Inspection device and inspection method

Abstract

The present invention addresses the problem of providing an inspection device that is capable of inspecting samples conveyed at high speed with high resolution, high sensitivity, high S / N, and high precision, and the main purpose of the present invention is to provide an inspection device that is equipped with at least: a conveyance mechanism that conveys samples to be inspected; a radiation source that emits radiation in a radial form into a region through which the conveyed samples pass; a detector that is disposed so as to be able to sense radiation that has passed through the conveyed samples and that converts the detected radiation into an electrical signal; a movement mechanism that moves the detector in the direction of conveyance of the samples by the conveyance mechanism; and an assessment means that determines the presence / absence of defects in the samples on the basis of sample image- information from the detector and obtained during a period in which the samples are passing through the region, and on the basis of movement information regarding the samples from the conveyance mechanism. The invention also proposes using this device to inspect for foreign matter in the samples.

Description

Inspection device and inspection method

[0001] The present invention relates to an inspection device and an inspection method for inspecting an object to be inspected (hereinafter sometimes referred to as a "sample") for minute defects and the like.

[0002] Generally, inspections are performed to detect minute metal impurities and defects in samples using inspection methods suited to each type. Non-destructive inspections involve, for example, irradiating the sample with electromagnetic waves such as X-rays and detecting the transmitted electromagnetic waves. Differences in the transmittance of the electromagnetic waves occur depending on the presence or absence of impurities or defects in the sample, as well as the material, and these differences can be detected to generate two-dimensional images, allowing the internal condition of the sample to be inspected.

[0003] In-line inspection, in which inspection is performed simultaneously with the manufacturing process on an assembly line, is known as a useful means for shortening takt time. For example, when an X-ray inspection device is used to detect foreign matter in a sample, a known device is used in-line inspection. The X-ray source and an X-ray detector with a linear array of pixels are arranged facing each other on a line substantially perpendicular to the surface on which the sample is placed, while the sample is moved in a predetermined direction by a conveying means. This device acquires image data at a period corresponding to the conveying speed of the sample, and linear X-ray transmission images generated based on the image data are sequentially arranged and combined to generate a two-dimensional image of the sample (see Patent Document 1).

[0004] JP 2015-64336 A JP 2020-3322 A JP 2020-143922 A JP 2000-221144 A

[0005] In the above-mentioned in-line inspection process, there is a demand for an inspection device and an inspection method that can accurately inspect minute defects with high resolution and in a short time. As an in-line inspection using electromagnetic waves such as X-rays, a method has been proposed in which a radiation source and a detector are fixed and the sample is transported between them for inspection, as described in Patent Documents 1, 2, and 3. However, if multiple defects overlap on a line connecting the X-rays and the detector, they are detected as a single defect, and defect information such as the number, size, and position of the defects becomes inaccurate, which can lead to an erroneous pass / fail judgment.

[0006] Furthermore, while increasing the sample transport speed is desirable for improving productivity, obtaining good images requires a high-scan rate detector capable of high-speed signal processing. Patent Document 2 addresses high-speed inspections by sharing the data readout from the detector's detection elements. However, acquiring images at high speeds results in only a short acquisition time, a small amount of electromagnetic wave per output, and a dark image with a low S / N ratio. Patent Document 3 proposes pre-setting a judgment area and transferring and processing data only for pixels corresponding to that area. However, even with these technologies, there are limits to achieving both high sample transport speed and high sensitivity. The faster the sample transport speed and the more accurate the inspection image, the more stringent the requirements for the detector. Furthermore, in the case of X-ray inspection, the conditions for the afterglow characteristics of the scintillator, a phosphor that converts X-rays into visible light, become more stringent, making it even more difficult to achieve high speed, high sensitivity, high S / N, and high resolution.

[0007] In Patent Document 4, imaging is performed while the detector is moving in the sample's travel direction, thereby reducing the relative speed difference between the detector and the sample, allowing the detector to follow high-speed transport. In fact, imaging while the detector is moving reduces image lag and improves temporal resolution. However, low-contrast defects, which have high electromagnetic transmittance and are difficult to distinguish from surrounding normal areas, are difficult to detect. Furthermore, as the sample transport speed increases, the overall image becomes darker, and the noise (i.e., image roughness) worsens relative to the overall brightness, resulting in a low S / N ratio, making low-contrast defects even more difficult to detect. While increasing the exposure time can brighten the image, this method results in a decrease in temporal resolution, resulting in a trade-off between temporal resolution and S / N.

[0008] SUMMARY OF THE INVENTION In view of the above problems, an object of the present invention is to provide an inspection apparatus capable of inspecting a sample transported at high speed with high resolution, high sensitivity, and high S / N ratio with high precision.

[0009] The inspection apparatus of the present invention that solves the above-mentioned problems comprises at least a transport mechanism that transports a sample to be inspected, a radiation source that radially irradiates a region through which the transported sample passes with radiation, a detector that is positioned so as to be able to detect radiation that has passed through the transported sample and that converts the detected radiation into an electrical signal, a movement mechanism that moves the detector in the direction in which the sample is transported by the transport mechanism, and determination means that determines the presence or absence of defects in the sample from image information of the sample from the detector obtained during the time the sample passes through the region and from movement information of the sample from the transport mechanism, and preferably the transport mechanism and movement mechanism are an inspection apparatus that moves the sample and the detector so that, when Vw is the movement speed of the sample by the transport mechanism and Vd is the movement speed component of the detector by the movement mechanism in the same direction as the movement direction of the sample, the ratio Vd / Vw of Vd to Vw is not more than twice the ratio (FDD / FOD) of the distance between the radiation source and the detector (FDD) to the distance between the radiation source and the sample.

[0010] Furthermore, the above-mentioned inspection device is more preferably an inspection device equipped with a thickness measuring device that measures the thickness of the sample, and a calculation means that calculates the position of a defect in the sample based on the thickness information of the sample measured by the thickness measuring device and the image information of the sample from the detector.

[0011] According to the present invention, it is possible to transport a sample at a high speed, exceeding the transport speed constraints imposed by the scan rate of the detector, and it is possible to obtain high-sensitivity, high-S / N, high-resolution images and inspect minute defects with high precision.

[0012] 1 is a schematic diagram showing a general configuration for explaining one embodiment of the present invention; FIG. 1 is a schematic diagram showing a general configuration for explaining one embodiment of the present invention; FIG. 1 is a schematic diagram showing an internal configuration for explaining one embodiment of the present invention; FIG. 1 is a schematic diagram showing the layout relationship of a radiation source, a sample, and a detector; FIG. 1 is a schematic diagram showing an example in which an endless annular conveying belt is used as the moving mechanism of the detection unit; FIG. 2 is a diagram explaining an example of the spatial layout of a group of defects; FIG. 2 is an explanatory diagram showing an image taken at time ta; FIG. 2 is an explanatory diagram showing an image taken at time tb; FIG. 2 is an explanatory diagram showing an image taken at time tc; FIG. 3 is a schematic diagram showing a projected image of defect 15-(1) at each time; FIG. 4 is a diagram explaining the step of determining the depth at which the defect exists; FIG. 4 is an image seen from above of the object to be inspected used for measurement in the Examples section, where (a) is the overall image and (b) is an enlarged image near an object equivalent to a foreign substance; FIG. 5 is an image of the area near an object equivalent to a foreign substance in the object to be inspected taken by an X-ray TDI camera, where the black dot pointed to by the arrow is the image of the object equivalent to a foreign substance.

[0013] The present invention will be described below with reference to the drawings, taking as an example an inspection using an X-ray generator as a radiation source. However, the specific example described below illustrates and explains one embodiment of the present invention, and the present invention should not be interpreted as being limited to this specific example. For example, it is easily understood that the present invention can be applied to other radiation sources as long as they can penetrate a sample. Of course, the specific example described below can be modified and implemented without departing from the spirit of the present invention.

[0014] As the most typical example, we will explain an example in which the radiation source and the detector are arranged on either side of the sample transport plane while the sample is being transported, and the sample transport plane and the detector movement line are parallel. Here, the distance at which the distance between the X-ray focus, which is the divergence point of the X-ray beam, and the detector is shortest is called the FDD (Focus to Detector Distance), and the distance at which the distance between the X-ray focus and the sample is shortest is called the FOD (Focus to Object Distance) (see Figure 5).

[0015] <Inspection Apparatus> FIGS. 1 and 2 are schematic diagrams illustrating the overall configuration of one embodiment of the inspection apparatus of the present invention. FIG. 1 shows an in-line inspection apparatus in which samples are transported sheet by sheet, while FIG. 2 shows an in-line inspection apparatus for roll-to-roll inspection of continuous films, sheets, etc. The inspection apparatus includes a housing 1. Housing 1 is equipped with an inlet 3 through which the sample is transported and an outlet 4 through which the sample is transported. When ionizing radiation such as X-rays is used as a radiation source, for the safety of operators, housing 1, inlet 3, and outlet 4 are equipped with shielding walls capable of blocking ionizing radiation, such as lead or stainless steel, on the inner walls, and the opening of outlet 3 is usually equipped with a rubber shielding cover to prevent leakage of ionizing radiation to the outside. In the case of FIG. 1 , a sample (such as sample 2(a) or sample 2(b)) is placed on a sample transport belt 6, which is moved by a sample transport roll 5 to introduce the sample into the housing. In the case of FIG. 2, the sample 2 comes into contact with the sample transport roll 5 and is transported directly into the housing 1 .

[0016] FIG. 3 is a schematic diagram showing the interior of FIG. 1 , and FIG. 4 is a schematic diagram showing the interior of FIG. 2 . In the case of FIG. 3 , the housing 1 contains the transported samples (sample 2(a) or sample 2(b)), a thickness measurement sensor 7 that measures the thickness of the sample, a radiation source 8 that irradiates radiation, and a detector 10 that detects radiation 9 and converts it into an electrical signal. In this example, the thickness measurement sensor 7 is located upstream inside the housing 1, just after the sample has passed through the inlet 3, and measures the thickness of the sample. The thickness measurement sensor 7 measures the height profile of the sample from the surface of the sample transport belt 6, thereby calculating the thickness H of each transported sample. In the case of FIG. 4 , since the sample is film-like, the thickness measurement sensors 7 are located above and below the sample, and the distance between the upper and lower measurement sensors and the sample is measured. If the distance between the upper measurement sensor and the sample 2 is Ht, the distance between the lower measurement sensor and the sample 2 is Hb, and the distance between the upper and lower measurement sensors is Hd, the thickness H of the sample 2 can be calculated as Hd - Ht - Hb. When radiation is emitted from the radiation source 8, which is a point light source, the radiation spreads radially, and the space irradiated by the radiation is usually conical. Because the radiation is emitted radially, defects located closer to the radiation source 8 in the sample are imaged as large images on the detector 10, while defects located further away are imaged as small images on the detector 10. When an X-ray source is used as the radiation source 8, additional equipment is required, such as a high-voltage generator that supplies high-voltage power to the X-ray tube and an X-ray controller that controls the tube voltage and tube current, but these are not shown in the figure. In addition, the signal (image information) output from the detector 8 is sent to an image processing device that processes the signal and is processed by a data processing unit (not shown) to determine whether the signal is good or bad. The signal processing unit and data processing unit can also be housed inside the housing 1.

[0017] <Thickness Measurement Sensor> The thickness measurement sensor 7 is used to accurately determine the surface position of the sample. Multiple sensors may be arranged in a direction in which more measurement points are desired, or one sensor may be capable of multi-point measurement. The thickness measurement sensor is installed, for example, upstream of the sample conveyor belt and performs sequential measurements while the sample is being conveyed. While a laser triangulation displacement meter, laser interferometer, ultrasonic distance meter, eddy current displacement sensor, or stylus displacement meter is used as the measurement instrument, laser triangulation displacement meter is preferred due to its non-contact measurement capability, its insensitivity to the material of the test object, its response speed, and ease of multi-point measurement. Specifically, a preferred method is the light-section method, in which a linear beam generated by a laser light source or the like is irradiated onto the sample and the trajectory of the reflected and scattered light corresponding to the surface shape is obtained as a height profile. The measurement instrument measures the thickness of the sample by reading the position of the light trajectory using a light-receiving means such as an image sensor and quantifying the surface shape of the sample surface.

[0018] <Radiation Source> In the inspection device of the present invention, the radiation source is not particularly limited as long as it can emit radiation that can penetrate the sample. However, a radiation source that generates electromagnetic waves is desirable because it is easy to handle and can be expected to provide high measurement accuracy, and an X-ray source is particularly preferred. The X-ray source that can be used is not particularly limited, and any known X-ray source can be used. For example, it is preferable to use a microfocus X-ray tube, which has an X-ray focus of 50 μm or less, as the X-ray source. There is no particular limit to the tube voltage of the X-ray tube. Note that the radiation source may be, instead of X-rays, a radiation source that can irradiate other radiations or electromagnetic waves, such as gamma rays or neutron rays, and irradiate radially from a point light source, i.e., a divergence point, depending on the type of sample and the inspection mode.

[0019] <Detector> A detector has the function of detecting radiation emitted from a radiation source and converting it into an electrical signal. There are no particular limitations on the type of detector as long as it is an element that detects the emitted radiation and generates an output according to the detection intensity. For example, if the radiation irradiating the sample is X-rays, an indirect conversion type detector element can be used, in which the X-rays are first converted into visible light by a scintillator and then received by a photodiode to generate an output, or a direct conversion type detector element that uses a semiconductor such as a-Se or CdTe and directly converts X-rays into an electrical signal to generate an output.

[0020] In addition, multiple detectors may be arranged, and examples include a line sensor in which elements are arranged in a straight line, a TDI (Time Delay Integration) sensor, or a detector having light-receiving elements in which elements are arranged in a planar manner, such as a CCD (Charge-Coupled Device) image sensor or a CMOS (Complementary Metal Oxide Semiconductor) image sensor, or a general-purpose two-dimensional X-ray detector, a flat panel detector (FPD). Indirect conversion FPDs are not subject to restrictions on operating temperature, etc., and are also superior in mechanical strength compared to direct conversion detectors. Therefore, indirect conversion X-ray detectors are easy to handle. Furthermore, indirect conversion FPDs preferably include a cell-type scintillator. In indirect conversion FPDs, a scintillator panel is used to convert radiation into visible light. The scintillator panel contains a phosphor that emits light in response to X-rays, such as cesium iodide (CsI). The phosphor emits visible light in response to the radiated X-rays. This light emission is converted into an electrical signal by a thin-film transistor (TFT), CCD, or CMOS, and the electrical signal is then transferred to convert the X-ray information into digital image information. However, indirect conversion FPDs tend to have poor image sharpness due to factors such as the scattering of visible light by the phosphor itself when the phosphor emits light. On the other hand, FPDs that employ a cell-type scintillator have phosphors filled in cells separated by partitions, thereby reducing the effects of scattering of visible light. As a result, FPDs equipped with a cell-type scintillator have high image sharpness and can detect minute defects and their locations with high accuracy. Because a large-area, highly sharp cell-type scintillator can be easily formed, a cell-type scintillator fabricated by photolithographically processing partitions using a photosensitive paste is more preferred.

[0021] Furthermore, in the present invention, it is preferable to use a flexible detector that has flexibility and can be bent. Taking a flat panel detector as an example, a flexible detector can be obtained by arranging a detection unit with pixels arranged in a matrix on a substrate formed of a flexible resin or the like that can be bent. Each pixel is provided with, for example, a photodiode that passes a current corresponding to the total energy of the received electromagnetic wave, and a pixel circuit that controls the driving of the photodiode. This allows each pixel to output an electrical signal corresponding to the total energy of the received light to a control unit.

[0022] <Detector Moving Mechanism> The inspection device of the present invention is equipped with a moving mechanism that moves the detector along the direction in which the sample is transported by the transport mechanism. As will be described later, the provision of such a moving mechanism makes it possible to perform inspections with high resolution, high sensitivity, and a high S / N ratio even at high speeds. The moving mechanism preferably has an endless, annular transport belt, and the detector is placed on this transport belt to perform the inspection.

[0023] A specific embodiment will be described with reference to FIG. 3. In the example of FIG. 3, the movement mechanism for moving the detector is driven by a driving means 13, with the conveyor belt 12 moving in the direction of sample transport. To capture images while moving the detector 10, stable movement, such as high uniformity of speed and linearity, is required. Therefore, it is preferable to provide an auxiliary means for assisting the smooth movement of the conveyor belt and a buffer means for absorbing tensile and compressive forces generated by the movement of the conveyor belt. Examples of such means include bearings and dampers with sufficient stress absorption capacity. To prevent tangling of the power supply cable for supplying power to the detector and the communication cable for transmitting and receiving signals for image data acquisition, which may hinder smooth movement when the detector is moved on the conveyor belt, the power supply cable can be connected to an external power cable using a rotating connector, such as a slip ring, that transmits power from an external rotating body. Wireless communication can also be used for the communication cable.

[0024] For example, when imaging using a detector, the timing for starting imaging is set for each row of pixels arranged in a direction (X direction) parallel to the conveying belt surface and perpendicular to the direction of movement of the detector, and image acquisition begins when a certain row of pixels enters the area irradiated with radiation from the radiation source, and ends when it leaves the irradiation range. Image data during the image acquisition period is temporarily stored in a storage device attached to the detector, and when image acquisition ends, the image data is transmitted to an external receiver. The timing for starting and ending image acquisition can be determined by placing sensors that can detect the passage of the detector upstream and downstream of the irradiation area and detecting the detector.

[0025] <Determination means for determining the presence or absence of defects> The determination means for determining the presence or absence of defects is connected to the detector, and acquires sample movement information from the transport mechanism as well as sample image information from the detector to detect defects present in the sample that is in the irradiation area, and has means (image information acquisition means, image synthesis means) for generating a mapping image in which the positions of defects are identified during the time the sample is present in the irradiation area, and preferably also has means for acquiring information from the thickness measuring device and further obtaining information in the sample thickness direction by calculation, and is composed of defect candidate detection means for calculating defect candidates from the acquired image, and means for determining whether the defect candidates correspond to defects.

[0026] <Image Information Acquisition Means> Signals detected by detector pixels 11 and converted into electrical signals output the intensity of the detected radiation as a brightness value, with pixels that detect strong radiation having a high brightness value (brighter) and pixels that detect weak radiation having a low brightness value (darker). If the size of detector pixel 11 is P [mm] and the desired resolution is Re [mm], then the magnification M must be set so that Re = P / M, as shown in Figure 5. In this case, the scan rate f [Hz], which is the number of times the detector captures images per second, is related to the time τ for one capture by τ = 1 / f [seconds]. If the sample transport speed is Vw [mm / second], the sample is transported at Vw [mm / second] for the length of the desired resolution Re, so the imaging time τ required for one detector capture is τ = Re / Vw = P / (M Vw) [seconds]. The higher the resolution, the smaller the detection pixel size, the larger the magnification, and the faster the sample transport speed, the shorter the time τ required for one image capture, requiring a detector with a high scan rate. When the detector is moved in the same direction as the sample transport direction at a detector speed Vd [mm / sec] by the detector movement mechanism, the detector pixels appear to move slower by Vd / M [mm / sec] than the sample transport speed Vw [mm / sec]. Therefore, the time τ required for one image capture by the detector 10 is τ = Re / |Vw - Vd / M| = P / |M Vw - Vd| [sec]. This increases the time available for one image capture, allowing detectors with low scan rates to achieve high temporal resolution and accommodate fast sample transport speeds. Conversely, increasing the allowable τ for a given resolution allows for faster sample transport speeds. For example, when imaging with a sample transport speed Vw = 100 mm / sec, a target resolution Re = 0.05 mm, a detector pixel size P = 0.1 mm, and a magnification of 2, if the detector transport speed Vd = 0 mm / sec, the imaging time (scan rate 2 kHz) is τ = 0.05 / 100 = 0.5 ms, and if the detector transport speed Vd = 150 mm / sec, imaging is possible in τ = 0.05 / (100 - 150 / 2) = 2 ms (scan rate 0.5 kHz). Furthermore, if the detector transport speed Vd = 200 mm / sec, an infinitely long imaging time is possible.On the other hand, at a speed where Vd / M significantly exceeds Vw, the speed difference becomes large, making imaging itself difficult unless the imaging time is shortened. In order to achieve high imaging accuracy and a higher level of high conveying speed, it is preferable to perform imaging by moving the detector so that the speed ratio Vd / Vw of Vd to Vw is equal to or less than twice the magnification factor M, and it is even more preferable to perform imaging by moving the detector so that Vd / Vw is equal to or less than the magnification factor M. Furthermore, by using a detector movement mechanism in which a flexible detector 14 having bendable flexibility is arranged as shown in Figure 6, and moving the conveying belt 12 on which the flexible detector 14 is arranged, it becomes possible to perform uninterrupted imaging of continuous samples such as roll-to-roll samples.

[0027] <Image Synthesis Method> It is possible to synthesize an image of the sample at a certain time from the intensity signals from the detector pixels 11 arranged on the detector 10 and the position information of the sample from the sample transport mechanism. If a defect exists in the sample, the defect will be recognized as a difference in contrast.

[0028] For example, let us consider a case where defects 15-(1), 15-(2), and 15-(3) of different sizes and shapes exist in a certain region of a sample, as shown in FIG. 7. This figure is a coordinate system in which the bottom surface of the sample is the XY plane with Z = 0, with the center of gravity of defect 15-(1) at coordinates (X1, Y1, Z1), the center of gravity of defect 15-(2) at coordinates (X2, Y2, Z1), and the center of gravity of defect 15-(3) at coordinates (X1, Y1, Z2). Consider images acquired by a detector when such a sample and detector are transported. FIGS. 8 to 10 show an example of continuous imaging performed while the sample and detector are moved from left to right in the figure (i.e., along the Y-axis). Among the continuously captured images, FIG. 8 shows the position of the sample at time ta and the image at that position. FIG. 9 shows the position of the sample at time tb and the image at that position. FIG. 10 shows the position of the sample at time tc and the image at that position. Note that ta<tb<tc. Furthermore, in the two-dimensional projection image 16 obtained at each time ta, tb, and tc, the image corresponding to defect 15-(1) is shown as 17-(1), the image corresponding to defect 15-(2) is shown as 17-(2), and the image corresponding to defect 15-(3) is shown as 17-(3). As can be seen from FIGS. 8 to 10, defects 15-(1) and 15-(2) exist at the same depth (Z1), and therefore, in the projection image 16, both defects appear to move at approximately the same distance. On the other hand, defects 15-(1) and 15-(3) exist at different depths (Z1 and Z2), and therefore, in the projection image 16, their apparent speeds of movement in the Y direction are different. Even if the X and Y coordinates are the same, separate images are obtained over a certain time period. In other words, the amount of movement in the projection image (projection movement amount) varies depending on the defect's depth in the sample. To explain this in more detail, FIG. 11 shows the defect 15-(1) and the corresponding image 17-(1) extracted at times ta, tb, and tc.Let Dab [mm] = Vw(tb-ta) be the movement of defect 15-(1) from time ta to time tb, Dbc [mm] = Vw(tc-tb) be the movement of defect 15-(1) from time tb to time tc, and let Lab be the movement of the image of the defect in the corresponding image from time ta to time tb, and Lbc be the movement of the image of the defect from time tb to time tc. If Mz1 = FDD / (FOD-Z1), which is a magnification factor Mz1 that takes into account the depth of defect 15-(1) within the sample, then Lab = Mz1·Dab = Vw(tb-ta)·FDD / (FOD-Z1), and Lbc = Mz1·Dbc = Vw(tb-ta)·FDD / (FOD-Z1). In other words, it can be seen that the movement of the image of the defect changes depending on the depth of the defect within the sample.

[0029] Another application method involves superimposing an image taken at a certain time on another image taken at a different time, shifted by the amount of defect movement, based on the movement behavior of the images of multiple defects. This allows the images of defects present at a particular depth to overlap, resulting in a high signal intensity and enabling the detection of even low-contrast defects. Specifically, as shown in FIG. 12, for example, by shifting the image taken at time ta (16 in FIG. 8) by −Lab and the image taken at time tc (16 in FIG. 10) by +Lbc using the defect image movement amounts Lab and Lbc calculated at depth Z1, centering the image taken at time tb (16 in FIG. 9), only defects present at depth Z1 can be positioned at approximately the same coordinates. Next, by superimposing the shifted images, defects 15(1) and 15-(2) present at depth Z1 overlap, resulting in a high signal intensity. On the other hand, defect 15-(3) that does not exist at depth Z1 has a different apparent movement speed, resulting in different coordinates, and therefore does not overlap to produce a large signal intensity. Therefore, by calculating the movement amount of the defect relative to Z2 in the same manner as above, shifting the positions by that amount, and adding them together, the position of defect 15-(3) that exists at depth Z2 can now be overlapped, resulting in a large signal intensity. By setting the number of divisions Nz within the sample thickness H and performing the same operation as above, ΔZ = H / Nz at each division, defects present at each depth can be extracted. Furthermore, if multiple defects overlap on the line connecting the X-ray and the detector, they will be detected as a single defect, but it is also possible to separate and extract the defects.

[0030] <Defect Determination> Defect candidates are extracted from the composite image obtained by the image synthesis method, and defects are detected using a brightness threshold in the bright direction and a brightness threshold in the dark direction, which can distinguish between defect candidates and non-defective areas. Detection of defect candidates may be narrowed down by narrowing down the area that satisfies the thresholds based on the size of the detection area, or by narrowing down the area based on the feature quantities of the detected shape. For example, if it is known that a foreign substance that will become a defect has an elongated shape in a specific direction due to a certain process, the orientation (angle) and thinness (aspect ratio) of the detected shape may be narrowed down as feature quantities. Furthermore, a spatial filter or the like may be used prior to detection based on the brightness threshold. Defect determination is performed using these methods.

[0031] The present invention will be explained in more detail below by way of examples, but the present invention is not limited to these examples and should not be interpreted as being limited to these examples. The materials and devices used in the examples are shown below.

[0032] (Inspection Object) A cylinder made of SUS304 with a diameter of 50 μm and a height of 50 μm was attached to a commercially available polyethylene terephthalate film with its bottom (or top) as the adhesive surface using epoxy resin to serve as an object equivalent to a foreign body 18 (manufactured by Toray Precision Co., Ltd.). A top view of the inspection object is shown in Figure 13. Two wires (markers 19) made of SUS304 with a diameter of 200 μm were placed on the film to indicate the location of the object equivalent to a foreign body in the captured image.

[0033] (Apparatus configuration, measurement) A commercially available X-ray tube (Hamamatsu Photonics K.K. L9181-02), a commercially available X-ray TDI camera (Hamamatsu Photonics K.K. C12300-121), a commercially available linear stage 1 for transporting the object to be inspected (SMC LEFB32T-1200-S5C5183), and a commercially available linear stage 2 for transporting the X-ray TDI camera (Zaber L40B1150-N2-KM02 X-MCC2-KX14B) were installed and X-ray imaging was performed. The X-ray tube, X-ray TDI camera, linear stage 1, and linear stage 2 were installed so that the distance FDD from the focus of the X-ray tube to the X-ray TDI camera was 325 mm and the distance FOD from the focus of the X-ray tube to the object to be inspected was 95 mm (see FIG. 5). The conveying directions of the linear stage 1 and the linear stage 2 are the same.

[0034] Next, the inspection object was placed on linear stage 1, and an X-ray TDI camera was placed on linear stage 2. X-rays were irradiated at an X-ray tube voltage of 40 kV and a tube current of 75 μA, and the linear stage 1 and the linear stage 2 were operated at the movement speed Vw and movement speed Vd conditions shown in Table 1, and images were taken with the X-ray TDI camera.

[0035] (Image Evaluation) In the images captured by the above device, the minimum luminance So of the portion where the foreign substance equivalent was captured, the average luminance Sb of the portion where the foreign substance equivalent was not captured, and the standard deviation σb of the luminance of the portion where the foreign substance equivalent was not captured were measured. Sb and σb were measured from 400 arbitrarily selected pixels excluding the location where the foreign substance equivalent was located in the image and its vicinity (a 20-pixel square (approximately 280 μm × 280 μm in this example) surrounding the location). The luminance was measured using the signal amount of the output X-ray image captured with an X-ray TDI camera. Table 1 shows the CNR (contrast-to-noise ratio), which is the value obtained by dividing the contrast C, which is the absolute value of the difference between So and Sb, by σb. The higher the CNR value, the greater the signal intensity contrast of the object being inspected compared to the noise in the portion where the object was not present, making it easier to detect.

[0036]

[0037] According to the present invention, transmission images are acquired while moving the detector in the same direction as the movement of the sample, so that projection images with high time resolution can be obtained even with a detector with a low scan rate.In addition, by acquiring images taken at various irradiation angles from a point light source at multiple locations, even if multiple defects overlap on the line connecting the radiation source and the detector, they can be separated and extracted.Furthermore, even low-contrast defects can be detected from multiple images as a large signal, allowing for high-speed and high-precision defect inspection.

[0038] 1: Housing 2, 2(a), 2(b): Sample 3: Inlet 4: Outlet 5: Sample transport roll 6: Sample transport belt 7: Thickness measurement sensor 8: Radiation source 9: Radiation 10: Detector 11: Detector pixel 12: Detector transport belt 13: Driving means for movement mechanism 14: Flexible detector 15-(1), 15-(2), 15-(3): Defects present in the sample 16: Projected image 17-(1), 17-(2), 17-(3): Images of defects present in the sample 18: Foreign matter equivalent (SUS304 cylinder) 19: Marker (SUS304 wire)

Claims

1. A inspection apparatus comprising at least: a transport mechanism for transporting a sample to be inspected; a radiation source that irradiates the area through which the transported sample passes with radiation in a radial direction; a detector that is disposed so as to be able to detect the radiation that has passed through the transported sample and converts the detected radiation into an electrical signal; a moving mechanism that moves the detector along the direction in which the sample is transported by the transport mechanism; and a determination means for determining the presence or absence of a defect in the sample from the image information of the sample from the detector and the movement information of the sample from the transport mechanism obtained during the time the sample passes through the area. The transport mechanism and the moving mechanism move the sample and the detector such that when the moving speed of the sample by the transport mechanism is Vw and the moving speed component of the detector by the moving mechanism in the same direction as the moving direction of the sample is Vd, the ratio Vd / Vw of Vd to Vw is not more than twice the ratio (FDD / FOD) of the distance (FDD) between the radiation source and the detector to the distance (FOD) between the radiation source and the sample.

2. Furthermore, the inspection apparatus according to claim 1, further comprising a thickness measuring instrument for measuring the thickness of the sample, and a calculating means for calculating the position of a defect in the sample based on the thickness information of the sample measured by the thickness measuring instrument and the image information of the sample from the detector.

3. The inspection apparatus according to claim 1 or 2, wherein a plurality of the detectors are arranged along the transport direction of the sample.

4. The moving mechanism is a moving mechanism having an endless annular transport belt, and the moving mechanism has a driving means for applying a driving force to the transport belt, an assisting means for assisting the smooth movement of the transport belt, and a buffering means for absorbing the tensile force and compressive force generated in the transport body as the transport belt moves.

5. The inspection apparatus according to claim 1 or 2, wherein the detector is a flexible detector having flexibility that can be curved, and the detector is arranged along the moving direction of the moving mechanism.

6. The inspection apparatus according to claim 1 or 2, wherein the radiation is an electromagnetic wave.

7. The inspection apparatus according to claim 1 or 2, wherein the detector is an indirect conversion type detector including a scintillator panel having a pixel structure filled with a phosphor that emits light by radiation in a space partitioned by a lattice-shaped partition formed on a substrate, and a photoelectric conversion element having a cell structure corresponding to the partition and performing photoelectric conversion on the light emitted from the phosphor.

8. A method for inspecting the presence or absence of defects inherent in a sample, comprising: passing a sample to be inspected through a region irradiated with radiation from a radiation source that irradiates radiation radially by a transport mechanism; detecting radiation transmitted through the sample passing through the region by a detector that is provided in a moving mechanism, is disposed so as to be capable of detecting the radiation transmitted through the sample passing through the region, and converts the detected radiation into an electrical signal, while moving the detector along the direction in which the sample is transported by the transport mechanism to perform detection of the radiation and conversion into an electrical signal; determining the presence or absence of defects in the sample from the image information of the sample from the detector obtained during the time when the sample passes through the region and the movement information of the sample from the transport mechanism. An inspection method, comprising: moving the sample and the detector by the transport mechanism and the moving mechanism such that, when the moving speed of the sample by the transport mechanism is Vw and the moving speed component of the detector by the moving mechanism in the same direction as the moving direction of the sample is Vd, the ratio Vd / Vw of Vd to Vw is not more than twice the ratio (FDD / FOB) of the distance (FDD) between the radiation source and the detector to the distance (FOB) between the radiation source and the sample.

9. The inspection method according to claim 8, further comprising measuring the thickness of the sample and calculating the position of the defect in the sample based on the measured thickness information and the image information of the sample from the detector.

10. The inspection method according to claim 8 or 9, wherein a plurality of the detectors are arranged along the transport direction of the sample.

11. The inspection method according to claim 8 or 9, wherein the moving mechanism is a moving mechanism including an endless annular transport belt, and the moving mechanism includes drive means for applying a driving force to the transport belt, auxiliary means for assisting the smooth movement of the transport belt, and buffer means for absorbing the tensile force and compressive force generated in the transport belt due to the movement of the transport belt.

12. The inspection method according to claim 8 or 9, wherein the radiation is an electromagnetic wave.

13. The inspection method according to claim 8 or 9, wherein the detector is an indirect conversion type radiation detector including a scintillator panel having a pixel structure filled with a phosphor that emits light by radiation in a space partitioned by lattice-shaped partition walls formed on a substrate, and a photoelectric conversion element having a cell structure corresponding to the partition and performing photoelectric conversion of the light emitted from the phosphor.