Detection method and device, electronic equipment and storage medium

Abstract

The detection method of the embodiments includes adjusting the relative angle of the to-be-detected member and the sensor according to the warping degree of any to-be-detected region in a segmented region of the to-be-detected member, so that the to-be-detected region is perpendicular to the optical axis of the sensor, the to-be-detected member includes a plurality of segmented regions, and the segmented region includes a plurality of to-be-detected regions; collecting a collection image of each to-be-detected region in the segmented region through the sensor; splicing the collection images of the plurality of to-be-detected regions to generate a segmented image; and splicing a plurality of segmented images to generate a detection image of the to-be-detected member. The detection method and device, the electronic equipment and the nonvolatile computer readable storage medium adjust the relative angle of the to-be-detected member and the sensor according to the warping degree of any to-be-detected region in a segmented region, the quality of the collected image is high, the number of adjustments is small, the splicing accuracy of the collected image is improved, and the accumulation amount and processing amount of splicing errors are small.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a detection method, a detection device, an electronic device and a nonvolatile computer readable storage medium. BACKGROUND

[0002] At present, when detecting a workpiece (such as a wafer, a display panel, etc.), since the collection accuracy is required to be very high, the field of view range of a sensor for collecting information is generally small, therefore, a large number of collection images are collected for each workpiece, and therefore, how to accurately combine a large number of collection images into a complete and clear workpiece image to ensure the detection effect of the workpiece image is a problem to be solved. SUMMARY

[0003] The present application provides a detection method, a detection device, an electronic device and a nonvolatile computer readable storage medium.

[0004] In a first aspect, the detection method of the present application includes adjusting the relative angle of a workpiece and a sensor according to the warping degree of any to-be-detected region in the segmented region of the workpiece, so that the to-be-detected region is perpendicular to the optical axis of the sensor, the workpiece includes a plurality of segmented regions, and the segmented region includes a plurality of to-be-detected regions; collecting collection images of each to-be-detected region in the segmented region by the sensor; splicing the collection images of a plurality of to-be-detected regions to generate a segmented image; and splicing a plurality of segmented images to generate a detection image of the workpiece, so as to detect the workpiece according to the detection image.

[0005] In a second aspect, the detection device of the present application includes an adjusting module, a collecting module, a first splicing module and a second splicing module. The adjusting module is used to move a workpiece to a predetermined position, so that an identification part of the workpiece is located in the field of view range of a shooting device; the collecting module is used to shoot the identification part by the shooting device to obtain a first image, and determine a first deviation according to the first image; the first splicing module is used to control the movement of a movement platform according to the first deviation, so that the identification part is located at a preset position in the field of view range, and the preset position is the position with the best focusing performance of the shooting device; and the second splicing module is used to shoot the identification part again by the shooting device when the identification part is located at the preset position to obtain a second image, and determine a second deviation according to the second image.

[0006] In a third aspect, the electronic device of the embodiments of the present application comprises a sensor, a motion platform and a processor, the motion platform is configured to adjust the relative angle between the object and the sensor according to the warping degree of any one of the sub-regions of the object, so that the sub-region is perpendicular to the optical axis of the sensor, the object comprises a plurality of the sub-regions, and the sub-regions comprise a plurality of the sub-regions; the sensor is configured to collect the image of each of the sub-regions; the processor is configured to stitch the images of the sub-regions to generate a segmented image, and stitch a plurality of the segmented images to generate a detection image of the object, and detect the object according to the detection image.

[0007] In a fourth aspect, a non-volatile computer readable storage medium comprising a computer program of the embodiments of the present application, when the computer program is executed by one or more processors, the processor executes the detection method. The detection method comprises adjusting the relative angle between the object and the sensor according to the warping degree of any one of the sub-regions of the object, so that the sub-region is perpendicular to the optical axis of the sensor, the object comprises a plurality of the sub-regions, and the sub-regions comprise a plurality of the sub-regions; collecting the image of each of the sub-regions by the sensor; stitching the images of the sub-regions to generate a segmented image, and stitching a plurality of the segmented images to generate a detection image of the object, and detecting the object according to the detection image.

[0008] The detection method, the detection device, the electronic equipment and the nonvolatile computer readable storage medium of the application, by dividing the to-be-detected member into a plurality of segmentation regions, then when image acquisition is performed on the to-be-detected region in each segmentation region, the to-be-detected member placement angle is adjusted through the warping degree of the to-be-detected region, so that the to-be-detected region is perpendicular to the optical axis of the sensor, and then the sensor collects the collection images of a plurality of to-be-detected regions in the current segmentation region. Since the plurality of to-be-detected regions in the same segmentation region are close in distance, when the optical axis of the sensor is perpendicular to one of the to-be-detected regions, the clarity of the collection images obtained by collecting the plurality of to-be-detected regions in the current segmentation region is high, and each segmentation region only needs to adjust the to-be-detected member once, so that the number of to-be-detected member adjustments is small, thereby preventing the number of to-be-detected member adjustments from affecting the subsequent splicing accuracy of the collection images, thereby ensuring the clarity of a single collection image while improving the splicing accuracy of the collection images. In addition, the application first splices the collection images in the same segmentation region to generate a plurality of segmentation images, and then splices a plurality of segmentation images to generate a detection image of the to-be-detected member. Compared with splicing all collection images as a detection image, there is a splicing error and a splicing processing amount when splicing any two adjacent collection images, resulting in a large cumulative amount of splicing error and a large splicing processing amount. Therefore, splicing the collection images in a single segmentation region first and then splicing adjacent segmentation images has a smaller cumulative amount of splicing error and processing amount, which can significantly reduce the splicing error and the splicing processing amount, and improve the accuracy and splicing efficiency of the final spliced detection image.

[0009] Additional aspects and advantages of the application will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following description and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0011] Figure 1 is a flowchart of the detection method of some embodiments of the application;

[0012] Figure 2 is a module schematic diagram of the detection device of some embodiments of the application;

[0013] Figure 3 is a planar schematic diagram of the electronic equipment of some embodiments of the application;

[0014] Figure 4is a planar schematic view of a test piece of certain embodiments of the present application;

[0015] Figure 5 is a flowchart of a detection method of certain embodiments of the present application;

[0016] Figure 6 and Figure 7 is a schematic diagram of the principle of a detection method of certain embodiments of the present application;

[0017] Figure 8 and Figure 9 is a flowchart of a detection method of certain embodiments of the present application;

[0018] Figure 10 is a schematic diagram of the principle of a detection method of certain embodiments of the present application;

[0019] Figure 11 is a flowchart of a detection method of certain embodiments of the present application;

[0020] Figure 12 and Figure 13 is a schematic diagram of the principle of a detection method of certain embodiments of the present application;

[0021] Figure 14 is a flowchart of a detection method of certain embodiments of the present application; and

[0022] Figure 15 is a connection diagram of a processor and a computer readable storage medium of certain embodiments of the present application. DETAILED DESCRIPTION

[0023] The embodiments of the present application will be further described below with reference to the drawings. The same or similar reference numerals are used throughout the drawings and identical or similar elements or elements having identical or similar functions are denoted with the same or similar reference numerals throughout. In addition, the embodiments of the present application described below with reference to the drawings are exemplary and are only used to explain the embodiments of the present application and cannot be understood as a limitation of the present application.

[0024] Referring to Figures 1 to 3 , the detection method of the embodiments of the present application comprises the following steps:

[0025] 011: adjusting the relative angle of the test piece 200 and the sensor according to the warping degree of any test region 211 in the segmented region 210 of the test piece 200, so that the test region 211 is perpendicular to the optical axis of the sensor, the test piece 200 comprises a plurality of segmented regions 210, and the segmented region 210 comprises a plurality of test regions 211;

[0026] 012: collecting a collection image of each test region 211 in the segmented region 210 by the sensor 40;

[0027] 013: stitching the acquisition images of the plurality of test regions 211 to generate a segmentation image; and

[0028] 014: stitching the segmentation images to generate a detection image of the test piece 200, so as to detect the test piece 200 according to the detection image.

[0029] The detection device 10 of the embodiment of the application comprises an adjusting module 11, an acquisition module 12, a first stitching module 13 and a second stitching module 14. The adjusting module 11 is configured to adjust the relative angle between the test piece 200 and the sensor 40 according to the warping degree of any test region 211 in the segmentation region 210 of the test piece 200, so that the test region 211 is perpendicular to the optical axis of the sensor 40; the acquisition module 12 is configured to acquire an acquisition image of each test region 211 in the segmentation region 210 by the sensor 40; the first stitching module 13 is configured to stitch the acquisition images of the plurality of test regions 211 to generate a segmentation image; and the second stitching module 14 is configured to stitch the segmentation images to generate a detection image of the test piece 200, so as to detect the test piece 200 according to the detection image. That is, the step 011 can be implemented by the adjusting module 11, the step 012 can be executed by the acquisition module 12, the step 013 can be executed by the first stitching module 13, and the step 014 can be executed by the second stitching module 14.

[0030] The electronic device 100 of the embodiment of the application comprises a processor 20, a motion platform 30 and a sensor 40. The motion platform 30 is configured to adjust the relative angle between the test piece 200 and the sensor 40 according to the warping degree of any test region 211 in the segmentation region 210 of the test piece 200, so that the test region 211 is perpendicular to the optical axis of the sensor 40. The test piece 200 comprises a plurality of segmentation regions 210, and each segmentation region 210 comprises a plurality of test regions 211. The sensor 40 is configured to acquire an acquisition image of each test region 211 in the segmentation region 210. The processor 20 is configured to stitch the acquisition images of the plurality of test regions 211 to generate a segmentation image, and stitch the segmentation images to generate a detection image of the test piece 200, so as to detect the test piece 200 according to the detection image. That is, the step 011 can be executed by the motion platform 30, the step 012 can be executed by the sensor 40, and the steps 013 and 014 can be executed by the processor 20.

[0031] Specifically, the electronic device 100 can be a measuring machine. It can be understood that the specific form of the electronic device 100 is not limited to the measuring machine, and can also be any device capable of detecting the test piece 200.

[0032] The electronic device 100 comprises a processor 20, a motion platform 30 and a sensor 40. The processor 20 and the sensor 40 can be arranged on the motion platform 30. The motion platform 30 can be used to carry a test piece 200, and the motion platform 30 moves to drive the sensor 40 and / or the test piece 200 to move, so that the sensor 40 collects information of the test piece 200.

[0033] For example, the motion platform 30 comprises an XY motion platform 31 and a Z motion platform 32, and the sensor 40 is arranged on the motion platform 30, specifically, the sensor 40 is arranged on the Z motion platform 32. The XY motion platform 31 is used to control the test piece 200 to move along a horizontal plane, so as to change the relative position of the test piece 200 and the sensor 40 in the horizontal plane, and the Z motion platform 32 is used to control the sensor 40 to move along a direction perpendicular to the horizontal plane, so that the three-dimensional position (i.e., the relative position in the horizontal plane and the relative position in the direction perpendicular to the horizontal plane) of the sensor 40 relative to the test piece 200 is realized by cooperation of the XY motion platform 31 and the Z motion platform 32. In addition, the Z motion platform 32 can also rotate the sensor 40 to change the relative angle of the sensor 40 and the test piece 200, or the XY motion platform 31 can rotate the test piece 200 to change the relative angle of the sensor 40 and the test piece 200.

[0034] It can be understood that the motion platform 30 is not limited to the above structure, as long as it can change the three-dimensional position and the relative angle of the sensor 40 relative to the test piece 200.

[0035] The sensor 40 can be one or more, and the multiple sensors 40 can be different types of sensors 40, for example, the sensor 40 can comprise a spectral confocal sensor 40, a visible light camera, an infrared camera, a depth camera, etc.

[0036] The test piece 200 can be a panel (such as a display panel, a touch panel, etc.) or a wafer. In the embodiments of the present application, the test piece 200 is taken as a wafer for example.

[0037] Please refer to Figure 4The motion platform 30 generally has a predetermined bearing area for placing the test piece 200, and after the test piece 200 is placed in the area, the information of the test piece 200 is collected. It can be understood that due to the manufacturing error of the test piece 200 or the forming structure of the test piece 200 itself, the warping degree of different positions on the surface of the test piece 200 can be different. Since the field of view range of the sensor 40 is relatively small compared with the test piece 200, the field of view range can only cover a part of the test piece 200, and when detecting, the test piece 200 is divided into a plurality of larger segmentation areas 210, and the plurality of segmentation areas 210 are further divided into a plurality of smaller test areas 211, such as test areas A1 to A9, test areas A10 to A18, test areas A19 to A27, test areas A28 to A36, each of which constitutes a segmentation area, and the field of view range of the sensor 40 can cover one or more test areas 211 (for example, one) at a time, and multiple collections are required to obtain all image information of the entire test piece 200.

[0038] Therefore, before the sensor 40 collects the collection image of each test area 211, the relative angle between the test piece 200 and the sensor 40 can be adjusted according to the warping degree of the test area 211, so that the test piece 200 can face the sensor 40, that is, the test area 211 is perpendicular to the optical axis of the sensor 40. At this time, the focusing effect of the sensor 40 on the test area 211 reaches the best, so that the clarity of the collection image of the test area 211 collected by the sensor 40 is higher.

[0039] Of course, in order to prevent the number of relative angle adjustments between the test piece 200 and the sensor 40 from affecting the splicing accuracy of the subsequent collection image, the present application selects one test area 211 in each segmentation area 210 as an alignment area, and then adjusts the angle according to the warping degree of the test area 211, so that the optical axis of the sensor 40 is perpendicular to the test area 211, and then collects the collection image of all test areas 211 in the current segmentation area 210 at the adjusted angle. The warping degree of each test area 211 can be preset, or the warping degree of the test area 211 can be determined by identifying the image of the test area 211 in the collection image after the sensor 40 collects the current test image; the alignment area can be the test area 211 located in the center area of the segmentation area 210 (such as test areas A5, A14, A23 and A32, etc.), and when the test area 211 in the center area faces the sensor 40, the relative angle between the surrounding other test areas 211 and the sensor 40 is small, thereby further improving the clarity of the collection image on the basis of ensuring a small number of angle adjustment times.

[0040] In other embodiments, when the difference between the warping degrees of different to-be-tested regions 211 in the same segmentation region 210 is too large (e.g., the difference between the warping degrees of any two to-be-tested regions 211 is greater than a preset warping threshold), the angle between the optical axis of the sensor 40 and the other to-be-tested regions 211 is still large after one of the to-be-tested regions 211 directly faces the sensor 40, so that the acquisition quality of the acquisition image of the other to-be-tested regions 211 collected by the sensor 40 is poor, which affects the splicing effect of the subsequent acquisition image. At this time, multiple alignment regions can be determined for the segmentation region 210, each alignment region corresponding to a to-be-tested region 211 of a part of the segmentation region 210, e.g., the segmentation region 210 is divided into four sub-segmentation regions 210, and then one alignment region is determined for each sub-segmentation region 210 to realize image acquisition of all to-be-tested regions 211 in the sub-segmentation region 210, so as to ensure that the angle between the optical axis of the sensor 40 and each to-be-tested region 211 is small (e.g., less than a preset angle) when the sensor 40 collects the acquisition image of the to-be-tested region 211, to ensure the acquisition quality of each acquisition image and improve the splicing effect of the acquisition image.

[0041] After obtaining all the acquisition images in each segmentation region 210, all the acquisition images in each segmentation region 210 can be spliced to generate multiple segmentation images, and then the multiple segmentation images can be spliced to generate a detection image of the to-be-tested piece 200. Compared with splicing all the acquisition images into a detection image, there is a splicing error and a splicing processing amount when splicing any two adjacent acquisition images, which results in a large cumulative amount of splicing error and a large splicing processing amount. Therefore, the cumulative amount of splicing error and the processing amount caused by splicing the acquisition images in a single segmentation region 210 first and then splicing adjacent segmentation images are both small, which can significantly reduce the splicing error and the splicing processing amount, and improve the accuracy and splicing efficiency of the final spliced detection image.

[0042] The detection method, the detection device 10 and the electronic equipment 100 of the present application divide the to-be-detected member 200 into a plurality of segmentation regions 210, and then adjust the placement angle of the to-be-detected member 200 by the degree of warping of the to-be-detected region 211 in each segmentation region 210 when collecting images of the to-be-detected region 211, so that the to-be-detected region 211 is perpendicular to the optical axis of the sensor 40, and then the sensor 40 collects the collection images of a plurality of to-be-detected regions 211 in the current segmentation region 210. Since the plurality of to-be-detected regions 211 in the same segmentation region 210 are close in distance, when the optical axis of the sensor 40 is perpendicular to one of the to-be-detected regions 211, the collection images of the plurality of to-be-detected regions 211 in the current segmentation region 210 have high clarity. Each segmentation region 210 only needs to adjust the to-be-detected member 200 once, so the number of adjustments of the to-be-detected member 200 is small, thereby preventing the number of adjustments of the to-be-detected member 200 from affecting the accuracy of splicing of the collection images, thereby ensuring the clarity of a single collection image while improving the accuracy of splicing of the collection images. In addition, the present application first splices the collection images in the same segmentation region 210 to generate a plurality of segmentation images, and then splices the plurality of segmentation images to generate a detection image of the to-be-detected member 200. Compared with splicing all collection images to generate a detection image, there is a splicing error and a splicing processing amount when splicing any two adjacent collection images, resulting in a large cumulative amount of splicing error and a large splicing processing amount. Therefore, splicing the collection images in a single segmentation region 210 first and then splicing adjacent segmentation images has a small cumulative amount of splicing error and a small processing amount, which can significantly reduce the splicing error and the splicing processing amount, and improve the accuracy and splicing efficiency of the final spliced detection image.

[0043] Please refer to Figure 2 、 Figure 3 and Figure 5 In some embodiments, step 013 comprises:

[0044] 0131: determining a first overlapping region of any two adjacent collection images; and

[0045] 0132: splicing any two adjacent collection images according to the first overlapping region to generate a segmentation image.

[0046] In some embodiments, the first splicing module 13 is further configured to determine a first overlapping region of any two adjacent collection images; and splice any two adjacent collection images according to the first overlapping region to generate a segmentation image. That is, steps 0131 and 0132 can be performed by the first splicing module 13.

[0047] In some embodiments, the processor 20 is further configured to determine a first overlap region of any two adjacent captured images; and stitch the any two adjacent captured images according to the first overlap region to generate a segmented image. That is, the step 0131 and the step 0132 can be performed by the processor 20.

[0048] Specifically, when stitching the captured images in the same segmented region 210, first, two adjacent captured images are obtained, then image recognition is performed to determine the overlap part (hereinafter referred to as the first overlap region) in the two adjacent captured images, and then the two adjacent captured images are stitched according to the first overlap region (for example, the two captured images are stacked so that the first overlap regions of the two captured images coincide with each other), and after the two captured images are stitched, a new captured image is obtained, then the new captured image and the adjacent captured image are again subjected to image recognition to determine the first overlap region of the new captured image and the adjacent captured image, so as to sequentially stitch all the captured images in the segmented region 210 into a segmented image.

[0049] For example, referring to Figure 6 and Figure 7 , the captured images are sequentially arranged according to the capture positions, the captured image P1 and the captured image P2 located at the upper left corner of the segmented region 210 are first stitched according to the first overlap region X1 of the two captured images, then the captured image P1 and the captured image P2 are combined into a new captured image E1, then the captured image E1 and the captured image P3 adjacent to the captured image E1 (i.e., adjacent to the captured image P2) are again stitched according to the first overlap region X1 of the two captured images, so as to sequentially stitch all the captured images in the segmented region 210 to generate a segmented image, such as sequentially stitching the plurality of captured images in the segmented region 210 to realize accurate stitching of the captured images. In other embodiments, the first overlap region X1 of any two adjacent captured images in the segmented region 210 can be first identified, and when stitching, the first overlap region X1 that has been identified is used to quickly stitch all the adjacent captured images in the segmented region 210 to obtain a segmented image, so as to improve the stitching efficiency.

[0050] For example, referring to Figure 2 , Figure 3 and Figure 8 , in some embodiments, the step 0132 comprises:

[0051] 01321: cropping the first overlap region of one of the any two adjacent captured images to generate a cropped captured image;

[0052] 01322: stitching the cropped captured image and the captured image having the same first overlap region and not being cropped to generate a segmented image.

[0053] In some embodiments, the first stitching module 13 is configured to crop the first overlap region of one of the two adjacent captured images to generate a cropped captured image, and stitch the cropped captured image and the other captured image having the same first overlap region to generate the segmented image. In other words, the step 1321 and the step 01322 can be performed by the first stitching module 13.

[0054] In some embodiments, the processor 20 is further configured to crop the first overlap region of one of the two adjacent captured images to generate a cropped captured image, and stitch the cropped captured image and the other captured image having the same first overlap region to generate the segmented image. In other words, the step 1321 and the step 01322 can be performed by the processor 20.

[0055] In detail, referring to Figure 6 and 7 , when stitching two adjacent captured images according to the first overlap region X1, the first overlap region X1 of one of the two captured images (e.g., the captured image P1) can be cropped to generate a cropped captured image, and then the cropped captured image and the other adjacent captured image (e.g., the captured image P2) can be stitched to generate a stitched captured image E1 which only retains the first overlap region X1 of one of the two captured images, so that all the adjacent captured images can be accurately stitched to generate the segmented image T0. In this way, the stitching of any two adjacent captured images can be quickly and accurately implemented by image cropping to generate the segmented image T0.

[0056] Referring to Figure 2 , Figure 3 and Figure 9 , in some embodiments, the step 013 further includes:

[0057] 0133: identifying a to-be-tested image region in the captured images corresponding to the to-be-tested region 211; and

[0058] 0134: stitching the to-be-tested image regions in the plurality of captured images to generate the segmented image.

[0059] In some embodiments, the first stitching module 13 is configured to identify a to-be-tested image region in the captured images corresponding to the to-be-tested region 211, and stitch the to-be-tested image regions in the plurality of captured images to generate the segmented image. In other words, the step 133 and the step 0134 can be performed by the first stitching module 13.

[0060] In some embodiments, the processor 20 is further configured to identify a to-be-tested image region corresponding to the to-be-tested region 211 in the captured image; and stitch the to-be-tested image regions in the plurality of captured images to generate the segmentation image. That is, the step 133 and the step 134 can be performed by the processor 20.

[0061] Specifically, it can be understood that, for the to-be-tested piece 200 such as a wafer or a panel, the surface thereof is composed of a plurality of circuits as minimum repeating units, and when the to-be-tested piece 200 is detected, only the circuit part is concerned, each minimum repeating unit of the to-be-tested piece 200 can be regarded as a to-be-tested region 211, and the field of view range of the sensor 40 can not be just to cover the to-be-tested region 211, but to cover a part of the region around the to-be-tested region 211, such as a part of the region around the to-be-tested region 211 which does not cover the circuit or a part of an adjacent to-be-tested region 211.

[0062] Therefore, referring to Figure 10 When stitching, the captured image P0 can be first subjected to image recognition to determine the to-be-tested image region R0 corresponding to the to-be-tested region 211 in the captured image, and then after the to-be-tested image region R0 corresponding to each captured image P0 is obtained, the adjacent to-be-tested image regions R0 can be directly stitched to obtain the segmentation image T0. It can be understood that there is actually no overlapping part between different to-be-tested regions 211, and therefore the overlapping part of the adjacent to-be-tested image regions R0 need not be identified, and the adjacent to-be-tested image regions R0 can be simply stitched, which is relatively simple and has high stitching accuracy. In addition, since the to-be-tested image region R0 is generally located in the central region of the captured image P0, the definition of the to-be-tested image region R0 is higher than that of the edge region of the captured image P0, and thus the definition of the stitched segmentation image T0 can be improved.

[0063] Referring to Figure 2 , Figure 3 and Figure 11 In some embodiments, the step 014 includes:

[0064] 0141: identifying a second overlapping region of any two adjacent segmentation images; and

[0065] 0142: stitching the any two adjacent segmentation images according to the second overlapping region to generate a detection image.

[0066] In some embodiments, the second stitching module 14 is further configured to identify a second overlapping region of any two adjacent segmentation images; and stitch the any two adjacent segmentation images according to the second overlapping region to generate a detection image. That is, the step 0141 and the step 0142 can be performed by the second stitching module 14.

[0067] In some embodiments, the processor 20 is further configured to identify a second overlapping region of any two adjacent segmented images; and stitch the any two adjacent segmented images according to the second overlapping region to generate the detection image. That is, the step 0141 and the step 0142 can be performed by the processor 20.

[0068] Specifically, in stitching the detection image of the test piece 200, first, two adjacent segmented images are obtained, then image recognition is performed to determine the overlapping part (hereinafter referred to as the second overlapping region) of the two adjacent segmented images, then the two adjacent segmented images are stitched according to the second overlapping region (for example, the two segmented images are stacked so that the second overlapping regions of the two segmented images are overlapped together), and after the two segmented images are stitched, a new segmented image is obtained, then the new segmented image and the adjacent segmented image are again subjected to image recognition to determine the second overlapping region of the new segmented image and the adjacent segmented image, thereby sequentially combining all the segmented images in the segmented region 210 into one segmented image.

[0069] For example, referring to Figure 12 and Figure 13 , the segmented image T1 and the segmented image T2 located at the upper left corner of the test piece 200 are first stitched according to the second overlapping region X2 of the two segmented images, then the segmented image T1 and the segmented image T2 are combined into a new segmented image E2, then the segmented image E2 and the segmented image T3 adjacent to the segmented image E2 (i.e., adjacent to the segmented image T1) are again stitched according to the second overlapping region X2 of the two segmented images, thereby sequentially stitching all the segmented images of the test piece 200 to generate the detection image M0, such as sequentially stitching the plurality of segmented images in the test piece 200, thereby achieving accurate stitching of the detection image M0. In other embodiments, the second overlapping region of any two adjacent segmented images in the test piece 200 can be identified first, and when stitching, the second overlapping region that has been identified is used to quickly stitch all the adjacent segmented images in the segmented region 210, thereby obtaining the detection image M0, which can improve the stitching efficiency.

[0070] When stitching two adjacent segmented images (such as the segmented images T1 and T2) according to the second overlapping region, the second overlapping region X2 of one of the two segmented images T1 can be cropped to generate a cropped segmented image, then the cropped segmented image and the adjacent segmented image T2 which is not cropped are stitched to generate a stitched segmented image which only retains the second overlapping region X2 of one of the two segmented images, thereby accurately stitching all the adjacent segmented images together to generate the detection image M0. In this way, through image cropping, the stitching of any two adjacent segmented images can be quickly and accurately achieved to generate the detection image M0.

[0071] For example, referring to Figure 2 ,Figure 3 and Figure 14 In some embodiments, the detection method further comprises:

[0072] 015: determining whether the collected images are valid;

[0073] 016: if yes, proceeding to a step of splicing the collected images of the plurality of to-be-detected regions 211 to generate the segmentation image.

[0074] In some embodiments, the detection device 10 further comprises a determination module 15. The determination module 15 is configured to determine whether the collected images are valid; and the first splicing module 13 is further configured to splice the collected images of the plurality of to-be-detected regions 211 to generate the segmentation image when the collected images are valid. That is, the step 015 can be performed by the determination module 15, and the step 016 can be performed by the first splicing module 13.

[0075] In some embodiments, the processor 20 is further configured to determine whether the collected images are valid; and splice the collected images of the plurality of to-be-detected regions 211 to generate the segmentation image when the collected images are valid. That is, the step 015 and the step 016 can be performed by the processor 20.

[0076] Specifically, before splicing the collected images, in order to ensure the accuracy of the segmentation image after splicing, it is necessary to determine the validity of the collected images first. It can be understood that the collected images may be abnormal when collected, such as abnormal collection parameters of the sensor 40, different collection parameters (such as different shooting distances) are used for different to-be-detected regions 211, resulting in that the collected images are too large or too small, at this time, the collected images are obviously not valid, which will seriously affect the accuracy of the segmentation image obtained by subsequent image splicing; or the sensor 40 misses the collected images of one or more to-be-detected regions 211, resulting in that the number of collected images is not equal to the preset number threshold (such as the preset number threshold can be determined according to the number of to-be-detected regions 211 and the number of to-be-detected regions 211 collected by the sensor 40 each time), at this time, the collected images cannot be spliced into a complete segmentation image, and obviously, the collected images are not valid; or the clarity of the spectral signal corresponding to the collected image is not greater than the preset clarity threshold, which indicates that the collection quality of the collected image is poor, therefore, at this time, the collected images are obviously not valid.

[0077] Therefore, when judging whether the collected image is valid, it can be judged whether the size of the collected image is located in the preset size range, whether the number of the collected images is equal to the preset number threshold, and / or whether the definition of the spectral signal corresponding to the collected image is greater than the preset definition threshold. In the embodiments of the present application, in order to ensure that the effectiveness of the collected image is the highest, so as to maximize the accuracy of the segmented image, only when the size of the collected image is located in the preset size range, the number of the collected images is equal to the preset number threshold, and the definition of the spectral signal of the collected image is greater than the preset definition threshold, it is determined that the collected image is valid. When the collected image is invalid, the collected image of the test piece 200 can be re-collected until the collected image is valid.

[0078] Referring to Figure 15 One or more of the embodiments of the present application include a non-volatile computer readable storage medium 300 containing a computer program 302, which, when executed by one or more processors 20, causes the processor 20 to perform the detection method of any of the above embodiments.

[0079] For example, in conjunction with Figures 1 to 3 When the computer program 302 is executed by one or more processors 20, the processor 20 performs the following steps:

[0080] 011: Adjust the relative angle of the test piece 200 and the sensor according to the warping degree of any test region 211 in the segmented region 210 of the test piece 200, so that the test region 211 is perpendicular to the optical axis of the sensor, the test piece 200 includes a plurality of segmented regions 210, and the segmented region 210 includes a plurality of test regions 211;

[0081] 012: Collect the collected image of each test region 211 in the segmented region 210 by the sensor 40;

[0082] 013: Splice the collected images of a plurality of test regions 211 to generate a segmented image; and

[0083] 014: Splice a plurality of segmented images to generate a detection image of the test piece 200.

[0084] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "exemplary embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the exemplary description of the above terms does not necessarily refer to the same embodiment or example. Also, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in an appropriate manner. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0085] Any process or method descriptions or descriptions of the flow diagrams in the specification or otherwise described herein can be understood as representing the steps of the code of the modules, segments or portions of the respective processes, including one or more steps for implementing specific logic functions or steps, and the scope of the preferred embodiments of the present application includes other implementations that can not be shown or discussed in the specification. It will be understood by those skilled in the art that the scope of the preferred embodiments of the present application includes additional implementations in which the functions can be performed in an order different from that shown or discussed, including functions performed in substantially simultaneous fashion, or in reverse order, as appropriate, according to the function involved, as will be understood by those skilled in the art.

[0086] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above-described embodiments within the scope of the present application.

Claims

1. A method of detection, characterized in that, The method comprises the following steps: adjusting the relative angle between the wafer and the sensor according to the warping degree of any one of the divided regions of the wafer, so that the divided region is perpendicular to the optical axis of the sensor, the wafer comprises a plurality of the divided regions, and the divided region comprises a plurality of the divided regions; collecting the collection image of each of the divided regions by the sensor; stitching the collection images of the divided regions to generate a divided image; stitching a plurality of the divided images to generate a detection image of the wafer, and detecting the wafer according to the detection image; the wafer is a wafer, and each minimum repeating unit of the wafer is a divided region; the stitching of the collection images of the divided regions to generate a divided image comprises: identifying the divided image region corresponding to the divided region in the collection image; and stitching the divided image regions in a plurality of the collection images to generate the divided image. the stitching of the collection images of the divided regions to generate a divided image comprises: determining a first overlapping region of any two adjacent collection images; and 2. The detection method according to claim 1, characterized in that, stitching any two adjacent collection images according to the first overlapping region to generate the divided image. the stitching of any two adjacent collection images according to the first overlapping region to generate the divided image comprises: cropping the first overlapping region of one of the two adjacent collection images to generate a cropped collection image; 3. The detection method according to claim 2, characterized in that, stitching the cropped collection image and the collection image with the same first overlapping region and not being cropped to generate the divided image. the stitching of a plurality of the divided images to generate a detection image of the wafer, and detecting the wafer according to the detection image comprises: identifying a second overlapping region of any two adjacent divided images; and 4. The method of claim 1, wherein, stitching any two adjacent divided images according to the second overlapping region to generate the detection image. The method further comprises the following steps: determining whether the collection image is valid; 5. The method of claim 1, wherein if yes, entering the step of stitching the collection images of the divided regions to generate a divided image. the determination of whether the collection image is valid comprises: determining whether the size of the collection image is within a preset size range, whether the number of the collection images is equal to a preset number threshold, and / or whether the clarity of the spectrum signal corresponding to the collection image is greater than a preset clarity threshold.

6. The detection method according to claim 5, characterized in that, The field of view of the sensor covers the divided region. The method comprises the following steps:

7. The method of claim 1, wherein, an adjusting module is configured to adjust the relative angle between the wafer and the sensor according to the warping degree of any one of the divided regions of the wafer, so that the divided region is perpendicular to the optical axis of the sensor, the wafer comprises a plurality of the divided regions, and the divided region comprises a plurality of the divided regions; 8. A detection device, characterized in that a collecting module is configured to collect the collection image of each of the divided regions by the sensor; a first stitching module is configured to stitch the collection images of the divided regions to generate a divided image; and a second stitching module is configured to stitch a plurality of the divided images to generate a detection image of the wafer, and detect the wafer according to the detection image. ​ a second splicing module, configured to splice a plurality of the segmented images to generate a detection image of the workpiece, so as to detect the workpiece according to the detection image; the workpiece is a wafer, and each minimum repeating unit of the workpiece is taken as a detection region. The first splicing module is further configured to identify a detection image region corresponding to the detection region in the collected image, and splice the detection image regions in a plurality of the collected images to generate the segmented image.

9. An electronic device, comprising: A sensor, a motion platform and a processor are included, the motion platform is configured to adjust a relative angle between the workpiece and the sensor according to a warping degree of any detection region in a segmented region of the workpiece, so as to make the detection region perpendicular to an optical axis of the sensor, the workpiece includes a plurality of the segmented regions, and the segmented region includes a plurality of the detection regions; the sensor is configured to collect a collected image of each detection region in the segmented region; the processor is configured to splice the collected images of a plurality of the detection regions to generate a segmented image; and splice a plurality of the segmented images to generate a detection image of the workpiece, so as to detect the workpiece according to the detection image; the workpiece is a wafer, and each minimum repeating unit of the workpiece is taken as a detection region. The processor is further configured to identify a detection image region corresponding to the detection region in the collected image; and splice the detection image regions in a plurality of the collected images to generate the segmented image.

10. A non-volatile computer readable storage medium including a computer program, the computer program, when executed by a processor, causes the processor to perform the detection method of any one of claims 1-7.

Images