Measuring device and sorting machine

Abstract

A measurement device for measuring the shape and / or dimension characteristics of the entire and / or part of an object is provided with: a transfer unit configured to transfer the object; an electromagnetic wave irradiation source configured to irradiate an electromagnetic wave to the object being transferred by the action of the transfer unit; a line sensor configured so as to have a plurality of electromagnetic wave detection elements linearly arranged in a first direction intersecting a transfer direction of the object, the line sensor detecting at least one of a reflected electromagnetic wave irradiated from the electromagnetic wave irradiation source and reflected by the object and a transmitted electromagnetic wave transmitted through the object; a transfer speed detection unit configured to detect a transfer speed of the object being transferred in a predetermined direction; and a feature determination unit configured to determine a feature of the object on the basis of the image acquired by the line sensor and the transfer speed in the predetermined direction.

Description

Technical Field

[0001] The present disclosure relates to a technique for measuring the shape and / or size characteristics of an entire object and / or a part thereof. Background Art

[0002] Conventionally, an optical sorter (hereinafter also simply referred to as a sorter) is known as follows: using light information obtained by an optical sensor when light is irradiated from a light source onto a sorting object (hereinafter also simply referred to as an object) being transferred, to determine the characteristics of the object (for example, whether it is a qualified product or a non-qualified product), and removing specific objects (for example, non-qualified products). For example, Japanese Patent Application Laid-Open No. 2000-197855 discloses a sorter that uses a line sensor in which a plurality of light receiving elements are linearly arranged in a direction crossing the transfer direction of the object as an optical sensor. With the line sensor, the illuminance of the light from the light source on each light receiving element is kept uniform, so that defects on the appearance of the sorting object can be detected with good accuracy.

[0003] However, for the purpose of detecting the shape and / or size characteristics, there is still room for improvement in the detection accuracy of the characteristics of a sorter equipped with a line sensor. Specifically, the line sensor synthesizes a plurality of linear images obtained by scanning a single object being transferred multiple times, thereby obtaining an image of the single object, and thus is affected by the transfer speed and / or transfer direction of the object.

[0004] For example, in a line sensor, it has the following characteristics: the resolution in a direction orthogonal to the direction (hereinafter also referred to as the arrangement direction) in which the plurality of light receiving elements are arranged changes according to the transfer speed of the object in that direction. Therefore, under processing conditions where it is impossible to keep the transfer speed of the object constantly consistent, the above characteristics will reduce the detection accuracy of the size and shape of the object in the obtained image.

[0005] Moreover, in a sorter, for example, a phenomenon may occur in which the transfer direction of the object inadvertently includes a component in the arrangement direction due to collision between the objects being transferred. In the case where such a phenomenon occurs, if the plurality of linear images obtained by the line sensor are simply synthesized, a shape that is deformed in the arrangement direction with respect to the actual shape of the object will be detected. This will reduce the detection accuracy of the shape of the object in the obtained image.

[0006] The above problems are not limited to sorters that irradiate light onto an object, and are common to various sorters equipped with an electromagnetic wave irradiation source and a line sensor for detecting electromagnetic waves. In addition, this problem is not limited to sorters, but is also common to measuring devices for measuring the shape and / or size characteristics of an object. Therefore, for a measuring device or sorter equipped with a line sensor, it is required to improve the detection accuracy of the shape and / or size characteristics of an object. Summary of the Invention

[0007] The present disclosure is achieved to solve at least a part of the above problems, and can be realized, for example, in the following manner.

[0008] According to a first aspect of the present disclosure, there is provided a measuring device for measuring the shape and / or size characteristics of the whole and / or a part of an object. The measuring device includes: a transfer unit configured to transfer the object; an electromagnetic wave irradiation source configured to irradiate electromagnetic waves onto the object being transferred by the action of the transfer unit; a line sensor configured to have a plurality of electromagnetic wave detection elements linearly arranged in a first direction intersecting the transfer direction of the object, and detect at least one of the reflected electromagnetic waves irradiated from the electromagnetic wave irradiation source and reflected by the object and the transmitted electromagnetic waves transmitted through the object; a transfer speed detection unit configured to detect the transfer speed of the object being transferred in a specified direction; and a feature determination unit configured to determine the characteristics of the object based on the image obtained by the line sensor and the transfer speed in the specified direction.

[0009] "The object being transferred by the action of the transfer unit" includes, for example, an object being transferred on the transfer unit and an object falling from the transfer unit. In addition, the electromagnetic wave irradiation source may irradiate at least one of visible light, near-infrared light, and X-rays.

[0010] According to this measuring device, the feature determination unit determines the characteristics of the object based on the image obtained by the line sensor and the transfer speed of the object being transferred in the specified direction. Therefore, it is possible to accurately determine the shape and / or size characteristics of the object by reflecting the difference in the transfer speed of the object in the specified direction (in other words, reducing or eliminating the influence of the difference in the transfer speed in the specified direction on the image).

[0011] According to a second aspect of the present disclosure, based on the first aspect, the feature includes a first feature quantity of the whole and / or a part of the object. The feature determination unit is further configured to correct the image acquired by the line sensor based on the transfer speed in a specified direction, and determine the first feature quantity based on the corrected image, or configured to correct the second feature quantity determined based on the image acquired by the line sensor based on the transfer speed in the specified direction, thereby determining the first feature quantity. According to this aspect, the image or the second feature quantity is corrected based on the transfer speed in the specified direction (in other words, correction is performed to reduce or eliminate the influence of the difference in the transfer speed in the specified direction on the image), so that the first feature quantity of the whole and / or a part of the object can be accurately determined. The first feature quantity can be, for example, at least one of the area, height, width, outer perimeter, and roundness of the whole and / or a part of the object. This is the same for the third aspect described later.

[0012] According to a third aspect of the present disclosure, based on the first aspect or the second aspect, the feature includes the mass of the whole and / or a part of the object. The feature determination unit is configured to determine the mass based on the first feature quantity of the whole and / or a part of the object. The feature determination unit is further configured to correct the image acquired by the line sensor based on the transfer speed in the specified direction, and acquire the first feature quantity based on the corrected image, or configured to correct the second feature quantity determined based on the image acquired by the line sensor based on the transfer speed in the specified direction, thereby acquiring the first feature quantity. According to this aspect, the image or the second feature quantity is corrected based on the transfer speed in the specified direction (in other words, correction is performed to reduce or eliminate the influence of the difference in the transfer speed in the specified direction on the image), so that the mass of the whole and / or a part of the object can be accurately determined. The mass can be, for example, the determination result of the quality based on a specified standard (the determination result of whether it is a qualified product or a non-qualified product), or the grade of the quality. In addition, the mass can also include the category of non-conformity.

[0013] According to a fourth aspect of the present disclosure, based on the first aspect or the second aspect, the specified direction includes a second direction orthogonal to the first direction. The feature includes the mass of the whole and / or a part of the object. The feature determination unit is further configured to compare the feature quantity determined based on the image acquired by the line sensor and the threshold determined based on the transfer speed in the second direction, thereby determining the mass. According to this aspect, the threshold is determined based on the transfer speed in the second direction (in other words, according to the change in the image size in the second direction caused by the change in the transfer speed in the second direction, the threshold is corrected in a manner that increases or decreases in the same direction as the change in the image size), so that the mass of the whole and / or a part of the object can be accurately determined.

[0014] According to the fifth aspect of the present disclosure, based on the second or third aspect, the specified direction includes a second direction orthogonal to the first direction. The feature determination unit is further configured to correct the size of the image in the second direction based on the transfer speed in the second direction, thereby correcting the image, or correct the feature quantity component in the second direction in the second feature quantity based on the transfer speed in the second direction, thereby correcting the second feature quantity. According to this aspect, even if the resolution in the second direction of the line sensor changes due to the difference in the transfer speed in the second direction of the object, the image or the second feature quantity can be corrected to reduce or eliminate the influence of the change in the resolution. Therefore, based on the corrected image or the corrected second feature quantity, the features of the shape and / or size of the object can be accurately determined.

[0015] According to the sixth aspect of the present disclosure, based on any one of the first to fifth aspects, the specified direction includes the first direction. The feature determination unit is further configured to correct the coordinate values in the first direction of the plurality of pixels constituting the image based on the transfer speed in the first direction, thereby correcting the image, and determine the features of the object based on the corrected image. According to this aspect, when the transfer speed of the object has a speed component in the first direction, that is, when the object is transferred while laterally moving relative to the intended transfer direction, the image can be corrected to reduce or eliminate the deformation of the object on the image due to the lateral movement. Therefore, based on the corrected image, the features of the shape and / or size of the object can be accurately determined.

[0016] According to the seventh aspect of the present disclosure, a sorting machine is provided. The sorting machine includes: a measuring device according to any one of the first to sixth aspects; and a sorting unit configured to sort the object based on the features determined by the feature determination unit. According to this sorting machine, the same effects as any one of the first to sixth aspects can be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram showing a schematic structure of an optical sorting machine according to an embodiment of the present disclosure.

[0018] Figure 2 It is a schematic diagram showing the arrangement of optical elements in a color sensor.

[0019] Figure 3 It is an explanatory diagram exemplifying a region captured by one scan in a sorting object.

[0020] Figure 4 It is an explanatory diagram showing an example of a method for calculating a color deviation amount.

[0021] Figure 5It is an explanatory diagram showing an example of a method for correcting an image representing the transfer speed in a second direction of an object.

[0022] Figure 6 It is an explanatory diagram showing an example of a method for correcting an image representing the transfer speed in a first direction of an object.

[0023] Figure 7 It is an explanatory diagram showing an example of a method for correcting an image representing the transfer speed in a first direction of an object. Detailed implementation mode

[0024] Figure 1 It is a schematic diagram showing the schematic structure of an optical sorter (hereinafter simply referred to as a sorter) 10 as an embodiment of the present disclosure. In the present embodiment, the sorter 10 is used to sort defective products (such as broken grains, undercooked grains, grains of different colors, damaged grains, dead rice, foreign objects (such as small stones, mud, glass pieces, etc.)) from rice grains (more specifically, brown rice or polished rice) as an example of an object to be sorted (hereinafter simply referred to as an object) 90. However, the object 90 is not limited to brown rice or polished rice, and can also be any granular material. For example, the object 90 can also be paddy rice, wheat grains, beans (soybeans, chickpeas, edamame, etc.), resin (granules, etc.), rubber sheets, etc.

[0025] As Figure 1 shown, the sorter 10 includes: an optical detection unit 20, a storage tank 71, a feeder 72, a chute 73, a qualified product discharge chute 74, a defective product discharge chute 75, a sorting unit 60, and a controller 80. The controller 80 controls all the operations of the sorter 10. The controller 80 also functions as a transfer speed calculation unit 81 and a feature determination unit 82. The functions of the controller 80 can be realized by a CPU executing a prescribed program, can also be realized by a dedicated circuit (such as a PLD, ASIC, etc.), or can be realized by a combination of a CPU and a dedicated circuit. In addition, the functions of the controller 80 can be assigned to an integrated single device, or can be distributed to multiple devices in a dispersed manner. The detailed content of the functions of the controller 80 will be described later.

[0026] The storage tank 71 temporarily stores the object 90. The feeder 72 supplies the object 90 stored in the storage tank 71 onto the chute 73 as an example of a transfer unit for transferring the object. The object 90 supplied onto the chute 73 slides downward on the chute 73 and falls from the lower end of the chute 73. The chute 73 has a prescribed width capable of allowing a plurality of objects 90 to fall simultaneously. As the transfer unit, a conveyor can also be used instead of the chute 73.

[0027] The optical detection unit 20 irradiates light on the object 90 that has slid down the chute 73 (i.e., the object 90 that is falling from the chute 73), and detects the light associated with the object 90 (specifically, the transmitted light that passes through the object 90 and / or the reflected light that is reflected by the object 90). In addition, in an alternative embodiment, light may also be irradiated on the object 90 that is sliding on the chute 73. Further, in the case of using a conveyor instead of the chute 73, light may also be irradiated on the object 90 that is being transported on the conveyor or the object 90 that is falling from the conveyor.

[0028] As Figure 1 shown, the optical detection unit 20 includes a first light source 31, a second light source 32, a first line sensor 40, and a second line sensor 50. The first light source 31 and the first line sensor 40 are arranged on one side (also referred to as the front side) with respect to the transfer path of the object 90 (in other words, the trajectory of the object 90 falling from the chute 73). On the other hand, the second light source 32 and the second line sensor 50 are arranged on the other side (also referred to as the rear side) with respect to the transfer path of the object 90.

[0029] The first light source 31 emits first light 33 toward a plurality of objects 90 being transferred (i.e., objects 90 falling from the chute 73). Similarly, the second light source 32 emits second light 34 toward a plurality of objects 90 being transferred. The first light 33 and the second light 34 each have a wavelength corresponding to red, a wavelength corresponding to green, and a wavelength corresponding to blue. In the present embodiment, the first light source 31 and the second light source 32 are so-called color LEDs. However, the light sources 31, 32 may also be any other light-emitting elements (such as halogen lamps). In addition, Figure 1 it, the number of the light sources 31, 32 is shown as one each, but the number of the light sources of at least one of the light sources 31, 32 may also be plural.

[0030] The first line sensor 40 and the second line sensor 50 detect the light associated with the object 90 being transferred. The front-side first line sensor 40 can detect the light 33 (hereinafter also referred to as the reflected light 33) emitted from the front-side first light source 31 and reflected by the object 90, and the light 34 (hereinafter also referred to as the transmitted light 34) emitted from the rear-side second light source 32 and passing through the object 90. The rear-side second line sensor 50 can detect the light 34 (hereinafter also referred to as the reflected light 34) emitted from the rear-side second light source 32 and reflected by the object 90, and the light 33 (hereinafter also referred to as the transmitted light 33) emitted from the front-side light source 31 and passing through the object 90.

[0031] It is detected by the line sensors 40 and 50 what kind of light is determined by the lighting patterns of the light sources 31 and 32. In the first lighting pattern in which the first light source 31 and the second light source 32 are lit simultaneously, the first line sensor 40 detects the light synthesized by the reflected light 33 and the transmitted light 34 (hereinafter also referred to as the reflected and transmitted light), and the second line sensor 50 detects the reflected and transmitted light synthesized by the reflected light 34 and the transmitted light 33. In the second lighting pattern in which the first light source 31 is lit and the second light source 32 is extinguished, the first line sensor 40 detects the reflected light 33, and the second line sensor 50 detects the transmitted light 33. In the third lighting pattern in which the first light source 31 is extinguished and the second light source 32 is lit, the first line sensor 40 detects the transmitted light 34, and the second line sensor 50 detects the reflected light 34. Which of the first to third lighting patterns to adopt can be arbitrarily determined according to the type and properties of the object 90 and the type of defective products to be removed. It is also possible to adopt only any one of the first to third lighting patterns. Or, two or more of the first to third lighting patterns may appear alternately at a predetermined time interval or according to a predetermined repetition rule.

[0032] In the present embodiment, the line sensors 40 and 50 are color CCD sensors. More specifically, each of the line sensors 40 and 50 has: a plurality of optical elements for detecting light having a wavelength corresponding to red (hereinafter referred to as R elements), a plurality of optical elements for detecting light having a wavelength corresponding to green (hereinafter referred to as G elements), and a plurality of optical elements for detecting light having a wavelength corresponding to blue (hereinafter referred to as B elements). R, G, and B respectively mean R, G, and B in the RGB color space. Each of these optical elements includes a condenser lens, a color filter, and a photoelectric conversion element. Each of the color filters has the property of transmitting light having a wavelength corresponding to the color of the light to be detected (for example, if it is an R element, it is red) and not transmitting light of other wavelengths. The line sensors 40 and 50 are not limited to CCD sensors, and may also be other types of line sensors such as CMOS sensors.

[0033] Figure 2It is a schematic diagram showing the configuration of the optical elements in the first-line sensor 40. As shown in the figure, the first-line sensor 40 is a so-called 3-line sensor, which includes: an R element group 44 in which a plurality of R elements 41 are arranged in a row, a G element group 45 in which a plurality of G elements 42 are arranged in a row, and a B element group 46 in which a plurality of B elements 43 are arranged in a row. The plurality of R elements 41, the plurality of G elements 42, and the plurality of B elements 43 are all linearly arranged in a first direction D1 (which is also the width direction of the chute 73) that intersects the transfer direction of the object 90 (the falling direction of the object 90). In other words, the R element group 44, the G element group 45, and the B element group 46 are arranged in parallel so as to be separated from each other in a second direction D2 that is orthogonal to the first direction D1. The second direction D2 is also the direction intended as the transfer direction of the object 90. However, in reality, there are cases where the object 90 has a velocity component in the first direction D1 due to collisions between the objects 90, etc. In that case, the transfer direction of the object 90 becomes a direction that intersects the second direction D2.

[0034] The separation distance between the R element group 44 and the G element group 45 is L1, the separation distance between the G element group 45 and the B element group 46 is L2, and the separation distance between the R element group 44 and the B element group 46 is L3 (= L1 + L2). Usually, L1 = L2, but L1 and L2 can also be different values. The second-line sensor 50 has the same structure as the first-line sensor 40, so its description is omitted.

[0035] The outputs from such line sensors 40 and 50, that is, the analog signals representing the detected light intensity, are converted into digital signals by an AC / DC converter (not shown). This digital signal is input to the controller 80 as image data. As is well known, in a line sensor, a linear image can be obtained through one scan. Therefore, by synthesizing a plurality of linear images obtained through multiple scans, an image having a specified height in the direction corresponding to the second direction D2 (an image having a size that can accommodate at least one object 90, typically an image having a size that can accommodate a plurality of objects 90) can be obtained. For example, as Figure 3 shown, through 10 scans (illustrated as a smaller number than the actual number for simplicity of explanation), an overall image of one object 90 is obtained for each of the colors R, G, and B. Figure 3 The numbers 1 to 10 shown indicate that the linear region marked with this number is the region photographed through the corresponding scan. For example, the region marked with "2" is the region for which image data is obtained through the second scan.

[0036] Based on the images obtained by the line sensors 40 and 50 like this, the controller 80 determines the characteristics of the object 90 for each object 90, as the processing of the feature determination unit 82. Such processing is performed for both the image obtained by the first line sensor 40 and the image obtained by the second line sensor 50. In the present embodiment, the characteristics include color characteristics (in other words, optical characteristics), and shape and / or size characteristics. In addition, the characteristics include characteristic quantities representing the characteristics by physical quantities, and quality determined based on the characteristic quantities.

[0037] The characteristic quantity of color includes the color gray-scale values of the respective pixels of the image of the object 90. The characteristic quantity of shape and / or size can include, for example, at least one of the area, height, width, outer perimeter, and roundness of the whole and / or a part of the object 90. Such a characteristic quantity can also be calculated as a value in units of the number of pixels constituting the image. Alternatively, the characteristic quantity can be calculated as a value in actual size units based on the value in units of the number of pixels and the resolutions of the line sensors 40 and 50.

[0038] In the present embodiment, "quality" includes, for example, the distinction between acceptable products (that is, rice grains with relatively high quality) and non-acceptable products (that is, rice grains with relatively low quality and / or foreign substances). Among them, "quality" can also include the category of non-conformance (for example, which one of broken rice, undercooked grains, grains of different colors, damaged grains, dead rice, and foreign substances it corresponds to). Alternatively, "quality" can also include the distinction between rice grains to be removed in the sorting unit 60 and rice grains not to be removed. In addition, "quality" includes the quality determined based on the color characteristics. Non-acceptable products determined based on the color characteristics can include, for example, undercooked grains, grains of different colors, damaged grains, dead rice, and foreign substances. Non-acceptable products determined based on the shape and / or size characteristics can include, for example, broken rice, pest-damaged grains, and foreign substances.

[0039] In the present embodiment, the feature determination unit 82 compares the characteristic quantity of color (in other words, the gray-scale value of the image data) with a pre-determined threshold value (in other words, whether the color-based characteristic quantity is within the pre-determined normal range), thereby determining whether the object 90 is an acceptable product or a non-acceptable product. Such a determination can also be made based on the representative value (average value, median value, maximum value, minimum value, etc.) of the gray-scale values of the multiple pixels constituting the image of the object 90. Alternatively, non-acceptable products can also include the object 90 with local non-conformance of a specified size or more. Such local non-conformance can also be determined based on the number of pixels with gray-scale values not within the normal range among the multiple pixels constituting the image of the object 90 being equal to or more than a specified number (in other words, the area of the non-conforming part being equal to or more than a specified value).

[0040] Further, in the present embodiment, the feature determination unit 82 compares the feature amounts of the shape and / or size with a predetermined threshold value (in other words, determines whether the feature amounts of the shape and / or size are within a predetermined normal range), thereby determining whether the object 90 is a qualified product or a non-qualified product.

[0041] The sorting unit 60 sorts the object 90 based on the features determined by the feature determination unit 82. This sorting is performed by a track change operation for changing the track of a specific object 90. Specifically, the sorting unit 60 includes: a plurality of nozzles 61, and valves 62 corresponding to the number of the nozzles 61 (in the present embodiment, the same as the number of the nozzles 61, but may be different from the number of the nozzles 61). The plurality of nozzles 61 are arranged in the width direction of the chute 73.

[0042] The plurality of nozzles 61 are respectively connected to a compressor (not shown) via the plurality of valves 62. The plurality of valves 62 are selectively opened according to a control signal from the controller 80, so that the plurality of nozzles 61 selectively jet air 63 toward the object 90 determined to be a non-qualified product (more precisely, the object 90 determined to be a non-qualified product based on the feature amount of color with respect to the image obtained by at least one of the line sensors 40 and 50, and the object 90 determined to be a non-qualified product based on the feature amount of shape and / or size). The object 90 determined to be a non-qualified product is blown away by the air 63, deviates from the track of falling from the chute 73 and is guided to the non-qualified product discharge chute 75 (represented as the object 91 in Figure 1 ). On the other hand, air 63 is not jetted toward the object 90 determined to be a qualified product. Therefore, the object 90 determined to be a qualified product does not change the falling track and is guided to the qualified product discharge chute 74 (represented as the object 92 in Figure 1 ).

[0043] In an alternative embodiment, instead of the structure of jetting air 63 toward the object 90 after falling from the chute 73, air 63 may be jetted toward the object 90 sliding on the chute 73 to change the transfer path of the object 90. In addition, as the transfer unit, a belt conveyor may be used instead of the chute 73. In this case, air may also be jetted toward the object falling from one end of the belt conveyor. Or, air may also be jetted toward the object being conveyed on the belt conveyor.

[0044] In a further alternative embodiment, instead of the structure of jetting air 63 toward the object 90 determined to be a non-qualified product, air 63 may be jetted toward the object 90 determined to be a qualified product (so-called reverse removal). In addition, the track change operation is not limited to the jetting of air 63, and any other known method may be adopted.

[0045] In the above-described sorter 10, in order to improve the accuracy related to the determination of the shape and / or size characteristics of the object 90, the transfer speed of the object 90 in a specified direction during transfer is detected, and based on this transfer speed, the shape and / or size characteristics are determined. Hereinafter, such a structure will be described in detail. In the present embodiment, the specified direction includes both the first direction D1 and the second direction D2. However, the specified direction may also be only one of the first direction D1 and the second direction D2.

[0046] First, with reference to Figure 4 and Figure 5 , a method for detecting the transfer speed of the object 90 in the specified direction will be described. Hereinafter, a method for detecting the transfer speed of the object 90 at the moment when the object 90 is photographed by the first line sensor 40 using the first line sensor 40 will be described in detail. The detection of such a transfer speed is executed as a process of the transfer speed calculation unit 81 of the controller 80.

[0047] In order to detect the transfer speed, the transfer speed calculation unit 81 first calculates the amount of color deviation for the color image obtained by the first line sensor 40. In the first line sensor 40, the R element group 44, the G element group 45, and the B element group 46 perform scanning simultaneously, but the R element group 44, the G element group 45, and the B element group 46 are separated from each other in the second direction D2 (refer to Figure 2 ), so strictly speaking, the photographed part of the object 90 is offset by the amount of this separation distance between the colors. Therefore, a color deviation occurs in the direction corresponding to the second direction D2 between the red image obtained by the R element group 44, the green image obtained by the G element group 45, and the blue image obtained by the B element group 46. And when the transfer direction of the object 90 includes a component in the first direction D1 due to the collision of the objects 90 with each other, etc. (in other words, when the transfer direction of the object 90 is a direction intersecting the second direction D2 (excluding the perpendicular direction)), a color deviation also occurs in the direction corresponding to the first direction D1 between the red image, the green image, and the blue image. The amount of such a color deviation is calculated in units smaller than 1 pixel, which is the unit constituting the image.

[0048] In the present embodiment, the transfer speed calculation unit 81 calculates the color deviation amounts S1rg, S2rg between the red image and the green image, the color deviation amounts S1gb, S2gb between the green image and the blue image, and the color deviation amounts S1rb, S2rb between the red image and the blue image. Figure 4 is an explanatory diagram showing an example of the calculation method of the color deviation amount. Hereinafter, with reference to Figure 4, the color deviation amounts S1rg and S2rg between the red image 92R and the green image 92G are described in terms of the particles of the object 90. In the calculation of the color deviation amount, the transfer speed calculation unit 81 first sets common coordinates for the red image 92R of 1 particle of the object 90 and the green image 92G of the same 1 particle of the object 90 at a resolution finer than 1 pixel. For example, coordinate points in units of 1 / 1000 pixels can be set for each of the X direction (the direction corresponding to the first direction D1) and the Y direction (the direction corresponding to the second direction D2). In this case, 1 million (= 1000 × 1000) coordinate points are assigned to 1 pixel. This resolution can be set to any value according to the desired calculation accuracy of the color deviation amount.

[0049] Next, the transfer speed calculation unit 81 calculates the red concentration center-of-gravity coordinate point 93R of the red image 92R and the green concentration center-of-gravity coordinate point 93G of the green image 92G based on the color gray-scale values of each coordinate point (the gray-scale values of 1 million coordinate points corresponding to 1 pixel are the same). For the coordinate values of the concentration center-of-gravity coordinate points, for each of the X coordinate and the Y coordinate, it can be calculated by dividing the sum of the values obtained by multiplying the coordinate values and the gray-scale values for each coordinate point by the sum of the gray-scale values of each coordinate point.

[0050] Then, the transfer speed calculation unit 81 calculates the color deviation amount based on the obtained concentration center-of-gravity coordinate points. For example, as Figure 4 shown, the transfer speed calculation unit 81 obtains the separation distance between the red concentration center-of-gravity coordinate point 93R and the green concentration center-of-gravity coordinate point 93G in the Y direction (that is, the direction corresponding to the second direction D2) as the color deviation amount S2rg (unit: pixel) in the Y direction between the red image 92R and the green image 92G. In addition, the transfer speed calculation unit 81 obtains the separation distance between the red concentration center-of-gravity coordinate point 93R and the green concentration center-of-gravity coordinate point 93G in the X direction (that is, the direction corresponding to the first direction D1) as the color deviation amount S1rg in the X direction between the red image 92R and the green image 92G. Although not shown in the figure, the transfer speed calculation unit 81 similarly calculates the color deviation amounts S2gb and S1gb in the Y direction and the X direction between the green image 92G and the blue image 92B, and the color deviation amounts S2rb and S1rb in the Y direction and the X direction between the red image 92R and the blue image 92B.

[0051] The calculation of the color deviation amount is not limited to the above method using the concentration center of gravity, and can be performed by any known method. In addition, instead of or in addition to the structure that calculates the color deviation amount for each particle of the object 90, the color deviation amount can be calculated for each particle group overlapping on the color image. In order to calculate the amount of color deviation for each particle and / or each particle group of the object 90, it is also possible to remove the image area outside the image area representing the particle or particle group of the object 90 (hereinafter also referred to as the blank area) from the calculation object area of the color deviation amount. The blank area can be easily removed by binarizing the color image. Thus, if the color deviation amount is calculated in units of particles or particle groups, the color deviation amount can be calculated with good accuracy. However, the transfer speed calculation unit 81 may also divide the color image into a plurality of regions (the size of the region can include a plurality of particles), and calculate the color deviation amount in each of the plurality of regions.

[0052] Next, the transfer speed calculation unit 81 calculates the transfer speed of the object 90 in the second direction D2 for each particle of the object 90 based on the color deviation amounts S2rg, S2gb, and S2rb in the Y direction (i.e., the direction corresponding to the second direction D2). If the time required for the first line sensor 40 to scan once is set as the scan time T, the transfer speed V2rg of the object 90 in the second direction D2 calculated based on the color deviation amount S2rg between the red image 92R and the green image 92G is obtained, for example, by the following formula (1). In formula (1), L1 > 0. In addition, in formula (1), the unit of the color deviation amount S2rg is a unit representing distance. For example, when the unit of the separation distance L1 and the color deviation amount S2rg is "mm", and the unit of the scan time T is "ms", the unit of the obtained transfer speed V2rg is "m / s". As described above, when the transfer speed calculation unit 81 obtains the color deviation amount S2rg in units of "pixels", the color deviation amount S2rg in units of "pixels" is multiplied by the size (mm) of the pixels in the second direction D2 of the first line sensor 40 (i.e., the size corresponding to 1 pixel) to obtain the color deviation amount S2rg in units of "mm". The transfer speed V2rg is not limited to formula (1), and can also be calculated by other formulas including the color deviation amount S2rg as a variable based on experiments or the like.

[0053] V2rg = (L1 + S2rg) / T ··· (1)

[0054] Similarly, the transfer speed V2gb of the object 90 in the second direction D2 calculated based on the color deviation amount S2gb is obtained by the following formula (2), and the transfer speed V2rb of the object 90 in the second direction D2 calculated based on the color deviation amount S2rb is obtained by the following formula (3). In formulas (2) and (3), L2 > 0, L3 > 0.

[0055] V2gb = (L2 + S2gb) / T ··· (2)

[0056] V2rb = (L3 + S2rb) / T ··· (3)

[0057] Further, the transfer speed calculation unit 81 calculates the transfer speeds V1rg, V1gb, and V1rb of the object 90 in the first direction D1 based on the color deviation amounts S1rg, S1gb, and S1rb in the X direction (i.e., the direction corresponding to the first direction D1), respectively. The transfer speeds V1rg, V1gb, and V1rb are obtained, for example, by the following formulas (4) to (6).

[0058] V1rg = S1rg / T ··· (4)

[0059] V1gb = S1gb / T ··· (5)

[0060] V1rb = S1rb / T ··· (6)

[0061] The transfer speed calculation unit 81 detects the representative value (such as their average value, median value, etc.) of the transfer speeds V2rg, V2gb, and V2rb obtained as described above as the transfer speed V2 of the object 90 in the second direction D2. Similarly, the transfer speed calculation unit 81 detects the representative value (such as their average value, median value, etc.) of the transfer speeds V1rg, V1gb, and V1rb as the transfer speed V1 of the object 90 in the first direction D1. In an alternative embodiment, the transfer speed calculation unit 81 may detect the transfer speed V2 based on one or two of the transfer speeds V2rg, V2gb, and V2rb, or may detect the transfer speed V1 based on one or two of the transfer speeds V1rg, V1gb, and V1rb.

[0062] Next, with reference to Figures 5 - 7 , a method for determining the shape and / or size characteristics of the object 90 based on the transfer speed will be described. Hereinafter, a method for determining the shape and / or size characteristics of the image obtained by the first line sensor 40 based on the transfer speed calculated from the image obtained by the first line sensor 40 will be described in detail. In the present embodiment, the feature determination unit 82 corrects the image of the object 90 based on the transfer speed V1 in the first direction D1 and the transfer speed V2 in the second direction D2, and determines the shape and / or size characteristics of the object 90 based on the corrected image. However, only one of the correction based on the transfer speed V1 and the correction based on the transfer speed V2 may be performed. The correction based on the transfer speed V1 and / or the transfer speed V2 is performed for each grain of the object 90.

[0063] First, the correction of the image based on the transfer speed V2 in the second direction D2 will be described. The resolution of the second direction D2 of the line sensor (in the Figure 2 example, the direction orthogonal to the direction in which the respective elements 41 to 43 are arranged) is determined by the distance that the object 90 moves in the second direction D2 during one scanning time. Specifically, the resolution of the second direction D2 is expressed as the product of the one scanning time and the transfer speed V2 of the object 90 in the second direction D2. Therefore, the size (in other words, the number of pixels) of the object 90 on the image corresponding to the direction of the second direction D2 (that is, Figure 4 the Y direction) varies according to the transfer speed V2. More specifically, the transfer speed V2 of the object 90 in the second direction D2 and the size of the object 90 on the image corresponding to the direction of the second direction D2 are in an inverse proportional relationship. On the other hand, the resolution of the first direction D1 of the line sensor is constant and does not depend on the transfer speed of the object 90.

[0064] In the line sensor having such characteristics, when the object 90 is transferred at a transfer speed V2 different from the pre-assumed transfer speed (that is, the design value regarding the transfer speed V2), the size of the object 90 on the image in the direction corresponding to the second direction D2 is different from the actual size. In that case, the measurement accuracy of the characteristic quantities of the shape and / or size decreases. And, the threshold value for determining the quality of the shape and / or size is preset based on the design value regarding the transfer speed V2 (in other words, based on the actual size of the object 90). Therefore, when the object 90 is transferred at a transfer speed V2 different from the pre-assumed transfer speed, the measurement accuracy of the quality of the shape and / or size also decreases. In order to suppress such a decrease in the measurement accuracy, in the present embodiment, the image obtained by the first line sensor 40 is corrected based on the transfer speed V2 in the second direction D2.

[0065] Specifically, the feature determination unit 82 corrects the size of the object 90 in the second direction D2 on the image obtained by the first line sensor 40 in a direction that reduces or eliminates the influence of the difference between the transfer speed V2 in the second direction D2 calculated by the transfer speed calculation unit 81 and the design value (hereinafter also referred to as the reference value VR) regarding the transfer speed V2 in the second direction D2 on the image (more specifically, the size of the image of the object 90 in the direction corresponding to the second direction D2).

[0066] For example, when the transfer speed V2 in the second direction D2 calculated by the transfer speed calculation unit 81 is greater than the reference value VR, the size of the object 90 in the direction corresponding to the second direction D2 on the image becomes smaller than the actual size. In that case, in order to reduce or eliminate this influence, the feature determination unit 82 corrects the image so that the size of the object 90 in the direction corresponding to the second direction D2 on the image becomes larger.

[0067] Conversely, when the transfer speed V2 in the second direction D2 calculated by the transfer speed calculation unit 81 is less than the reference value VR, the size of the object 90 in the direction corresponding to the second direction D2 on the image becomes larger than the actual size. In that case, in order to reduce or eliminate this influence, the feature determination unit 82 corrects the image so that the size of the object 90 in the direction corresponding to the second direction D2 on the image becomes smaller.

[0068] As described above, the actual transfer speed V2 of the object 90 in the second direction D2 and the size of the object 90 on the image in the direction corresponding to the second direction D2 are in an inverse proportional relationship. Therefore, if the size (i.e., the number of pixels) of the object 90 in the direction corresponding to the second direction D2 in the image obtained by the first line sensor 40 is multiplied by the ratio V2 / VR for correction, the influence of the change in the transfer speed V2 on the image can be minimized (i.e., the size of the object 90 on the image can be corrected to the actual size). In Figure 5 shows an example of such correction. When the transfer speed V2 is less than the reference value VR, the object 94 (the image of the object 90 before correction) is reduced only in the Y direction corresponding to the second direction D2 at the ratio V2 / VR, thereby obtaining the object 93 (the image of the object 90 with a large actual size). On the other hand, when the transfer speed V2 is greater than the reference value VR, the object 95 (the image of the object 90 before correction) is enlarged only in the Y direction corresponding to the second direction D2 at the ratio V2 / VR, thereby obtaining the object 93 (the image of the object 90 with a large actual size).

[0069] The feature determination unit 82 determines the features of the shape and / or size of the object 90 (i.e., the feature amount and the quality based on the feature amount) according to the image corrected based on the transfer speed V2 in this way. According to such correction, even if the resolution of the second direction D2 of the line sensor 40 changes due to the change in the transfer speed V2, and an image of the object 90 with a size different from the actual size in the direction corresponding to the second direction D2 is obtained, the image can be corrected so that the size in the second direction D2 is close to the actual size or becomes the actual size. Therefore, the feature determination unit 82 can accurately determine the features of the shape and / or size of the object 90 based on the corrected image.

[0070] Next, the correction of the image based on the transfer speed V1 in the first direction D1 will be described. When the object 90 has a speed component in the first direction D1, if the images of each scan are simply combined, Figure 3 the linear regions marked with numbers 1 to 10 as shown will be periodically shifted in the direction corresponding to the first direction D1. Therefore, a shape that is deformed in the direction (Y direction) corresponding to the first direction D1 with respect to the actual shape of the object 90 will be detected, and as a result, the measurement accuracy of the features of the shape and / or size of the object 90 will be degraded. Therefore, in order to suppress such a decrease in measurement accuracy, in the present embodiment, the image based on the transfer speed V1 in the first direction D1 is corrected.

[0071] Specifically, the feature determination unit 82 corrects the amount of offset in the coordinate values in the first direction D1 of the plurality of pixels constituting the image of the object 90 due to the transfer speed V1 in the first direction D1. Specifically, in the image region of the object 90 obtained by the Nth (N is a natural number) scan and the image region of the object 90 obtained by the (N + 1)th scan, the coordinate value of the object 90 in the first direction D1 is offset by the distance L4 that the object 90 moves in the first direction D1 during one scan time of the first line sensor 40. Therefore, the feature determination unit 82 corrects in the following manner: taking the first scan of one object 90 as a reference, the coordinate values of the image of the object 90 obtained by the second and subsequent scans are returned to the original position by the amount of offset from the reference. The correction amount (actual distance) of the coordinate value of the image region of the object 90 obtained by the Mth (M is an integer of 2 or more) scan is expressed as (M - 1) × L4. If the value of (M - 1) × L4 is divided by the size of the pixel in the first direction D1 of the first line sensor 40 (i.e., the size corresponding to 1 pixel), the correction amount in units of the number of pixels can be calculated.

[0072] Figure 6 and Figure 7 An example of such correction is shown. Each grid in the figure represents 1 pixel constituting the image. In addition, the numerical values in the figure show the coordinate values of the XY coordinate system defined by the X direction corresponding to the first direction D1 and the Y direction corresponding to the second direction D2. In this example, the Y-direction coordinate values are assigned for each scan of the first line sensor 40. That is, the Y coordinate value shows the linear image region obtained by which scan. Figure 6 The hatched grids show the pixels representing the object 90 (represented as the object 96) on the image before correction, Figure 7 The hatched grids show the pixels representing the object 90 (represented as the object 97) on the corrected image.

[0073] In addition, Figure 6 and Figure 7 illustrate a case where the direction of the assumed transfer speed V1 is the negative direction of the X coordinate and the distance L4 (the distance that the object 90 moves in the first direction D1 during one scanning time) is equal to the resolution of the first direction D1 of the line sensor 40. As shown in the figure, taking the line with a Y coordinate value of 2 as a reference, for each increase of 1 in the Y coordinate value, the X coordinate value increases by the distance L4 of the X coordinate (here, the distance of one pixel amount), and correct Figure 6 the image of the object 96 shown (the image of the object 90 that is photographed in a deformed shape due to the transfer speed V1 in the first direction D1). Thus, it is corrected to Figure 7 the image of the object 97 shown (the image of the object 90 in its actual shape).

[0074] The feature determination unit 82 determines the shape and / or size features of the object 90 based on the image corrected based on the transfer speed V1 in this way. According to this correction, even if the object 90 is transferred while laterally shifting with respect to the intended transfer direction (i.e., the second direction D2) (i.e., while offsetting in the first direction D1) and an image of the object 90 deformed in the direction corresponding to the first direction D1 is obtained, the image can be corrected to reduce or remove the deformation. Therefore, the feature determination unit 82 can accurately determine the shape and / or size features of the object 90 based on the corrected image.

[0075] In addition, the transfer speed calculation unit 81 may also calculate the transfer speeds V1 and V2 based on the color deviation amount of the image acquired by the second line sensor 50. In this case, the feature determination unit 82 may also correct the image acquired by the second line sensor 50 based on the transfer speeds V1 and V2 acquired by the second line sensor 50 and determine the shape and / or size features of the object 90 based on the corrected image.

[0076] In the present embodiment, the transfer speed calculation unit 81 corrects the images of the corresponding objects 90 based on the transfer speeds V1 and V2 calculated for each object 90. That is, the transfer speeds V1 and V2 used for correction are values inherent to each object 90. With this configuration, the transfer speed calculation unit 81 can perform more accurate correction for each object 90 using the more accurate transfer speeds V1 and V2. As a result, the features of the shape and / or size can be determined more accurately. However, the transfer speeds V1 and V2 detected for at least one object 90 at a specified timing may also be commonly used for correcting the images of multiple objects 90. In this case, the transfer speeds V1 and V2 used for correction may be updated, for example, at a specified frequency (e.g., every 5 minutes). Alternatively, the transfer speeds V1 and V2 detected when the sorter 10 starts operating may also be continuously used for correction during subsequent operation. According to the above configuration, the computational load on the transfer speed calculation unit 81 can be reduced. Also, as the transfer speeds V1 and V2, representative values (e.g., average values) of multiple objects 90 may be commonly used for correcting the images of the multiple objects 90.

[0077] In an alternative embodiment, instead of the configuration of correcting the image based on the transfer speed V2 in the second direction D2, the feature determination unit 82 may directly correct the feature amount of the shape and / or size based on the transfer speed V2 in the second direction D2. In this alternative embodiment, in order to obtain a feature amount (hereinafter also referred to as the first feature amount) equivalent to the feature amount of the shape and / or size determined based on the corrected image, the feature determination unit 82 first determines the feature amount (second feature amount) of the shape and / or size from the image (image to which the above correction is not applied) acquired by the first line sensor 40 (or the second line sensor 50). Next, the feature determination unit 82 corrects the second feature amount by correcting the feature amount component in the direction corresponding to the second direction D2 in the second feature amount based on the transfer speed V2 in the second direction D2, thereby obtaining the first feature amount.

[0078] For example, when the feature quantity to be obtained as the first feature quantity is the area or height of the object 90 (the length in the direction corresponding to the second direction D2), the feature determination unit 82 determines the value obtained by multiplying the second feature quantity by the ratio V2 / VR as the first feature quantity. When the feature quantity to be obtained as the first feature quantity is the width of the object 90 (the length in the direction corresponding to the first direction D1), the feature determination unit 82 determines the second feature quantity as the first feature quantity without correction. When the feature quantity to be obtained as the first feature quantity is the outer perimeter of the object 90, no correction is made to the length component (also referred to as Lx) in the direction corresponding to the first direction D1 in the outer perimeter, but correction is made by multiplying only the length component (also referred to as Ly) in the direction corresponding to the second direction D2 by the ratio V2 / VR, and the value of Lx + Ly × V2 / VR is determined as the first feature quantity.

[0079] The components Lx and Ly can be determined, for example, by the method described in Japanese Patent Application Laid-Open No. 2012-127706. Since this method is well-known, for a brief description, first, the image is binarized, and the pixel range representing the object 90 is determined. Next, a window of a specified size (such as 2×2 pixels, 3×3 pixels, etc.) is applied to the pixel range representing the object 90. Depending on which position within the window the pixel representing the object 90 is located, it is possible to determine whether the pixel within the window is a pixel representing the outer perimeter of the object 90 or a pixel representing the inside of the object 90, and it is possible to determine which of the components Lx and Ly the pixel representing the outer perimeter of the object 90 corresponds to. While moving the application position of the window pixel by pixel, this determination process is repeated until the attributes of all pixels representing the object 90 are determined.

[0080] If the first feature quantity of the shape and / or size is obtained in this way, the feature determination unit 82 determines the quality of the shape and / or size based on the first feature quantity by the above method. As described above, even if the second feature quantity is corrected based on the transfer speed V2 instead of correcting the image based on the transfer speed V2 to obtain the first feature quantity, the same effect as in the case of correcting the image is obtained.

[0081] In a further alternative embodiment, the feature determination unit 82 may also determine (correct) the threshold value for determining the quality of the shape and / or size based on the transfer speed V2 in the second direction D2, instead of correcting the structure of the image based on the transfer speed V2 in the second direction D2. As described above, the size of the object 90 on the image in the direction corresponding to the second direction D2 is VR / V2 times the actual size. Therefore, the feature determination unit 82 corrects the threshold value by multiplying, for example, the design value of the threshold value related to the direction corresponding to the second direction D2 by the ratio VR / V2. That is, the feature determination unit 82 corrects the threshold value so that the threshold value changes at the same ratio according to the change in the image size of the second direction D2 caused by the change in the transfer speed V2.

[0082] For example, the feature determination unit 82 may also correct by multiplying the threshold value related to the area or height (the length in the direction corresponding to the second direction D2) of the object 90 by the ratio VR / V2. In this case, the feature determination unit 82 compares the feature amount of the shape and / or size calculated from the uncorrected image with the corrected threshold value, thereby determining whether the object 90 is a qualified product or a non-qualified product. With this configuration, it is also possible to accurately determine the quality of the shape and / or size of the object 90.

[0083] In a further alternative embodiment, the transfer speeds V1 and V2 may be detected by any method other than the above method. For example, the sorting machine 10 may also be provided with an area sensor for detecting the transfer speeds V1 and V2. In this case, the feature determination unit 82 calculates the transfer speeds V1 and V2 based on the moving distance and the moving time between two timings for the same object 90 on two images obtained by the area sensor at two different timings. The moving distance may be calculated based on the difference in the positions of the density centers of gravity of the object 90 in each of the two images in the same manner as in the above example. The moving time can be calculated based on the known scanning time. Alternatively, the sorting machine 10 may use two line sensors respectively arranged at different detection positions (the photographing positions of the object 90 on the transfer path) instead of that area sensor.

[0084] The embodiments have been described above, but the above embodiments are for easy understanding of this explanation and do not limit the present invention. The present invention can be changed and improved without departing from its gist, and the present invention includes its equivalents. In addition, within the scope of being able to solve at least a part of the above problems or within the scope of achieving at least a part of the effects, any combination or any omission of the respective components described in the claims and the specification can be made.

[0085] For example, the line sensors 40 and 50 may also be any color sensors in which at least two of the R element group, G element group, and B element group are arranged so as to be separated from each other in the second direction D2, instead of the above-described 3-wire sensors. For example, the line sensors 40 and 50 may also be color sensors of a Bayer array. In this case, the transfer speeds V1 and V2 can be determined based on the color deviation amount between the red image and the blue image obtained from the R element group and the B element group.

[0086] Alternatively, the feature determination unit 82 may also determine the features of the shape and / or size of a part of the object 90 based on the transfer speed V1 and / or the transfer speed V2. Such a part may also be a part of the object 90 having a feature of a specified color. For example, in order to determine the color quality of the object 90 based on whether there is a local defect larger than a specified size, the feature determination unit 82 may also determine the feature amount of the shape and / or size of the area having a local defect (in other words, the area composed of a pixel group whose color gray value is not within the normal range on the image) based on the transfer speed V1 and / or the transfer speed V2.

[0087] Alternatively, the sorter 10 may also replace the visible light source (the light sources 31 and 32 in the above example) or, on top of that, include an electromagnetic wave irradiation source that irradiates the object 90 with electromagnetic waves having an arbitrary wavelength. Such electromagnetic waves may be, for example, at least one of X-rays and near-infrared rays. In this case, the sorter 10 may also include a line sensor having a plurality of electromagnetic wave detection elements corresponding to the wavelength of the electromagnetic waves irradiated by the electromagnetic wave irradiation source. The above various corrections can also be applied to the images obtained using such an electromagnetic wave irradiation source and line sensor.

[0088] The present invention is not limited to the sorter and can be implemented in various ways. For example, the present invention can also be implemented as a measuring device for measuring the features of the shape and / or size of the whole and / or a part of an object. Such a measuring device may, for example, include the Figure 1 sorter 10 shown, with the sorting unit 60 removed. In this case, the controller 80 may also output the determined features (or their statistics) of the object 90 to any device (such as a display, a communication interface, a storage medium, a printing device, etc.).

[0089] Explanation of reference numerals

[0090] 10...optical sorter; 20...optical detection unit; 31...first light source; 32...second light source; 33...first light; 34...second light; 40...first line sensor; 41...R element; 42...G element; 43...B element; 44...R element group; 45...R element group; 46...R element group; 50...second line sensor; 60...sorting unit; 61...nozzle; 62...valve; 63...air; 71...storage slot; 72...feeder; 73...chute; 74...qualified product discharge slot; 75...defective product discharge slot; 80...controller; 81...transfer speed calculation unit; 82...feature determination unit; 90, 91, 92, 93, 94, 95, 96, 97...object; 92G...green image; 92R...red image; 93G...green concentration centroid coordinate point; 93R...red concentration centroid coordinate point; D1...first direction; D2...second direction.

Claims

1. A measuring device for measuring the shape and / or dimensional characteristics of an entire object and / or a part thereof, characterized in that, Comprising: A transfer unit configured to transfer the object; An electromagnetic wave irradiation source configured to irradiate the object being transferred by the action of the transfer unit with electromagnetic waves; A line sensor configured to have a plurality of electromagnetic wave detection elements linearly arranged in a first direction intersecting the transfer direction of the object, and to detect at least one of the reflected electromagnetic waves irradiated from the electromagnetic wave irradiation source and reflected by the object and the transmitted electromagnetic waves transmitted through the object; A transfer speed detection unit configured to detect the transfer speed of the object being transferred in a specified direction; and A feature determination unit configured to determine the feature of the object based on the image obtained by the line sensor and the transfer speed in the specified direction.

2. The measuring device according to claim 1, wherein The feature includes a first feature quantity of the whole and / or a part of the object, The feature determination unit is further configured to correct the image obtained by the line sensor based on the transfer speed in the specified direction, and determine the first feature quantity based on the corrected image, or is configured to correct a second feature quantity determined based on the image obtained by the line sensor based on the transfer speed in the specified direction, thereby determining the first feature quantity.

3. The measuring device according to claim 1 or 2, wherein The feature includes the mass of the whole and / or a part of the object, The feature determination unit is configured to determine the mass based on the first feature quantity of the whole and / or a part of the object, The feature determination unit is further configured to correct the image obtained by the line sensor based on the transfer speed in the specified direction, and obtain the first feature quantity based on the corrected image, or is configured to correct a second feature quantity determined based on the image obtained by the line sensor based on the transfer speed in the specified direction, thereby obtaining the first feature quantity.

4. The measuring device according to claim 1 or 2, wherein The specified direction includes a second direction orthogonal to the first direction, The feature includes the mass of the whole and / or a part of the object, The feature determination unit is further configured to compare the feature quantity determined based on the image obtained by the line sensor with a threshold value determined based on the transfer speed in the second direction, thereby determining the mass.

5. The measuring device according to claim 2 or 3, wherein The specified direction includes a second direction orthogonal to the first direction, The feature determination unit is further configured to correct the size of the image in the second direction based on the transfer speed in the second direction, thereby correcting the image, or to correct the feature quantity component in the second direction in the second feature quantity based on the transfer speed in the second direction, thereby correcting the second feature quantity.

6. The measuring device according to any one of claims 1 to 5, wherein The specified direction includes the first direction, The feature determination unit is further configured to correct the coordinate values in the first direction of a plurality of pixels constituting the image based on the transfer speed in the first direction, thereby correcting the image, and determine the feature of the object based on the corrected image.

7. A sorting machine, characterized in that, Comprising: The measuring device according to any one of claims 1 to 6; and A sorting unit configured to sort the objects based on the features determined by the feature determination unit.

Images