Automatic optical detection method and system

Abstract

The present invention proposes an automatic optical inspection method and system, the method comprising: step 1, material loading; step 2, global positioning; step 3, coordinate conversion, so as to obtain the central image coordinates (x i ,y i ) coincides with the center of the high-resolution camera image (x i ',y i '); Step 4, manually select a qualified test object and move it under the high-resolution camera; Step 5, perform autofocus to obtain a high-definition image of the test object, and the programming module sets parameters according to the high-definition image; Step 6, automatic detection. The present invention is suitable for tiny test objects that cannot be positioned on the conveyor, and has the advantages of high precision, strong compatibility and fast detection speed.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic optical detection and visual detection, and in particular to an automatic optical detection method and system. Background Art

[0002] With the development of machine vision technology, automatic optical inspection equipment is widely used in high-end precision inspection scenarios in the field of integrated circuits and semiconductors, replacing the human eye to complete product quality inspection. Automatic optical inspection equipment can greatly improve the inspection speed and accuracy, improve inspection efficiency, and prevent inspection inconsistencies caused by human eye fatigue.

[0003] At present, the mainstream automatic optical inspection equipment is mainly online, which uses a conveyor to transport the parts to be inspected to the designated location, and controls the movement of the camera based on the programmed program to perform defect detection. For example, the Chinese invention patent "An automatic optical inspection device and method suitable for artificial intelligence PCB detection" with application number CN201811526166.1 proposes an automatic optical inspection device and method for PCB detection, including a transmission platform, a first mobile bracket, a second mobile bracket and a scanning camera. This method is often complex to program and is suitable for large-scale and large-sized inspection objects, but not for small inspection objects that cannot be positioned on the conveyor. Summary of the invention

[0004] The purpose of the present invention is to provide an automatic optical detection method and system, which is suitable for tiny detection objects that cannot be positioned on a conveyor device, and has the advantages of high accuracy, strong compatibility, and fast detection speed. The technical solution adopted by the present invention is as follows:

[0005] An automatic optical detection method comprises the following steps:

[0006] Step 1, loading, specifically including the following steps:

[0007] Place the material box containing the object to be inspected into the clamping unit, which automatically clamps the material box; the inspection object is placed in the material box; the clamping unit is installed on the XY motion platform to drive the inspection object to achieve planar motion;

[0008] Step 2: Global positioning, including the following steps:

[0009] Control the XY platform to move the material box to the bottom of the wide-view camera, take a global picture of the material box, execute the global positioning algorithm, and obtain the number n of detection objects in the material box and the central image coordinates (x i ,y i ), i∈(1,n);

[0010] Step 3: Coordinate transformation to obtain the center image coordinates (xi ,y i ) coincides with the center of the high-resolution camera image (x i ',y i '), specifically comprising the following steps:

[0011] Step 3A: Obtain the corresponding relationship P between the image pixel distance and the physical distance in the X direction x , specifically including the following steps:

[0012] Control the XY motion platform to move the Mark point to the bottom of the wide-view camera, capture the global image, and record the image coordinates (x1, y1) of the Mark point in the global image; the Mark point is an arbitrary point on the material box;

[0013] Control the XY platform to move a distance l1 in the X direction, and record the image coordinates (x2, y1) of the Mark point in the global image after the movement;

[0014] The corresponding relationship between the image pixel distance and the physical distance in the X direction can be obtained:

[0015] Step 3B: Obtain the corresponding relationship P between the image pixel distance and the physical distance in the Y direction y , specifically including the following steps:

[0016] Control the XY motion platform to move the Mark point to the bottom of the wide-field camera, capture the global image, and record the image coordinates (x1, y1) of the Mark point in the global image;

[0017] Control the XY platform to move a distance l2 in the Y direction, and record the image coordinates (x1, y2) of the Mark point in the global image after the movement;

[0018] The corresponding relationship between the image pixel distance and the physical distance in the Y direction can be obtained:

[0019] Step 3C: First, control the XY motion platform to move the clamping unit to the position directly below the wide-view camera and record the coordinates (x c ,y c );

[0020] Then, move the clamping unit to the bottom of the high-resolution camera so that the Mark point is located at the center of the high-resolution camera image and record the physical coordinates of the Mark point at this time (x' c ,y' c );

[0021] Step 3D: For each detected object in step 2, the central image coordinate (x i ,y i), i ∈ (1, n), the corresponding physical coordinates (x i ', y i ) can be obtained, where i ∈ (1, n), and:

[0022]

[0023] Step 4: Manually select a qualified detection object and move it under the high-resolution camera;

[0024] Step 5: Perform autofocus to obtain a high-definition image of the detection object. The programming module sets the detection algorithm parameters according to the high-definition image and saves the program template;

[0025] Step 6: Automatic detection, specifically including the following steps:

[0026] Step 6A: According to the physical coordinates (x i ', y i ) output in Step 3D, control the XY platform to move the i-th detection object to under the high-resolution camera;

[0027] Step 6B: The autofocus module outputs a high-definition image of the current detection object and sends it to the message queue through the image acquisition module;

[0028] Step 6C: The image processing module obtains the image from the message queue, performs detection according to the program template in the programming module, outputs the detection result, and saves the detection result to the database module; record i = i + 1;

[0029] Step 6D: If i < n, repeat Steps 6A to 6C until i = n, that is, all detection objects are detected.

[0030] Preferably, in Step 5, the programming module specifically performs the following steps:

[0031] Step 5A: Set the light source type and brightness;

[0032] Step 5B: Set the ignored area;

[0033] Step 5C: Set the detection sensitivity and detection threshold;

[0034] Step 5D: Save the program template.

[0035] Preferably, in Step 5, the autofocus specifically includes the following steps:

[0036] Step 51: Move the detection object under the high-resolution camera and turn on the camera for real-time photographing;

[0037] Step 52, setting a focus distance Δ, controlling the Z-axis lifting mechanism to move a distance Δ in the negative direction, and then controlling the Z-axis lifting mechanism to move a distance 2Δ in the positive direction;

[0038] During the movement, high-resolution camera images are collected in real time to obtain an image set {P1, P2, ...P j}, and its corresponding Z-axis coordinate set {z1, z2, ... z j};

[0039] Where j is the total number of images collected in real time; P j is the jth image;

[0040] Step 53: for the image set {P1, P2, ... P j}, and obtain the definition set {S1, S2, ... S j};

[0041] Among them, S j is the clarity of the jth image;

[0042] Step 54: Take the index i of the image with the highest resolution max =Max({S1, S2, ... S j}), then the Z-axis coordinate of the image with the highest definition is

[0043] Step 55: Control the Z axis to move to Retake an image and the image will be a high-definition image with clear focus.

[0044] Preferably, step 5 further includes the following steps:

[0045] Manual review: Manual review of the results of automatic detection, re-evaluation of missed or misjudged workpieces, and update of the results to the database module.

[0046] An automatic optical inspection system, comprising:

[0047] The image acquisition module realizes the control and image acquisition of multiple cameras, which includes: a large field of view camera and a high-resolution camera. The large field of view camera is used for positioning the detection object, and the high-resolution camera is used for high-precision detection of the detection object. The high-resolution camera, the large field of view camera and the light source module are installed on the Z-axis lifting mechanism;

[0048] Image processing module, which realizes asynchronous processing of images and outputs detection results to the database module;

[0049] The motion control module realizes the control of each motion axis, which includes an XY motion platform and a Z-axis lifting platform. The clamping unit is installed on the XY motion platform, and the XY motion platform drives the detection object to realize planar motion; the Z-axis lifting platform includes the Z-axis lifting mechanism;

[0050] An auto-focus module realizes automatic focusing of the image of the detection object;

[0051] Programming module, used to set detection algorithm parameters and create program templates;

[0052] Automatic calibration module, used to calibrate the image coordinate system and motion coordinate system;

[0053] The clamping unit is used to clamp the material box.

[0054] Preferably, the light source module realizes parameter setting and light switching control of the light source, including: a coaxial light source for a wide-view camera and a combined light source;

[0055] The coaxial light source of the wide field of view camera is installed below the wide field of view camera to provide lighting for the wide field of view camera;

[0056] The combined light source includes a combined coaxial light source and a combined ring light source, which are installed below the high-resolution camera. The combined coaxial light source provides bright field illumination for the high-resolution camera, and the combined ring light source provides dark field illumination.

[0057] Compared with the prior art, the advantages of the present invention are:

[0058] 1. This method has the advantages of high precision and high speed: including loading, global positioning, coordinate transformation, setting program templates and automatic detection. Therefore, you only need to click on the detection object on the global image, and the XY platform can move the corresponding detection object to the bottom of the high-resolution camera. Subsequently, the program module, autofocus module, and image processing module cooperate to realize automatic detection of all detection objects. Therefore, this method has the advantages of high precision and high speed.

[0059] 2. It is suitable for tiny inspection objects that cannot be positioned on the conveyor. It solves the problem that the high-precision inspection method for tiny parts in the prior art, such as bare chips, still relies on manual inspection using tools such as microscopes, which has low inspection efficiency, high labor intensity, and frequent missed inspections caused by eye fatigue. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A framework diagram for automatic optical inspection;

[0061] Figure 2 It is a structural diagram of automatic optical inspection;

[0062] Figure 3A schematic diagram of the material box and the arrangement of its parts taken by the wide-view camera 8;

[0063] Figure 4 A flowchart of an automatic optical detection method in the present invention;

[0064] Figure 5 It is a schematic diagram of the global positioning result;

[0065] Figure 6 A schematic diagram of a program template;

[0066] Figure 7 Schematic diagram of detection algorithm parameters.

[0067] 1. XY motion platform; 2. Clamping unit; 3. Combined ring light source; 4. Combined coaxial light source; 5. Coaxial light source for wide field of view camera; 6. High-resolution camera; 7. Z-axis lifting platform; 8. Wide field of view camera. DETAILED DESCRIPTION

[0068] The automatic optical detection method and system of the present invention will be described in more detail below in conjunction with the schematic diagram, wherein the preferred embodiment of the present invention is shown, and it should be understood that those skilled in the art can modify the present invention described herein and still achieve the beneficial effects of the present invention. Therefore, the following description should be understood as being widely known to those skilled in the art and not as a limitation of the present invention.

[0069] like Figures 1-2 , an automatic optical inspection system, comprising:

[0070] The image acquisition module realizes the control and image acquisition of multiple cameras, and includes: a large field of view camera 8 and a high-resolution camera 6.

[0071] The large-viewing-field camera 8 is used to detect the positioning of the object, and the high-resolution camera 6 is used to detect the object with high precision; the high-resolution camera 6, the large-viewing-field camera 8 and the light source module are installed on the Z-axis lifting mechanism to realize vertical movement.

[0072] Control the two cameras separately, set parameters such as camera exposure and frame rate, and realize image acquisition of the two cameras.

[0073] The image processing module realizes asynchronous processing of images and outputs the detection results to the database module. Specifically, the image acquisition module puts the images into a sequential cache, and the image processing module then obtains them from the cache in sequence to realize asynchronous processing.

[0074] Specifically, the image processing module adopts multi-threading, based on the producer-consumer model. The image acquisition module runs on the producer thread and sends the acquired image to the message queue. The image processing module runs on the consumer thread and reads the image from the message queue and performs asynchronous processing.

[0075] The motion control module realizes the control of each motion axis, which includes an XY motion platform 1 and a Z-axis lifting platform 7. The clamping unit 2 is installed on the XY motion platform 1. The XY motion platform 1 drives the detection object to realize planar motion; the Z-axis lifting platform 7 includes a Z-axis lifting mechanism;

[0076] The auto-focus module realizes the automatic focusing of the detection object image.

[0077] Programming module, used to set detection algorithm parameters and create program templates;

[0078] The automatic calibration module is used to calibrate the image coordinate system and the motion coordinate system.

[0079] The clamping unit 2 is used to clamp the material box. A Mark point is fixed on the clamping unit 2 for hand-eye calibration of the image acquisition system and the XY motion platform 1.

[0080] The light source module realizes the parameter setting and light switching control of the light source, including: a coaxial light source 5 for a large-viewing-field camera and a combined light source.

[0081] The wide-view-field camera coaxial light source 5 is installed below the wide-view-field camera 8 to provide lighting for the wide-view-field camera 8 .

[0082] The combined light source includes a combined coaxial light source 4 and a combined annular light source 3, which are installed below a high-resolution camera 6. The combined coaxial light source provides bright field illumination for the high-resolution camera, and the combined annular light source provides dark field illumination.

[0083] The parameter module is used to make the autofocus module and set the focus parameters of the autofocus module.

[0084] like Figure 4 , an automatic optical detection method, comprising the following steps:

[0085] Step 1: Loading.

[0086] Place the material box containing the object to be tested into the clamping unit 2, which automatically clamps the material box; place the test object in the material box in order, such as Figure 3 As shown; the clamping unit 2 is installed on the XY motion platform 1, driving the detection object to achieve planar motion;

[0087] Step 2: Global positioning, such as Figure 5 shown.

[0088] Control the XY platform to move the material box to the bottom of the wide-view camera 8, take a global picture of the material box, execute the global positioning algorithm, and obtain the number n of detection objects in the material box and the central image coordinates (x i ,y i ), i∈(1,n);

[0089] The steps of the global positioning algorithm are as follows:

[0090] Step 2A: De-noising and histogram equalization of the acquired global image to enhance the image;

[0091] Step 2B: Perform threshold segmentation and binarization on the global image to segment the detection object area;

[0092] Step 2C: Extract the edge of the binary image and segment the contour of the detected object;

[0093] Step 2D: Fit the object contour and find the center point, which is the coordinate of the center point of the detected object

[0094] Step 3: Coordinate transformation to obtain the center image coordinates (x i ,y i ) coincides with the center of the high-resolution camera image (x i ',y i ').

[0095] Specific:

[0096] Step 3A: Obtain the corresponding relationship P between the image pixel distance and the physical distance in the X direction x , specifically including the following steps:

[0097] Control the XY motion platform 1 to move the Mark point to below the wide-view camera 8, capture a global image, and record the image coordinates (x1, y1) of the Mark point in the global image; wherein the Mark point is an arbitrary point on the material box;

[0098] Control the XY platform to move a distance l1 in the X direction, and record the image coordinates (x2, y1) of the Mark point in the global image after the movement;

[0099] The corresponding relationship between the image pixel distance and the physical distance in the X direction can be obtained:

[0100] Step 3B: Obtain the corresponding relationship P between the image pixel distance and the physical distance in the Y direction y , specifically including the following steps:

[0101] Control the XY motion platform 1 to move the Mark point to below the wide-view camera 8, capture a global image, and record the image coordinates (x1, y1) of the Mark point in the global image;

[0102] Control the XY platform to move a distance l2 in the Y direction, and record the image coordinates (x1, y2) of the Mark point in the global image after the movement;

[0103] The corresponding relationship between the image pixel distance and the physical distance in the Y direction can be obtained:

[0104] Step 3C: First, control the XY motion platform 1 to move the clamping unit 2 to the position directly below the wide-view camera 8, and record the coordinates (x c ,y c );

[0105] Then, move the clamping unit 2 to the bottom of the high-resolution camera 6 so that the Mark point is located at the center of the high-resolution camera image, and record the physical coordinates (x' c ,y' c );

[0106] Step 3D: For each detected object in step 2, the central image coordinate (x i ,y i ), i∈(1,n), we can get its corresponding physical coordinates (x i ',y i '), i∈(1,n), where:

[0107]

[0108] Therefore, after coordinate conversion, click the detection object on the global image (change the center image coordinates (x i ,y i ) is sent to the control system), the XY platform can move the corresponding detection object to the bottom of the high-resolution camera 6. Among them, the coordinates of the XY platform and the physical coordinates of the detection object share a coordinate system, namely the motion coordinate system.

[0109] Step 4: Manually select a qualified inspection object and move it under the high-resolution camera 6.

[0110] Step 5: Execute autofocus to obtain a high-definition image of the detection object. The programming module sets the detection algorithm parameters (such as Figure 7 ), and save the program template, such as Figure 6 shown.

[0111] Specific:

[0112] The programming module specifically performs the following steps:

[0113] Step 5A: Select the combined light source type (coaxial light source or ring light source) of the high-resolution camera and set the brightness of the light source used. The brightness range is 0 to 255.

[0114] Step 5B, set the detection area, set the area to be detected and the area to be ignored in the detection object, and automatically select the corresponding area by setting the threshold, or intercept the image with the mouse and select the intercepted area. ;

[0115] Step 5C, set the detection sensitivity. The detection sensitivity can be used to set the detection accuracy of the entire system. The smaller the sensitivity, the smaller the minimum defect that can be detected and the higher the detection accuracy. The larger the sensitivity, the larger the minimum defect that can be detected and the lower the detection accuracy.

[0116] Step 5D, save the program module.

[0117] The automatic focusing in step 5 and step 6 specifically includes the following steps:

[0118] Step 51, move the detection object to below the high-resolution camera 6, and turn on the camera to take pictures in real time;

[0119] Step 52, setting a focus distance Δ, controlling the Z-axis lifting mechanism to move a distance Δ in the negative direction, and then controlling the Z-axis lifting mechanism to move a distance 2Δ in the positive direction;

[0120] During the movement, high-resolution camera images are collected in real time to obtain an image set {P1, P2, ...P j}, and its corresponding Z-axis coordinate set {z1, z2, ... z j};

[0121] Where j is the total number of images collected in real time; P j is the jth image;

[0122] Step 53: for the image set {P1, P2, ... P j}, and obtain the definition set {S1, S2, ... S j};

[0123] Among them, S j is the clarity of the jth image;

[0124] Step 54: Take the index i of the image with the highest resolution max =Max({S1, S2, ... S j}), then the Z-axis coordinate of the image with the highest definition is

[0125] Step 55, control the Z-axis to move to Take another image. At this time, the image is a high-definition image with clear focus.

[0126] Step 6, automatic detection.

[0127] Step 6A, according to the physical coordinates (x i ', y i ') output in Step 3D, control the XY platform to move the i-th detection object to below the high-resolution camera 6;

[0128] Step 6B, the automatic focusing module outputs a high-definition image of the current detection object and sends it to the message queue through the image acquisition module;

[0129] Step 6C, the image processing module obtains the image from the message queue, performs detection according to the current program, outputs the detection result, and saves the detection result to the database module; record i = i + 1.

[0130] Among them, the detection result includes the chip number, qualified / unqualified, result description, the program template used, and the detection time.

[0131] Step 6D: If i < n, repeat Steps 6A to 6C until i = n, that is, all detection objects are detected.

[0132] Step 7, manual review. Manually review the results of the automatic detection, re-determine the missed or misjudged workpieces, and update the results to the database module.

[0133] The above is only the preferred embodiment of the present invention and does not impose any limitation on the present invention. Any person skilled in the art within the technical field of the present invention, without departing from the technical solution of the present invention, makes any form of equivalent substitution or modification and other changes to the technical solution and technical content disclosed by the present invention, all of which belong to the content of the technical solution of the present invention and still fall within the protection scope of the present invention.

Claims

1. An automatic optical detection method, characterized in that: It includes the following steps: Step 1: Loading, which specifically includes the following steps: Put the cartridge containing the object to be detected into the clamping unit, and the clamping unit automatically clamps the cartridge; the object to be detected is placed in the cartridge; the clamping unit is installed on the XY platform and drives the object to be detected to achieve planar movement; Step 2: Global positioning, which specifically includes the following steps: Control the XY platform to move the material box to the bottom of the wide-view camera, take a global picture of the material box, execute the global positioning algorithm, and obtain the number n of detection objects in the material box and the central image coordinates (x i ,y i ), i∈(1,n); Step 3: Coordinate transformation to obtain the center image coordinates (x i ,y i ) coincides with the center of the high-resolution camera image (x i ′ ,y i ′ ), specifically comprising the following steps: Step 3A: Obtain the corresponding relationship P between the image pixel distance and the physical distance in the X direction x , specifically including the following steps: Control the XY platform to move the Mark point under the large field of view camera, take a global image, and record the image coordinates (x1, y1) of the Mark point in the global image; wherein, the Mark point is an arbitrary point on the cartridge; Control the XY platform to move a distance l1 along the X direction, and record the image coordinates (x2, y1) of the Mark point in the global image after the movement; The corresponding relationship between the image pixel distance and the physical distance in the X direction can be obtained: Step 3B: Obtain the corresponding relationship P between the image pixel distance and the physical distance in the Y direction y , specifically including the following steps: Control the XY platform to move the Mark point under the large field of view camera, take a global image, and record the image coordinates (x1, y1) of the Mark point in the global image; Control the XY platform to move a distance l2 along the Y direction, and record the image coordinates (x1, y2) of the Mark point in the global image after the movement; The corresponding relationship between the image pixel distance and the physical distance in the Y direction can be obtained: Step 3C: First, control the XY platform to move the clamping unit to the position directly below the wide-view camera and record the coordinates of the Mark point in the global image (x c ,y c ); Then, move the clamping unit to the bottom of the high-resolution camera so that the Mark point is located at the center of the high-resolution camera image and record the physical coordinates (x c ′ ,y c ′ ); Step 3D: For each detected object in step 2, the central image coordinate (x i ,y i ), i∈(1,n), we can get its corresponding physical coordinates (x i ′ ,y i ′ ), i∈(1,n), where: Step 4: Manually select a qualified object to be detected and move it under the high-resolution camera; Step 5: The automatic focusing module performs automatic focusing to obtain a high-definition image of the object to be detected. The programming module sets the detection algorithm parameters according to the high-definition image and saves the program template; Step 6: Automatic detection, which specifically includes the following steps: Step 6A: According to the physical coordinates (x i ′ ,y i ′ ), control the XY platform to move the i-th detection object to the bottom of the high-resolution camera; Step 6B: The automatic focusing module outputs the high-definition image of the current object to be detected and sends it to the message queue through the image acquisition module; Step 6C: The image processing module obtains the image from the message queue, performs detection according to the program template in the programming module, outputs the detection result, and saves the detection result to the database module; record i = i + 1; Step 6D: If i < n, repeat steps 6A to 6C until i = n, that is, all objects to be detected are detected.

2. The automatic optical inspection method according to claim 1, characterized in that: In step 5, the programming module specifically performs the following steps: Step 5A: Set the light source type and brightness; Step 5B: Set the ignored area; Step 5C: Set the detection sensitivity and detection threshold; Step 5D: Save the program template.

3. The automatic optical detection method according to claim 2, characterized in that: In step 5, the automatic focusing specifically includes the following steps: Step 51: Move the object to be detected under the high-resolution camera and turn on the real-time camera shooting; Step 52: Set the focusing distance Δ, control the Z-axis lifting mechanism to move a distance Δ in the negative direction, and then control the Z-axis lifting mechanism to move a distance 2Δ in the positive direction; During the movement, high-resolution camera images are collected in real time to obtain an image set {P1, P2, ...P j }, and its corresponding Z-axis coordinate set {z1, z2, ... z j }; Where j is the total number of images collected in real time; P j is the jth image; Step 53: for the image set {P1, P2, ... P j }, and obtain the definition set {S1, S2, ... S j }; Among them, S j is the clarity of the jth image; Step 54: Take the index i of the image with the highest resolution max =Max({S1, S2, ... S j }), then the Z-axis coordinate of the image with the highest definition is Step 55: Control the Z axis to move to Retake an image and the image will be a high-definition image with clear focus.

4. The automatic optical inspection method according to claim 1, characterized in that: After step 5, the following steps are also included: Manual review: Manually review the results of the automatic detection, re-determine the missed or misjudged workpieces, and update the results to the database module.

5. An automatic optical inspection system, used to implement the automatic optical inspection method according to any one of claims 1 to 4, characterized in that: It includes: An image acquisition module, which realizes the control and image acquisition of multiple cameras. It includes: a large field of view camera and a high-resolution camera. The large field of view camera is used for the positioning of the object to be detected, and the high-resolution camera is used for the high-precision detection of the object to be detected; the high-resolution camera, the large field of view camera and the light source module are installed on the Z-axis lifting mechanism; An image processing module, which realizes the asynchronous processing of images and outputs the detection results to the database module; The motion control module realizes the control of each motion axis, which includes an XY platform and a Z-axis lifting platform. The clamping unit is installed on the XY platform, and the XY platform drives the detection object to realize planar motion; the Z-axis lifting platform includes the Z-axis lifting mechanism; An auto-focus module realizes automatic focusing of the image of the detection object; Programming module, used to set detection algorithm parameters and create program templates; Automatic calibration module, used to calibrate the image coordinate system and motion coordinate system; The clamping unit is used to clamp the material box.

6. The automatic optical inspection system according to claim 5, characterized in that: The light source module realizes the parameter setting and light switching control of the light source, including: a coaxial light source for a large-viewing-field camera and a combined light source; The coaxial light source of the wide field of view camera is installed below the wide field of view camera to provide lighting for the wide field of view camera; The combined light source includes a combined coaxial light source and a combined ring light source, which are installed below the high-resolution camera. The combined coaxial light source provides bright field illumination for the high-resolution camera, and the combined ring light source provides dark field illumination.

Images