by 1 mm at a time, for example, based on 11 mm (in this case, a delay of 55 psec).

[0112] Furthermore, in order to receive light for 5 nsec before the end timing of the accumulation time, the operation time of the image intensifier 200 and the light source 103 may be delayed by 45 to 55 nsec (set to a fixed delay time after grasping the individual variations of the elements) for the timing of the accumulation time. In this case, taking into consideration the response of the light source 103 and the response of the image intensifier 200, tuning on the order of psec is performed to match the timing including the individual variations of the response of the light source 103, the image intensifier 200, and the light receiving element circuit using the delay circuit 210 shown in Fig. 20A and Fig. 20B, and it becomes possible to obtain a signal with high gain and good S / N. In other words, tuning is performed to remove the scattered light component and obtain a signal with high gain and high S / N.

[0113] According to the above results, it is preferable to set the operations of the light receiving element, the light source 103, and the image intensifier 200 as follows. That is, it is preferable that the gate operation time of the image intensifier 200 and the light emission time of the light source 103 for illuminating the inspection object are shorter than the accumulation time of the light receiving element. Also, in one cycle of a series of operations of the light receiving element, the light source 103, and the image intensifier 200, it is preferable that (1) the light emission start time and the light emission end time of the light source 103 are earlier than the accumulation operation start time of the light receiving element, and (2) the gate operation start time of the image intensifier 200 is earlier than the accumulation operation end time of the light receiving element, and the accumulation operation end time of the light receiving element is earlier than the gate operation end time of the image intensifier 200. It is preferable that at least one of the image intensifier 200 and the light source 103 is provided with a delay circuit 210, and the operation times of the light source 103 and the image intensifier 200 can be adjusted.

[0114] 11. Modifications of the optical system 22A and 22B show schematic diagrams of a reflective optical system. In FIG. 22A and FIG. 22B, a view from the main scanning direction and a view from a direction perpendicular to the main scanning direction are shown side by side. In FIG. 22 and FIG. 22B, a reflective coaxial optical system is shown, and a polarizing beam splitter 231 (hereinafter referred to as PBS 231) and a λ / 4 wave plate 232 can separate the irradiated light and the reflected light and measure the light. In FIG. 22A and FIG. 22B, the PBS 231 is disposed between the image intensifier 200 and the light receiving lens 11. In FIG. 22A, the λ / 4 wave plate 232 is disposed between the light receiving lens 11 and the image intensifier 200, and in FIG. 22B, the λ / 4 wave plate 232 is disposed between the light receiving lens 11 and the object to be inspected. However, the λ / 4 wave plate 232 may be disposed between the light receiving lens 11 and the aperture 21.

[0115] In this manner, the PBS 231 is disposed on the optical axis of the light receiving lens 11 between the image intensifier 200 and the light receiving lens 11 on the light receiving element array 120 side of the telecentric optical system 300. Also, a λ / 4 wave plate 232 is disposed either between the light receiving lens 11 on the light receiving element array 120 side of the telecentric optical system 300 and the PBS 231, or between the light receiving lens 11 on the inspection object side of the telecentric optical system 300 and the inspection object, or between the light receiving lens 11 on the light receiving element array 120 side of the telecentric optical system 300 or the light receiving lens 11 on the inspection object side and the aperture 21. The PBS 231 and the λ / 4 wave plate 232 have a width in the main scanning direction longer than that of the light receiving element array 120.

[0116] The S-polarized beam emitted from the light source 103 such as a diode is almost totally reflected by the polarizing beam splitter 231 (hereinafter referred to as PBS231), enters the λ / 4 wave plate 232, becomes circularly polarized, and enters the object to be inspected through the lens of the illumination optical system. Next, the reflected component of the object to be inspected becomes circularly polarized in the opposite direction and enters the light receiving optical system and the λ / 4 wave plate 232 again, becomes a P-polarized beam, and enters the PBS231. The P-polarized beam passes through the PBS231 and enters the MCP202, where the light intensity is amplified and then enters the light receiving element array 120, thereby obtaining a multiplied electrical output. As with the transmission type, it is possible to obtain only the straight light component, and the optical axis is automatically aligned between the illumination optical system and the light receiving optical system, which is an advantage that it is much easier to handle than the transmission type. Note that other light sources may be used as long as they have excellent polarization characteristics, light intensity, and responsiveness.

[0117] The disadvantage of the reflective type compared to the through-beam type is that the received light intensity is about three orders of magnitude lower for the through-beam type inspection target. In this case, using a two-stage MCP202 makes it possible to compensate for the decrease in light intensity. By replacing the one-stage MCP202 with a two-stage, the light intensity improves by more than three orders of magnitude, and an output equivalent to that of the through-beam type can be obtained.

[0118] In addition, in the transmission type, in addition to the combination of light source (linearly polarized light source) 103, PBS 231, and λ / 4 wave plate 232 used in the reflection type, a polarizing plate having the same polarization direction as the P-polarized component transmitted through PBS 231 may be used in the light receiving section. Needless to say, by adding the polarizing plate to the optical system of Figure 22A or Figure 22B, a straight light component with an even better degree of polarization can be obtained.

[0119] 22C and 22D are schematic diagrams of other reflective optical systems. Fig. 22C shows an object-side telecentric optical system, and Fig. 22D shows a double-telecentric optical system. Fig. 22C and Fig. 22D show λ / 4 wave plates 232 facing both sides of light receiving lens 11, but it is sufficient that λ / 4 wave plates 232 are provided facing either one side of light receiving lens 11.

[0120] 22C, a PBS 231 is disposed on the optical axis of the light receiving lens 11, between the image intensifier 200 and the light receiving lens 11 of the telecentric optical system 300, and in the vicinity of the aperture 21. However, a configuration in which the PBS 231 having the aperture 21 on the entrance / exit surface is disposed in the vicinity of the intersection of the light beams may be used. Alternatively, a λ / 4 wave plate 232 is disposed between the light receiving lens 11 of the telecentric optical system 300 and the object to be inspected, or a λ / 4 wave plate 232 is disposed between the light receiving lens 11 and the PBS 231, in the vicinity of the light receiving lens 11. The width of the PBS 231 and the λ / 4 wave plate 232 in the main scanning direction is longer than that of the light receiving element array 120. In addition, in an object-side telecentric optical system as shown in FIG. 22C, the light-receiving lens 11 on the light-receiving element array 120 side and the light-receiving lens 11 on the inspection object side in the double-telecentric optical system may be replaced by one or a set (one group) of light-receiving lenses 11 including a Gaussian lens consisting of multiple lenses, typically an achromat, a separate achromat, or an apochromat.

[0121] In FIG. 22D, the PBS 231 is disposed between the image intensifier 200 and the light receiving lens 11 on the light receiving element array 120 side of the telecentric optical system 300 on the optical axis of the light receiving lens 11. However, the PBS 231 having the aperture 21 on the entrance / exit surface may be disposed near the intersection of the light beams. Alternatively, the λ / 4 wave plate 232 is disposed between the light receiving lens 11 on the inspection object side of the telecentric optical system 300 and the inspection object, or the λ / 4 wave plate 232 is disposed near the light receiving lens 11 between the light receiving lens 11 and the PBS 231. The width of the PBS 231 and the λ / 4 wave plate 232 in the main scanning direction is longer than that of the light receiving element array 120. In FIG. 22D, all the λ / 4 wave plates 232 are disposed near the light receiving lens 11, but they may be disposed in any one of the locations in FIG. 22D.

[0122] 12. Modifications of the light receiving optical system Next, another embodiment of the light receiving optical system will be described. Fig. 23 shows a case where light is irradiated from a light source (polarized light source) 103 and received by a multi-stage (two stages in Fig. 23) telecentric optical system 300, and includes a telecentric optical system 300 (telecentric optical system 301) on the light source 103 side and a telecentric optical system 300 (telecentric optical system 302) on the light receiving element array 120 side. The telecentric optical system 301 includes a pair of light receiving lenses 311 and an aperture 211 provided therebetween. The telecentric optical system 302 includes a pair of light receiving lenses 312 and an aperture 212 provided therebetween. Fig. 23 shows a case where the diameter of the aperture 211 on the light source 103 side is made small and the aperture 212 on the light receiving element array 120 side at the rear stage is made to be of normal size. That is, two sets of telecentric optical systems 300 are arranged on the optical axis, and the diameter (aperture diameter) of the aperture 21 included in each telecentric optical system 300 is different. Specifically, the aperture 211 arranged on the upstream side of the telecentric optical system 300 is smaller than the aperture 212 on the downstream side. It is preferable that the diameter of the aperture 21 included in each telecentric optical system 300 is approximately the diffraction limit of the light receiving lens 11 or a diameter equivalent to approximately one pixel of the light receiving element.

[0123] In the embodiment, the diameter of the aperture 211 of the telecentric optical system 301 is 10 μm, and the diameter of the aperture 212 of the telecentric optical system 302 is 800 μm. This makes the diffraction effect prominent in the aperture 211, and the light that has been converted into a parallel light beam by the light receiving lens 311 on the inspection target side has an airy disc divergence angle θd of θd≈12.5° when emitted from the aperture 211 due to the aperture 211 having a diameter of 10 μm. By utilizing the diffraction divergence and moving the light receiving lens 311 at the rear stage of the telecentric optical system 301 slightly closer to the aperture 211, the light emitted from the aperture 211 passes through the light receiving lens 311, and while diffusing, passes through the light receiving lens 312 at the front stage of the telecentric optical system 302, and enters the aperture 212. The spread angle θd' of the Airy Disc due to the diffraction effect of the aperture 212 is θd'≈0.08°, which is very small, that is, the diffraction effect of the rear aperture 212 can be ignored. That is, the front (first) aperture 211 spreads the light beam, and the rear (second) aperture 212 captures the vicinity of the center of the spread light and returns it to a parallel light beam, so that only the straight light component can be selectively received. Therefore, the straight light component of the light scattered by foreign objects and defects in the inspection object can be more selectively received. As a result, foreign objects and defects buried in the inspection object can be detected with even higher accuracy.

[0124] In this embodiment, the telecentric optical system 300 has two stages, but if the number of stages is increased to three or four, more straight light components can be selectively received. In that case, only the aperture 21 on the light receiving element array 120 side is set to an aperture diameter capable of receiving parallel light beams, and the other apertures 21 are set to a diameter taking into consideration the spread caused by diffraction. The amount of spread light increases due to the diffraction effect, and the straight light components become weak, but this can be dealt with by making the image intensifier 200 multi-stage. For example, if it is assumed for simplicity that the diffracted light has a flat intensity distribution with respect to the optical axis, the amount of light is reduced to 1 / 6400 by the two-stage telecentric optical system 300, so a one-stage image intensifier 200 with a multiplication factor of several thousand to several tens of thousands is used. In the case of a three-stage aperture 21, if both of the upstream apertures 21 are 20 μm, the amount of light that ultimately passes through the aperture 21 on the light receiving element array 120 side is (1 / 1600) × (1 / 1600) ≈ 1 / 2,560,000. Therefore, in this case, a two-stage image intensifier 200 with a multiplication factor of hundreds of thousands to millions should be used.

[0125] In addition, the light receiving lens 311 at the rear of the telecentric optical system 301 arranged on the light source 103 side is slightly closer to the aperture 211. By doing so, the light beam spreads, and the spread of the spread light beam is reduced by the light receiving lens 312 at the front of the telecentric optical system 302, and the spread of the spread light beam approaches a parallel light beam, and only the parallel light beam that is originally required by the telecentric optical system 302 is selected by the aperture 212 and imaged on the light receiving element array 120. In other words, by once spreading the light beam and selecting a component of the spread light beam that is close to the optical axis, only the further straight light component can be received. Furthermore, in the method of extracting the straight light component by the time difference shown in Figures 18 to 21, in the combination of the image intensifier 200 and the light receiving element array 120 without using a lens system, the light receiving lens 11 in Figure 16A may be removed and only the image intensifier 200 and the light receiving element array may be arranged to face the inspection object.

[0126] FIG. 26A shows a schematic diagram of the case where the lens system is removed. If the solid angle (half angle) between the object to be inspected and one pixel of the light receiving element is about 1°, the light beam can be narrowed down to a value close to the NA of the telecentric optical system 300 used in the present invention, and the distance between the light receiving system and the object to be inspected in this case is about 7500 mm when NA=0.02. If it is equivalent to 600 dpi (equivalent to a pixel size of 42 μm), the optical path difference between adjacent pixels can be made relatively compact by bending the optical path many times using a member such as a plane mirror. The reflectance of the reflecting surface is 99.5%. Reflecting surfaces using dielectric multilayer films are generally implemented, and even with a number of reflections of about n=30 times, the final intensity of the light beam at the time of emission is 85% or more compared to the intensity at the time of incidence, so there is no problem in practical use. FIG. 26B shows a schematic diagram of the case where the optical path is bent and compacted using a plane mirror 303. Also, a reflective type can be realized by arranging the PBS 231 in front of the light receiving section of the image intensifier 200 in the same manner as the telecentric optical system 300 described above, collimating the light emitted from the linearly polarized light source by a collimation means (not shown), and using a means for shaping the collimated light into a sheet (not shown), adjusting it so that it is perpendicular to the angle of incidence of the transmitted light to the image intensifier 200, and arranging a λ / 4 wavelength plate and a polarizing filter near the prism 304 as in the above-described embodiment (variation example).

[0127] 26B, light from an object to be inspected is bent by a right-angle prism 304, and then repeatedly bent between a pair of parallel plane mirrors 303. The distance between the plane mirrors 303 is about 250 mm, and the light is bent 28 times, resulting in an optical path length of about 7500 mm. In this case, the optical path difference between adjacent pixels is about 1.5 mm, which is almost the same as the maximum resolution of 1.5 mm that can be discriminated by time difference, and therefore satisfies performance almost equivalent to that of the telecentric optical system 300.

[0128] 26A and 26B, the light receiving lens 11 and the aperture 21 are removed from the telecentric optical system 300, and a right-angle prism 304 and a pair of plane mirrors 303 are arranged opposite to the position of the light receiving lens 11. The pair of plane mirrors 303 are arranged so as to form parallel planes perpendicular to the conveying direction, and the right-angle prism 304 is arranged at one end of the plane mirror 303 on the side of the inspection object. The light beam emitted from the inspection object that enters the right-angle prism 304 is perpendicular to the direction in which the inspection object is conveyed, and the light beam subsequently enters the inclined surface that serves as the total reflection surface of the right-angle prism 304. One of the sides that form the right angle of the right-angle prism 304 is non-parallel to the plane mirror 303. The light beam emitted from the right-angle prism 304 is incident non-perpendicularly on the reflecting surface of one of the plane mirrors 303, and travels a long optical path in the depth direction of the plane mirror 303 while being reflected and bent between the pair of plane mirrors 303, and then emerges from the other end of the plane mirror 303 on the opposite side to the object to be inspected. The emitted light beam is incident on the light-receiving element array 120 via the image intensifier 200. A slit (not shown) having a long opening in the main scanning direction for blocking disturbance light may be arranged at the end of the plane mirror 303.

[0129] 13.Application to Tomography The inspection apparatus described above was intended for real-time inspection, but the inspection apparatus described in this embodiment is intended mainly for offline inspection. The aforementioned transmission type and reflection type optical systems can receive straight light by using the image intensifier 200 and delay circuit 210, so that it is possible to obtain a three-dimensional tomographic image (three-dimensional tomographic image or three-dimensional tomographic video) by rotating the inspection object and the optical system relatively and using Radon transform, two-dimensional Fourier transform, or successive approximation method used in X-ray CT. FIG. 24A shows an embodiment (schematic diagram). The configuration within the thick frame in the lower right of FIG. 24A corresponds to FIG. 16A, FIG. 22A, FIG. 22B, and FIG. 23. Also, the system may be configured using only the optical system and the image intensifier 200 without using the delay circuit 210.

[0130] Incidentally, the light receiving optical system described above is based on the telecentric optical system 300, but tomography does not require an imaging optical system. In order to obtain a tomographic image by determining the transmittance at an arbitrary position of the object to be inspected, the light emitted from the light source 103 is made incident on the object to be inspected, and the parallel light beam component of the beam emitted from the object to be inspected is received by the light receiving element array 120, and its output is obtained. Therefore, it is not necessarily necessary to form an image. The telecentric optical system 300 plays a role in making the light beam incident on the light receiving element array 120 as a light beam close to a parallel light beam by making the aperture diameter used therefor extremely small. However, if the aperture diameter is simply made small, the light emitted from the aperture 21 will spread due to the diffraction effect, so it is necessary to make only the light beam near the center, which is exactly the light path, incident on the light receiving element array 120. For this reason, two stages of telecentric optical systems 301, 302 are used, and the aperture 211 is narrowed in the telecentric optical system 301 in the front stage to diffract the incident light beam, and then the central part of the diffracted light is guided to the light receiving element array 120 by the telecentric optical system 302 in the rear stage in order to receive only approximately the central part of the diffracted light.

[0131] The present invention is a system similar to the parallel beam method and fan beam method in X-ray CT, but it uses the near-infrared region and can make lenses for the telecentric optical system 300 using glass materials or resins with refractive indexes, so that the light emitted from the light source 103 can be made into a parallel sheet beam relatively easily, the pixel size of the light receiving element can be made very small, and the scanning speed of the light receiving element array 120 is fast, so that a resolution and inspection speed exceeding those of X-ray CT can be realized. Furthermore, the light receiving optical system can be extended, so that scanning in the main scanning direction such as helical scanning used in X-ray CT is not necessary. However, helical scanning can also be used. In this case, the object to be inspected may be moved in the main scanning direction, or the entire optical system may be moved relative to the object to be inspected. By using an infrared light source, the problem of radiation exposure due to X-rays is also eliminated. In addition, by changing the distance (diameter) between the illumination optical system and the light receiving optical system in the gantry (broken line) shown in FIG. 24A, inspection according to the size of the object to be inspected is possible.

[0132] FIG. 24B is a schematic diagram showing a staggered arrangement of rectangular light receiving lenses 11 and light receiving element array 120 as shown in FIG. 25A, which will be described later, in a gantry. In this embodiment, the light receiving lenses 11 are arranged in one row. In FIG. 25A, the image intensifier 200 is omitted. Also, a sheet beam light source having a light intensity distribution with a rectangular cross section that is somewhat wide and long in the perpendicular direction to the circumference is arranged on a circumference centered on the light receiving element array 120, facing the light receiving element array 120. The light receiving element array 120 is a multi-line, and for example, a 256-line CCD sensor is commercially available. The CCD sensor has a lower readout frequency than the CMOS sensor, but has a good S / N ratio. The line rate is about 1 KHz on the market.

[0133] The configuration shown in FIG. 25A can be used to configure a three-dimensional tomographic image forming apparatus (three-dimensional tomographic image forming apparatus or three-dimensional tomographic video forming apparatus) using the above-mentioned optical line sensor. This three-dimensional tomographic image forming apparatus has an illumination optical system 100 that irradiates a parallel beam having a rectangular cross section in the optical axis direction and is arranged opposite a telecentric optical system 300. The illumination optical system 100 may include a light source 103, a collimator 130, and a condenser lens 104. The illumination optical system 100 and the telecentric optical system 200 are arranged on a circumference and rotate on the circumference. Light that has passed through an inspection object arranged approximately at the center of the circumference is received by the telecentric optical system 300, and a three-dimensional tomographic image (three-dimensional tomographic image or three-dimensional tomographic video) is constructed by a predetermined calculation. In the example of Figure 25A, multiple pairs of illumination optical systems 100 and telecentric optical systems 200 are arranged on a circumference, but it is sufficient that at least one pair of illumination optical systems 100 and telecentric optical systems 200 is arranged on a circumference and rotates on that circumference.

[0134] When the width Pg of the gantry in the rotation direction perpendicular to the main scanning direction of the light receiving element array 120 is Pg=2.5 mm, the gantry is rotated at a pitch of 2.5 mm, and when the radius rg of the gantry is rg=250 mm, the circumference Cg of the gantry is Cg≈1570 mm. Therefore, the reading of each multi-line light receiving element array 120 is a parallel output, and the number of repeated readings required for one rotation is 627 times, so the time Tg required for one rotation is Tg≈0.628 sec. Therefore, even including other signal processing, measurement and inspection can be performed in a short time. The number of repetitions of one light receiving element (multi-line sensor) is 628 times for one rotation of the gantry, so the angle pitch θg is θg≈0.57°. Figure 27 shows a schematic diagram of the case where the gantry is rotated at a pitch of 5 degrees, 10 degrees, and 20 degrees without gaps. In order to obtain data without gaps, the number of lines of the light receiving element array 120 or the length in the main scanning direction (number of elements) is changed. However, if compressed sensing technology is used during image reconstruction, thinning processing to a certain extent is also possible, enabling faster reading. For example, the above-mentioned θg ≒ 0.57° may be thinned out and output as θg = 1°.

[0135] Here, the tomography image processing method will be briefly described with reference to Fig. 24A. Considering the rotation coordinate system in which the number of lines is inclined at a certain angle θ with respect to the xy coordinates of the object to be inspected, one line segment is defined from the center of the circle to the circumference, and the relationship between the xy coordinates and the st coordinates for the inclined thick line in Fig. 24A is expressed by the following formula (1).

number

[0136] If the absorptance of the object being inspected at any point on the thick line s at the wavelength used is μt(s, θ), then the value integrated on the t axis represents all absorption on the thick line s. In other words, the total absorption amount μ(s, θ) at the angle θ on s is expressed by the following formula (2).

number

[0137] Generally, in xy coordinates, the Fourier transform Fμ(u, v) of μ(x, y) is expressed by the following formula (3).

number

[0138] Now, if the Fourier transform of μ(s, θ) in FIG. 24A with respect to the variable s is taken as Fμ(s, θ), Fμ(s, θ) is expressed by the following equation (4).

number

[0139] Then, Fμ(s, θ) is transformed into the following equation (5) by the inverse Radon transform, so that the concentration μ(x, y) at each point of the xy coordinates is obtained, and a tomographic image of the object to be inspected is obtained.

number

[0140] The above is a mathematical method for continuous values, and actual data is discrete. Therefore, when performing the transformation of equation (5), errors in the high frequency range become large, so a filter that emphasizes high frequency components is applied. There are various well-known filters, but they are omitted in the present invention. Also, obtaining equation (5) is called "reconstruction of a tomographic image" or "restoration of a tomographic image", and in addition to the above-mentioned Fourier transform method (discrete Fourier transform), there is also an iterative approximation method using simultaneous equations for the reconstruction of a tomographic image. In the present invention, the reconstruction of various well-known tomographic images is omitted. In the present invention, the method using the Fourier transform method will be called the "analytical reconstruction method" and the method using the iterative approximation method or the like will be called the "algebraic reconstruction method".

[0141] Furthermore, the calculation times required for Radon (inverse) transform, one-dimensional Fourier transform, two-dimensional Fourier (inverse) transform, and CT image reconstruction methods such as filtered back projection and iterative approximation are very short, making this method sufficient for tomography used in offline inspection.

[0142] FIG. 25B is a schematic diagram showing a case where the main scanning direction of the light receiving element array 120 is parallel to the direction in which the entire optical system rotates in the gantry. In FIG. 25B, the main scanning direction is approximately perpendicular to the tangent of the circumference on which the optical axis of at least one set of the illumination optical system 100 and the telecentric optical system 200 rotates, and the light receiving element array 120 is approximately parallel to the main scanning direction. In this case, the length of the light receiving element array 120 in the main scanning direction is approximately 15 mm, and the light receiving elements are arranged in a direction perpendicular to the main scanning direction (sub-scanning direction) without any gaps. If the first row of the light receiving element array 120 is three lines, the second row of the light receiving element array 120 and the third row of the light receiving element array 120 are shifted in a direction perpendicular to the rotation direction (usually the sub-scanning direction) to fill the gaps between the lines. In addition, the light receiving lens 11 has a rectangular shape.

[0143] Fig. 25C shows a schematic diagram of a light receiving optical system composed of a light receiving element array 120 and a light receiving lens 11. The image intensifier 200 is omitted. In Fig. 25C, the light receiving lens 11 of the telecentric optical system 300 has a rectangular shape in cross section in the optical axis direction.

[0144] However, a gap may be provided in the sub-scanning direction, and sub-scanning (short-distance helical scanning) may be performed for the gap during one rotation of the gantry. The number of rows of light-receiving element array 120 in the rotation direction of the gantry shown in Figures 25B and 25C is three. Therefore, it is sufficient to rotate at a pitch of about 45 mm. Alternatively, the signal capture interval may be set at a pitch of 45 mm in synchronization with the rotation of the gantry. The shape of the cross section of light-receiving lens 11 in the optical axis direction is rectangular.

[0145] As described above, in the case of FIG. 25A, the angle pitch θg during rotation in the gantry is θg≒0.57°, while in the case of FIG. 25B, θg≒10°. Therefore, if the line rate of the light receiving element array 120 in FIG. 25A and FIG. 25B is the same, a detection speed of about 18 times can be realized. That is, Tg≒36 msec, which is almost close to the TV rate, enables dynamic biological observation. By increasing or decreasing the radius of the gantry, it is possible to handle people and opaque micro animals as the object of inspection. Furthermore, if the optical system shown in FIG. 25A and FIG. 25B is made into a plurality of N sets and output in parallel to the subsequent calculation processing, the inspection time becomes N times faster on the same circumference, and not only a three-dimensional tomographic image but also a three-dimensional tomographic video can be acquired. For example, if the number of sets in FIG. 25A is increased to 18 sets, the inspection speed becomes the same as that in FIG. 25B, and a three-dimensional tomographic video can be acquired. FIG. 27 is a schematic diagram showing the relationship between the rotation angle pitch and the length of the light receiving element array 120. In FIG.

[0146] 14. Reflection Tomography Variation 1. FIG. 28A shows a reflection type tomography. X-ray tomography is a transmission type, and the light source 103 and the light receiving system face each other with the gantry at the center. Therefore, it is difficult to align the optical axis, and the size of the device is large. In this modification, a reflection type that overcomes this weakness is proposed. In this modification, the aperture 21 of the reflection type optical system is one, and instead the focal length of the light receiving lens 11 is made longer. In addition, regarding the NA of the light receiving lens 11, the depth of field of the light receiving lens 11 (the light receiving lens 11 on the side of the inspection object) is made deeper than the thickness of the inspection object in the optical axis direction. This makes it possible to receive a component close to a linear light. Although the amount of light received is less than that of the transmission type, the output intensity on the light source 103 side is adjusted so that an amount of light sufficient to satisfy the gain (photon counting) of the light receiving element is obtained, and the image intensifier 200 is made multi-staged to increase the sensitivity.

[0147] In this modification, since the reflecting optical system is arranged on the circumference of the gantry, the circumference must be secured, but since it is possible to use a part of the circumference for image reconstruction, the tomography device can be made compact. In the modification 1, the depth of field of the light receiving lens 11 on the inspection object side is made deeper than the thickness of the inspection object in the optical axis direction, but as shown in FIG. 28B, a collimator 130 may be arranged on the light source 103 side to make the light beam incident on the inspection object approximately parallel to the optical axis (parallel light beam). In addition, the focal position of the telecentric optical system 300 on the inspection object side may be approximately coincident with the focal position of the light receiving lens 11 on the light receiving element array 120 side.

[0148] The configuration of FIG. 28A or FIG. 28B is specifically as follows. (1) While maintaining the NA of the telecentric optical system itself, the collimator 130 on the light source 103 side collimates the light incident on the inspection object. (2) The focal length of the telecentric optical system is made long, and the NA is made equal to the collimation angle of the light source 130. (3) The aperture diameter is set to be equal to the beam divergence angle due to the diffraction limit (the NA of the telecentric optical system and the divergence angle of the light source 103 are set to be equal when converted into NA). (4) The depth of field of the telecentric optical system is set to be equal to or greater than the thickness of the object to be inspected.

[0149] Variation 2. Fig. 28C shows an optical system capable of receiving a more parallel light beam than the other modified examples. In this modified example, the optical system of Fig. 23 is arranged after the optical system shown in Fig. 28A, making it possible to receive a more parallel light beam. Since the optical system of Fig. 28A and the optical system of Fig. 23 both have a deep depth of field and focal depth, it is sufficient to approximately match the focal position of the optical system of Fig. 28A on the side of the inspection object with the focal position of the optical system of Fig. 23 on the side of the light receiving element array 120.

[0150] The above-mentioned modified example realizes tomography capable of acquiring a three-dimensional tomographic image (three-dimensional tomographic image or three-dimensional tomographic video) simply by moving in an arc, which allows the device to be made smaller, simpler, and less expensive. The three-dimensional tomographic image forming device according to the above-mentioned modified example has an illumination optical system 100 that irradiates a parallel beam having a rectangular cross section in the optical axis direction, and the illumination optical system 100 and a telecentric optical system 300 are arranged on a circumference. The telecentric optical system 300 rotates on the circumference or a part of the circumference, and light reflected from an inspection object arranged at approximately the center of the circumference is received by the telecentric optical system 300, and a three-dimensional tomographic image is constructed by a predetermined calculation.

[0151] In the three-dimensional tomographic image forming apparatus according to the above modification, the main scanning direction may be substantially perpendicular to a tangent to a circumference on which the optical axes of at least one set of the illumination optical system 100 and the telecentric optical system 300 rotate, and the light receiving element array 120 may be substantially parallel to the main scanning direction. The light receiving lens 11 of the telecentric optical system 300 may have a rectangular shape in a cross section in the optical axis direction. A plurality of sets of the illumination optical system 100 and the telecentric optical system 300 may be arranged on the circumference. It goes without saying that the above-mentioned reflective optical system can be used not only for tomography but also as a foreign body / defect detection device. [Explanation of symbols]

[0152] 10 Light source section 11 Receiving lens 12 Light receiving part 20 focal plane 103 Light source 104 Condenser Lens 105 Cylindrical Lens 110 Lens Holder 111 Shielding part 120 Photodetector array 131 Red Laser Diode 132 Green Laser Diode 133 Blue Laser Diode 134 Light source board 135 Heatsink

Claims

1. An optical line sensor that reads an object to be inspected being transported in the sub-scanning direction using a reading line extending in the main scanning direction, Multiple light-receiving lenses arranged along the main scanning direction, It comprises a plurality of light-receiving elements arranged in a line along the main scanning direction, which receive light that has passed through the plurality of light-receiving lenses, The plurality of light-receiving lenses are arranged spaced apart from each other so as not to overlap in the field of view of each light-receiving lens. The aforementioned multiple light-receiving lenses, together with the aperture, constitute a telecentric optical system. The plurality of light-receiving elements are arranged in an array to form a light-receiving element array that forms at least one row or more of the reading lines. An optical line sensor characterized in that an image intensifier, including a photocathode, a microchannel plate, and a phosphor surface, is positioned opposite the light-receiving surfaces of the plurality of light-receiving elements.

2. The gate operation time of the image intensifier and the emission time of the light source illuminating the object to be inspected are both shorter than the storage time of the light-receiving element. In one cycle of the series of operations of the light-receiving element, the light source, and the image intensifier, The light source's light emission start time and light emission end time are earlier than the light receiving element's storage operation start time. The start time of the gate operation of the image intensifier is earlier than the end time of the accumulation operation of the photodetector, and the end time of the accumulation operation of the photodetector is earlier than the end time of the gate operation of the image intensifier. At least one of the image intensifier and the light source is equipped with a delay circuit, which allows the operating times of the light source and the image intensifier to be adjusted. The optical line sensor according to claim 1, characterized in that the light-receiving lens is formed in a rectangular shape when viewed from a direction perpendicular to the main scanning direction and the sub-scanning direction, and the optical axis of the light-receiving lens is located in the center of the width of the light-receiving lens in the sub-scanning direction.

3. At least two sets of the telecentric optical systems are arranged on the optical axis, and the aperture located upstream of the telecentric optical system is smaller than the aperture located downstream. The optical line sensor according to claim 1, characterized in that the plurality of light-receiving lenses are spaced apart from each other so as to be less than or equal to the width of the light-receiving lens in the main scanning direction.

4. The aforementioned multiple light-receiving elements are arranged in two or more rows in an array to constitute a plurality of light-receiving element arrays. The optical line sensor according to claim 1, characterized in that the plurality of light-receiving lenses are arranged in a number corresponding to the plurality of light-receiving element arrays, and the optical axis of the light that passes through each light-receiving lens and is guided to each light-receiving element array penetrates the central part of each light-receiving element array, or a position parallel to the sub-scanning direction from the central part of each light-receiving element array.

5. The aforementioned arrays of multiple photodetectors are arrays of multiple photodetectors arranged in each of the two rows of reading lines, each array being shorter than the respective reading line, and the arrays of photodetectors arranged in one reading line and the arrays of photodetectors arranged in the other reading line are arranged alternately in a staggered pattern along the main scanning direction. The light-receiving lens is formed in a rectangular shape when viewed from a direction perpendicular to the main scanning direction and the sub-scanning direction. The optical line sensor according to claim 4, characterized in that the light-receiving lens corresponds one-to-one with each of the staggered-arranged light-receiving element arrays, and the optical axis of the light-receiving lens penetrates the central part of the staggered-arranged light-receiving element arrays.

6. It is further equipped with multiple light sources that illuminate the object to be inspected, The plurality of light sources are arranged parallel to the reading line, the optical axes of the plurality of light sources intersect with the optical axes of light that passes through the plurality of light-receiving lenses and is guided to the plurality of light-receiving elements, and are positioned at any position on a virtual plane that intersects with the optical axes of light that passes through the plurality of light-receiving lenses and is guided to the plurality of light-receiving elements, and furthermore, the light sources are positioned in the center between adjacent light-receiving lenses. The optical line sensor according to claim 1, characterized in that the plurality of light sources include light sources having a plurality of different wavelengths, and the plurality of light sources having a plurality of different wavelengths are treated as one unit, with a plurality of such units of light sources arranged in the main scanning direction.

7. The system further comprises a focusing lens that concentrates light beams from the aforementioned multiple light sources, The optical line sensor according to claim 6, characterized in that the focusing lens includes a first focusing lens in which the power in the main scanning direction is greater than the power in the sub-scanning direction, and a second focusing lens in which the power in the sub-scanning direction is greater than the power in the main scanning direction.

8. Further comprising a focusing lens for concentrating light beams from the plurality of light sources, The optical line sensor according to claim 6, characterized in that the condensing lens is a single lens body, and the power of the lens body in the main scanning direction is greater than the power in the sub-scanning direction.

9. The optical line sensor according to claim 7, characterized in that the power in the main scanning direction can be adjusted by the first and second focusing lenses.

10. The first focusing lens is a lenticular lens or a series of prisms. The optical line sensor according to claim 7, characterized in that the second focusing lens is a Fresnel lens or a cylindrical lens.

11. The plurality of light sources include white LEDs, or include red LEDs, green LEDs and blue LEDs, or include laser diodes. A light source substrate on which the aforementioned multiple light sources are mounted, The optical line sensor according to claim 6, further comprising a heat sink attached to the light source substrate.

12. The aforementioned plurality of light-receiving lenses are arranged in a single row along the main scanning direction. The plurality of light-receiving lenses are arranged such that the ends of adjacent light-receiving lenses in the main scanning direction are connected to each other. The aforementioned plurality of light-receiving lenses have the same shape when viewed from a direction perpendicular to the main scanning direction. The optical line sensor according to claim 1, characterized in that the plurality of light-receiving lenses are arranged in a state in which adjacent light-receiving lenses are inverted and connected.

13. The aforementioned multiple light-receiving elements are arranged in two or more rows in an array to constitute a plurality of light-receiving element arrays. The array of multiple light-receiving elements is positioned at the center of each of the multiple light-receiving lenses in the main scanning direction, and is arranged alternately in a staggered pattern along the main scanning direction. The optical line sensor according to claim 12, further comprising a plurality of light sources that illuminate the object to be inspected and which correspond one-to-one with the plurality of light-receiving element arrays.

14. The plurality of light sources are located near the ends of each of the plurality of photodetector arrays in a cross-sectional view along the optical axis of the light source, and in the central part of the plurality of photodetector arrays facing each other in the sub-scanning direction. The optical line sensor according to claim 13, characterized in that the plurality of light sources have multiple lines of illumination light intensity distribution in the sub-scanning direction corresponding to the plurality of light-receiving element arrays.

15. The device further comprises a shielding portion that extends from between the ends of adjacent light-receiving lenses in the main scanning direction toward the plurality of light-receiving elements and has a width at least equal to or greater than the width of the light-receiving lenses in the sub-scanning direction, The optical line sensor according to claim 12, characterized in that the shielding portion protrudes from the light-receiving lens in the direction of the object to be inspected.

16. The aforementioned microchannel plate has an auto-gate function, and the light source side has a function to adjust the light intensity in conjunction with the auto-gate function. (a) A polarizing beam splitter is positioned on the optical axis of the light-receiving lens between the image intensifier and the light-receiving lens on the light-receiving element array side of the telecentric optical system, A λ / 4 wave plate is placed between the light-receiving lens on the light-receiving element array side of the telecentric optical system and the polarizing beam splitter, or between the light-receiving lens on the object-to-inspection side of the telecentric optical system and the object-to-inspection, or between the light-receiving lens on the light-receiving element array side of the telecentric optical system or the light-receiving lens on the object-to-inspection side and the aperture. The polarization beam splitter and the λ / 4 wave plate have a width in the main scanning direction that is longer than the photodetector array, and the photodetector lens has a width in the sub-scanning direction that is smaller than the width in the main scanning direction, or (b) A polarizing beam splitter is positioned on the optical axis of the light-receiving lens between the image intensifier and the light-receiving lens on the light-receiving element array side of the telecentric optical system, and near the aperture, or a polarizing beam splitter with apertures on its input and output surfaces is positioned near the intersection of each light beam. In the telecentric optical system, either a λ / 4 wave plate is placed between the light-receiving lens on the object-to-inspection side and the object-to-inspection, or a λ / 4 wave plate is placed near the light-receiving lens between the light-receiving lens and the polarizing beam splitter. The optical line sensor according to claim 1, characterized in that the polarization beam splitter and the λ / 4 wave plate have a width in the main scanning direction that is longer than the photodetector array, and the photodetector lens has a width in the sub-scanning direction that is smaller than the width in the main scanning direction.

17. Removing the light-receiving lens and aperture from the telecentric optical system, a right-angle prism and a pair of planar mirrors are positioned opposite each other at the position of the light-receiving lens, and the pair of planar mirrors are positioned so as to be parallel planes perpendicular to the transport direction, the right-angle prism is positioned at one end of the planar mirror on the side facing the object to be inspected, the light ray emitted from the object to be inspected that is incident on the right-angle prism is a light ray perpendicular to the direction in which the object to be inspected is transported, the light ray then incident on the inclined surface that is the total reflection surface of the right-angle prism, one of the right-angle sides of the right-angle prism is non-parallel to the planar mirror, the light ray emitted from the right-angle prism is incident non-perpendicular to the reflection surface of one of the planar mirrors, travels along a long optical path in the depth direction of the planar mirrors while reflecting and bending between the pair of planar mirrors, and then is emitted from the other end of the planar mirror opposite to the object to be inspected, and the emitted light ray is incident on the light-receiving element array via the image intensifier, The optical line sensor according to claim 1, characterized in that a slit having an opening long in the main scanning direction for blocking ambient light is arranged at the end of the plane mirror.

18. Either the depth of field of the light-receiving lens on the object to be inspected is greater than the thickness of the object in the optical axis direction, or a collimator is placed on the light source side to make the light beam incident on the object to be inspected parallel to the optical axis. The optical line sensor according to claim 16, characterized in that the focal position of the telecentric optical system on the object to be inspected side is matched with the focal position of the light-receiving lens on the light-receiving element array side.

19. An image processing method using an optical line sensor as described in claim 13, For output signals from two photodetectors positioned at the same location in the main scanning direction but separated in the sub-scanning direction, one of the output signals from one photodetector overlaps with the output signal from the other photodetector. This signal is then combined with the other signal by applying a correction based on the ratio of the two output signals, which has been pre-corrected using a reference medium, to obtain a single line of output signals corresponding to the reading line. An image processing method characterized in that, with respect to the output signals from two photodetectors that are spaced apart in the sub-scanning direction at the same position in the main scanning direction, if the output signal from one photodetector is lower than the output signal from the other photodetector and the output signal from one photodetector falls below a threshold, the output signal from the other photodetector is interpolated, and the interpolated output signal is combined with the output signal from a photodetector at another position in the main scanning direction relative to the one photodetector to obtain a single line of output signals corresponding to the reading line.

20. An image processing system using an optical line sensor as described in claim 13, For output signals from two photodetectors positioned at the same location in the main scanning direction but separated in the sub-scanning direction, one of the output signals from one photodetector overlaps with the output signal from the other photodetector. This signal is then combined with the other signal by applying a correction based on the ratio of the two output signals, which has been pre-corrected using a reference medium, to obtain a single line of output signals corresponding to the reading line. An image processing system characterized in that, with respect to the output signals from two photodetectors that are spaced apart in the sub-scanning direction at the same position in the main scanning direction, if the output signal from one photodetector is lower than the output signal from the other photodetector and the output signal from one photodetector falls below a threshold, the output signal from the other photodetector is interpolated, and the interpolated output signal is combined with the output signal from a photodetector at another position in the main scanning direction relative to the one photodetector to obtain a single line of output signals corresponding to the reading line.

21. A three-dimensional tomographic imaging apparatus using an optical line sensor as described in claim 1, A parallel beam with a rectangular cross-section in the optical axis direction is irradiated, and an illumination optical system is positioned opposite the telecentric optical system, and the illumination optical system and the telecentric optical system are arranged on a circumference. At least one pair of the illumination optical system and the telecentric optical system rotates on the circumference, and the light transmitted through the object to be inspected, which is placed in the center of the circumference, is received by the telecentric optical system, and a three-dimensional tomographic image is constructed by predetermined calculations. A three-dimensional tomography apparatus characterized in that the main scanning direction is perpendicular to the tangent to the circumference on which the optical axes of at least one pair of illumination optical systems and the telecentric optical system rotate, and the photodetector array is parallel to the main scanning direction.

22. The light-receiving lens of the telecentric optical system has a rectangular shape in the cross-section along the optical axis. The three-dimensional tomography apparatus according to claim 21, characterized in that a plurality of sets of the illumination optical system and the telecentric optical system are arranged on the circumference.

23. The three-dimensional tomography apparatus according to claim 21, characterized in that two or more sets of the telecentric optical systems are arranged on the optical axis, and the diameter of the aperture included in each telecentric optical system is different.

24. A three-dimensional tomography apparatus according to claim 21, characterized in that one or more sets of telecentric optical systems are arranged on the optical axis, and the diameter of the aperture included in each telecentric optical system is the diffraction limit of the light-receiving lens or the diameter corresponding to one pixel of the light-receiving element.

25. A three-dimensional tomographic imaging apparatus using an optical line sensor as described in claim 18, The illumination optical system has a parallel beam that irradiates with a rectangular cross-section in the optical axis direction, and the illumination optical system and the telecentric optical system are arranged on a circle. The telecentric optical system rotates on or around the circumference, receives light reflected from the object to be inspected located in the center of the circumference, and constructs a three-dimensional tomographic image through predetermined calculations. The main scanning direction is perpendicular to the tangent to the circumference on which the optical axes of at least one pair of illumination optical systems and the telecentric optical system rotate, and the photodetector array is parallel to the main scanning direction. The light-receiving lens of the telecentric optical system has a rectangular shape in the cross-section along the optical axis. A three-dimensional tomography apparatus characterized in that multiple sets of the illumination optical system and the telecentric optical system are arranged on the circumference.

26. An image processing method characterized by obtaining a three-dimensional tomographic image using a three-dimensional tomographic image forming apparatus according to claim 21 or 25, and using either an analytical reconstruction method or an algebraic reconstruction method as an image reconstruction method.

Images