A fast automated optical detection system and method

Abstract

The present invention discloses a fast automated optical inspection system and method, wherein the method comprises the following steps: sequentially combining a global image and a local image to plan an inspection path; controlling the positions of a first image acquisition device and a second image acquisition device according to the planned inspection path, and acquiring image information of a workpiece to be inspected; and optically inspecting the workpiece to be inspected according to the image information. The present invention realizes fast positioning by combining a global image and a local image, and predicts the camera motion path, thereby realizing high-speed automatic path planning, indirectly improving the efficiency of optical inspection, and can be widely used in the field of optical inspection technology.

Description

Technical Field

[0001] The present invention relates to the field of optical detection technology, and in particular to a rapid automated optical detection system and method. Background Art

[0002] As industrial systems enter an era of automation and intelligence, people have higher and higher requirements for surface defect detection and dimension measurement of industrial products. At the same time, due to the increase in labor costs, the demand for automated testing is becoming more and more urgent. Especially for high-precision products, such as display panels, FPC cables, chips, etc., in the production process, they must not only meet the requirements of detection speed and accuracy, but also consider labor costs and automation costs. The demand for online automated testing equipment is becoming more and more urgent.

[0003] At present, the commonly used inspection methods for high-precision products are: (1) Ultrasonic inspection: Through the interaction between ultrasonic waves and the object to be inspected, the object to be inspected is subjected to microscopic inspection, geometric property measurement, inspection and characterization of changes in organizational structure and mechanical properties. This method has specific requirements for materials and is more difficult to inspect for objects with complex shapes or irregular shapes. Since the resolution is limited by the hardware of its wavelength, the spatial resolution of ultrasonic inspection is in the order of hundreds of microns. (2) Magnetic particle inspection. This method is a contact inspection method that uses magnetic powder to accumulate near the defect to detect defects on the surface of ferromagnetic materials. This technology can achieve 0.1 mm accuracy inspection and is effective for the inspection of defects and cracks on the surface of steel materials or workpieces, but it is only applicable to ferromagnetic materials, and the residual magnetism may have a certain impact on the surface of the object to be inspected. (3) Penetrant inspection. This method uses capillary phenomena to check for surface defects of materials. It is also a contact inspection method and has certain limitations. (4) Visual inspection: Through machine vision products, with a certain light source system, the image information of the object to be inspected is collected and transmitted to the image processing software for processing. Visual inspection methods have advantages that other inspection methods do not have: (a) It can realize non-contact measurement, reducing the possibility of damage to the object being inspected during the inspection process; (b) It is highly adaptable. Good inspection results can be achieved by selecting appropriate light source systems and image acquisition systems for different objects being inspected; (c) It is highly stable. Compared with traditional inspection methods such as manual inspection, visual inspection has strong stability and can achieve high-intensity continuous inspection under certain hardware conditions.

[0004] Although visual inspection has strong performance, there are still some challenges for product inspection with high precision requirements. Among them, slow inspection speed is one of the problems. The current high-precision AOI inspection system needs to be equipped with a high-precision image acquisition system, which has a slow image acquisition speed. At the same time, the physical size of the inspected object acquired in a single acquisition is small, resulting in a slow inspection speed for the entire system. At the same time, because the AOI system uses fixed-path scanning imaging, it has very strict requirements for workpiece positioning, and positioning the workpiece itself is a very time-consuming process.

[0005] Glossary:

[0006] AOI: Automated Optical Inspection, automatic optical inspection.

[0007] FPC: Flexible Printed Circuit, flexible circuit board. Summary of the invention

[0008] In order to solve the above technical problems, an object of the present invention is to provide a system and method capable of rapidly performing automated optical inspection.

[0009] The first technical solution adopted by the present invention is:

[0010] A rapid automated optical detection system comprises an image processing module, a motion platform, a first image acquisition device, a second image acquisition device and a third image acquisition device, wherein the image processing module is connected to the motion platform, the first image acquisition device, the second image acquisition device and the third image acquisition device respectively;

[0011] The first image acquisition device is used to acquire a local image of the workpiece to be inspected;

[0012] The second image acquisition device is used to acquire image information of the workpiece to be inspected;

[0013] The third image acquisition device is used to acquire a global image of the workpiece to be inspected;

[0014] The motion platform is used to control the first image acquisition device and the second image acquisition device to move together according to the received movement information;

[0015] The image processing module is used to plan the detection path by combining the global image and the local image in sequence, generate movement information according to the planned detection path and send it to the motion platform, and perform optical detection on the workpiece to be detected according to the image information.

[0016] Furthermore, the second image acquisition device adopts a high-resolution industrial camera.

[0017] Furthermore, the mobile platform includes an X-axis moving mechanism, a Y-axis moving mechanism and a Z-axis moving mechanism.

[0018] Furthermore, both the first image acquisition device and the second image acquisition device are provided with a light source.

[0019] The second technical solution adopted by the present invention is:

[0020] A rapid automated optical detection method comprises the following steps:

[0021] S1, combine the global image and the local image to plan the detection path;

[0022] S2, controlling the positions of the first image acquisition device and the second image acquisition device according to the planned detection path, and acquiring image information of the workpiece to be detected;

[0023] S3. Optically inspect the workpiece to be inspected based on the image information.

[0024] Further, the step S1 specifically includes the following steps:

[0025] After preprocessing the global image and the local image, the edge features of the global image and the local image are extracted respectively;

[0026] Acquire the current coordinate position of the second image acquisition device according to the edge feature;

[0027] The next coordinate position of the second image acquisition device is obtained according to the current coordinate position.

[0028] Furthermore, the step of controlling the positions of the first image acquisition device and the second image acquisition device according to the planned detection path in step S2 is specifically as follows:

[0029] After the second image acquisition device has acquired the image information of the current coordinate position, the positions of the first image acquisition device and the second image acquisition device are moved according to the next coordinate position;

[0030] Get the local image at the current coordinate position.

[0031] Further, the step S3 specifically includes the following steps:

[0032] After image information corresponding to each coordinate position is stitched together, a stitched image is obtained;

[0033] It is determined whether the stitched image meets the preset requirements. If so, feature extraction is performed on the stitched image, and whether the workpiece to be inspected has defects is analyzed based on the features; otherwise, the process returns to step S1.

[0034] Furthermore, an automatic focusing step is included, and the automatic focusing step is specifically:

[0035] Acquire a grayscale characteristic value of the image according to the image information, and determine whether the second image acquisition device is in a focused state according to the grayscale characteristic value;

[0036] When it is detected that the second image acquisition device is in a non-focusing state, the position of the second image acquisition device in the Z-axis direction is adjusted according to the focusing state to focus the second image acquisition device.

[0037] Furthermore, the method further includes a position correction step, wherein the position correction step is specifically:

[0038] The image information collected at the current position is combined with the preset coordinate position to analyze and determine whether a position deviation occurs, and when a position deviation is detected, the position of the second image collection device is adjusted according to the position deviation.

[0039] The beneficial effects of the present invention are as follows: the present invention realizes rapid positioning by combining global images with local images, and predicts the camera motion path, thereby realizing high-speed automatic path planning, and indirectly improving the efficiency of optical detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a structural block diagram of a rapid automated optical detection system of the present invention;

[0041] Figure 2 is a flowchart of the steps of a rapid automated optical detection method of the present invention;

[0042] Figure 3 is a schematic diagram of the mechanism of the motion platform in a specific implementation manner;

[0043] Figure 4 is a schematic diagram of the operation of the first image acquisition device and the third image acquisition device in a specific implementation manner;

[0044] Figure 5 yes Figure 4 A schematic diagram of a global image acquired by a third image acquisition device;

[0045] Figure 6 yes Figure 4 A schematic diagram of a local image captured by the first image capture device;

[0046] Figure 7 It is a schematic diagram of the operation of the first image acquisition device and the second image acquisition device in a specific implementation manner. DETAILED DESCRIPTION

[0047] like Figure 1 and Figure 3As shown, this embodiment provides a rapid automated optical detection system, including an image processing module, a motion platform 4, a first image acquisition device 1, a second image acquisition device 2 and a third image acquisition device 3, wherein the image processing module is connected to the motion platform 4, the first image acquisition device 1, the second image acquisition device 2 and the third image acquisition device 3 respectively;

[0048] The first image acquisition device 1 is used to acquire a local image of the workpiece 5 to be inspected;

[0049] The second image acquisition device 2 is used to acquire image information of the workpiece 5 to be inspected;

[0050] The third image acquisition device 3 is used to acquire a global image of the workpiece 5 to be inspected;

[0051] The motion platform 4 is used to control the first image acquisition device 1 and the second image acquisition device 2 to move together according to the received movement information;

[0052] The image processing module is used to plan the detection path by combining the global image and the local image in sequence, generate movement information according to the planned detection path and send it to the motion platform 4, and perform optical detection on the workpiece 5 to be detected according to the image information.

[0053] When the workpiece 5 to be inspected needs to be inspected, the workpiece 5 to be inspected is placed under the second image acquisition device 2, referring to Figure 3 , the first image acquisition device 1 and the second image acquisition device 2 are both installed on the motion platform 4, and move with the motion of the motion platform 4. The first image acquisition device 1 performs local imaging of the workpiece 5 to be detected, that is, only collects the local image of the workpiece 5 to be detected, and the third image acquisition device 3 is used to image the entire workpiece 5 to be detected, that is, collects the global image of the workpiece 5 to be detected, and the position of the third image acquisition device 3 is a fixed position. Among them, the first image acquisition device 1 and the third image acquisition device 3 can be implemented by ordinary industrial cameras, or by high-resolution industrial cameras. The second image acquisition device 2 is used to collect image information of the workpiece 5 to be detected, and the image information can be used for optical detection.

[0054] In the existing optical inspection technology, before the image of the workpiece to be measured is collected, it is necessary to position the workpiece to be measured, and then use a fixed path to scan and image the workpiece to be measured. Since the path is preset and fixed and cannot be flexibly changed, it is necessary to continuously scan the entire workpiece to be measured and obtain the scanned image of the entire workpiece to be measured. In fact, in the existing inspection, there is no need to scan the workpiece to be measured as a whole, so it takes too much time to scan the image, which seriously affects the efficiency of the inspection.

[0055] Therefore, this embodiment proposes a flexible scanning system, referring to Figure 4 In this system, the third image acquisition device 3 acquires the global image 6, and the first image acquisition device 1 acquires the local image 7, and matches the local image 7 to the global image 6, so as to quickly obtain the current position information and complete the positioning work; wherein, Figure 5 is a global image 6 of the workpiece 5 to be detected, Figure 6 It is a partial image 7 of the workpiece 5 to be inspected. The image processing module is provided with a preset program, which can predict the next coordinate position according to the current position information. Therefore, when the second image acquisition device 2 acquires the image information of the current position, it can quickly move to the next coordinate position, instead of adopting the existing method of slowly acquiring image information while moving. In addition, the system can flexibly change the detection path according to the design and obtain the corresponding image information without the need to scan and image the entire system, so as to achieve targeted results, greatly improve the efficiency of detection, indirectly reduce the production cost, and can be widely used in automated optical detection scenarios.

[0056] When the system performs the initial detection, visual calibration is required. The visual calibration can be planned based on the physical size information that can be collected by a single image, and the frequency of image collection can be planned. These technologies can be implemented using existing technologies.

[0057] As a further preferred embodiment, the second image acquisition device adopts a high-resolution industrial camera.

[0058] The second image acquisition device of the system may adopt a high-resolution industrial camera, or a high-definition camera or a camera with ordinary pixels. In the present embodiment, a high-resolution industrial camera is adopted to improve the accuracy of detection and better meet the high requirements of optical detection.

[0059] As a further preferred embodiment, the mobile platform includes an X-axis moving mechanism, a Y-axis moving mechanism and a Z-axis moving mechanism.

[0060] Since the high-precision AOI measurement accuracy requirements are all in the micron level, the resolution of its optical lens is also in the micron level. Therefore, due to the limitations of the current industrial camera imaging surface size and optical lens processing, the imaging depth of the telecentric lens used for high-precision photography is relatively small. This leads to the focus requirement for industrial cameras during photography. Even a small deviation will cause blurred imaging, and ultimately cause the detection algorithm to fail. In current technology, when scanning and imaging the workpiece to be inspected, the main method is to change the position of the X-axis and Y-axis. When blurred imaging occurs, no effective solution can be taken, which greatly affects the detection accuracy.

[0061] In this embodiment, when the second image acquisition device acquires image information of the workpiece to be detected, it can be moved on the XY detection plane through the X-axis moving mechanism and the Y-axis moving mechanism, and the height between the second image acquisition device and the workpiece to be detected is adjusted through the Z-axis moving mechanism to achieve automatic focusing, thereby greatly improving the detection accuracy.

[0062] Reference Figure 7 As a further preferred embodiment, both the first image acquisition device and the second image acquisition device are provided with a light source.

[0063] Reference Figure 7 The first image acquisition device includes a first lens 9, a first industrial camera 10 and a first light source 8, and the second image acquisition device includes a second lens 4, a second industrial camera 5 and a second light source 3.

[0064] Through the high-precision image acquisition system composed of the second lens 4, the second industrial camera 5 and the second light source 3, the reflected light of the workpiece 1 to be detected passes through the main optical axis 2 of the second lens, and the image acquisition of the workpiece 1 to be detected with micron-level accuracy can be achieved, and the image processing module performs algorithm processing to obtain information such as defects and dimensions of the workpiece to be detected. In the ordinary image acquisition system composed of the first lens 9, the first industrial camera 10 and the first light source 8, the reflected light of the workpiece 1 to be detected passes through the main optical axis 7 of the first lens, and an image with a certain accuracy can be acquired. The image information is converted into position information through the image processing module, and the position of the image acquisition at the next moment is predicted through the algorithm. The first image acquisition device acquires the current local image as feedback information, which is used to adjust the position of the second image acquisition device, thereby realizing closed-loop feedback, improving the accuracy of position movement, and indirectly improving the accuracy of detection. Among them, by configuring the light source, it is more conducive to the second image acquisition device to focus and obtain image information.

[0065] like Figure 2 As shown, this embodiment also provides a rapid automated optical detection method, comprising the following steps:

[0066] A1. Combine the global image and the local image to plan the detection path;

[0067] A2. Control the positions of the first image acquisition device and the second image acquisition device according to the planned detection path, and acquire image information of the workpiece to be detected;

[0068] A3. Perform optical inspection on the workpiece to be inspected based on the image information.

[0069] The embodiment of the present invention combines the global image and the local image to plan the detection path. The image processing module flexibly controls the movement of the first image acquisition device and the second image acquisition device according to the planned detection path, and collects the image information of the workpiece to be detected through the second image acquisition device, thereby avoiding the use of traditional methods to perform overall imaging of the workpiece to be detected, thereby greatly improving the detection efficiency. In the process of detection path planning, the global image and the local image can be combined to plan the later multi-step coordinate positions. For example, for the workpiece to be detected with a relatively simple shape (such as a rectangle or other regular shapes), the multi-step coordinate positions can be predicted. The next step coordinate position can also be planned in combination with the global image and the local image. For example, for some irregularly shaped workpieces to be detected, it is difficult to plan the multi-step coordinate positions, so the next step coordinate position is planned.

[0070] By combining global images with local images, rapid positioning is achieved and the camera motion path is predicted, thereby realizing high-speed automatic path planning and indirectly improving the efficiency of optical inspection.

[0071] Wherein, step A1 specifically includes steps A11 to A13:

[0072] A11, after preprocessing the global image and the local image, edge features of the global image and the local image are extracted respectively;

[0073] A12. Acquire the current coordinate position of the second image acquisition device according to the edge feature;

[0074] A13. Obtain the next coordinate position of the second image acquisition device according to the current coordinate position.

[0075] The preprocessing includes processing such as region segmentation, extraction and filtering, and feature extraction is performed on the preprocessed image, such as edge extraction, inflection point extraction, etc., followed by feature conversion, converting the image information into motion space position information, and obtaining the current coordinate position of the second image acquisition device. In combination with a preset algorithm (such as an image tracking prediction processing algorithm), the next coordinate position of the second image acquisition device is obtained according to the current coordinate position, and the next coordinate position can be used to control the rapid movement of the second image acquisition device.

[0076] Wherein, step A2 specifically includes steps A21 to A22:

[0077] A21, after the second image acquisition device completes acquiring image information of the current coordinate position, the positions of the first image acquisition device and the second image acquisition device are moved according to the next coordinate position;

[0078] A22. Obtain a local image of the current coordinate position, and collect image information of the workpiece to be inspected.

[0079] Each time the second image acquisition device moves to a position, it acquires image information of the workpiece to be inspected within the position range, and after the acquisition is completed, it quickly moves to the next position coordinate, rather than moving continuously, so that the image information of the workpiece to be inspected can be more effectively acquired. Among them, the local image of the current coordinate position can be used as feedback information, and closed-loop feedback can be performed to correct the predicted position information.

[0080] Wherein, step A3 specifically includes steps A31 to A32:

[0081] A31, stitching the image information corresponding to each coordinate position to obtain a stitched image;

[0082] A32. Determine whether the stitched image meets the preset requirements. If so, extract features from the stitched image and analyze whether the workpiece to be inspected has defects based on the features. Otherwise, return to step A1.

[0083] The image information collected at each coordinate position is stitched to obtain a stitched image. The stitched image can be an image of the entire workpiece to be detected, or a partial image of the workpiece to be detected, which is set according to specific detection requirements. In addition, in the face of irregular images, image acquisition can be performed more effectively. For example, if the shape of the workpiece to be detected is annular, according to traditional detection technology, it is necessary to collect the area in the middle of the ring, while in this embodiment, the acquisition can be carried out in a targeted manner, so there is no need to collect the area in the middle of the ring, which further improves the efficiency of the acquisition. The technology for stitching images can adopt existing technology, and there is no special requirement in this embodiment. The workpiece to be detected is optically inspected according to the stitched image to determine whether the workpiece to be detected has defects.

[0084] As a further preferred embodiment, an automatic focusing step is further included, and the automatic focusing step is specifically:

[0085] Acquire a grayscale characteristic value of the image according to the image information, and determine whether the second image acquisition device is in a focused state according to the grayscale characteristic value;

[0086] When it is detected that the second image acquisition device is in a non-focusing state, the position of the second image acquisition device in the Z-axis direction is adjusted according to the focusing state to focus the second image acquisition device.

[0087] The method of this embodiment also provides a high-speed automatic focusing function, which can achieve micron-precision adjustment of the camera, avoid blurry images, and indirectly improve detection efficiency and detection accuracy.

[0088] Specifically, automatic focusing can be achieved based on the grayscale value characteristics of the current image. When in the focused state, the mean square error value of the image is relatively large. When in the unfocused state, the mean square error value is relatively small. Rapid adjustment can be achieved based on the grayscale value of the image; in addition, the current focus state can also be judged based on the boundary information of the image. The present invention is not limited to the above two methods for achieving rapid focusing of the image in the Z direction.

[0089] As a further preferred embodiment, the method further includes a position correction step, wherein the position correction step is specifically:

[0090] The image information collected at the current position is combined with the preset coordinate position to analyze and determine whether a position deviation occurs, and when a position deviation is detected, the position of the second image collection device is adjusted according to the position deviation.

[0091] First, the image processing module inputs the local image obtained by the first image acquisition device into the global image obtained by the third image acquisition device, determines the preliminary position of the current second image acquisition device and the workpiece to be detected, and predicts the next coordinate position of the motion control system at the same time. Through this method, dynamic path planning can be achieved without the need for traditional continuous point-by-point scanning. At the same time, after the second image acquisition device is moved to the next coordinate point, the image captured by the second image acquisition device is compared with the set position to generate a position deviation. This feedback is a negative feedback system, and accurate positioning of the position is achieved through PD control. High-speed detection can be achieved through the combination of feedforward and feedback. Among them, the preset coordinate position is the next step coordinate position predicted in the previous step.

[0092] In summary, the optical detection system and method of the present invention have the following beneficial effects compared with the prior art:

[0093] (1) Automatic path planning. For industrial online detection, product materials are mostly placed randomly. Even if they are positioned by certain positioning fixtures, the positioning accuracy will not be very high. For high-precision industrial online detection, such as defect detection of lines in FPC cables and defect detection of glass boundaries, scanning and imaging are required during the detection process according to a certain path to achieve overall detection of the target. In addition, fixed path detection causes the system to usually adopt a manual method with a microscope for some special application scenarios. Due to manual judgment, the efficiency is low and the error is large. The present invention utilizes a combination of global images and local images to achieve rapid positioning, and predicts the camera motion path, thereby achieving high-speed automatic path planning. And through the optimization of the trajectory through closed-loop feedback, the difficulty of positioning the workpiece is avoided, the efficiency and quality of detection are greatly improved, and the requirements of rapid detection and high-precision detection are well met.

[0094] (2) Autofocus. Due to the limitation of lens design, the depth of field of high-precision imaging telecentric lenses is very small, so the control of imaging distance is very critical. The present invention realizes rapid adjustment of the distance between the camera and the target to be measured by analyzing the current image features, thereby realizing an autofocus method to compensate for the problem of insufficient depth of field and realize high-speed fast-reading focus. That is, the features of the image formed by the second image acquisition device and the local image features are used to realize distance prediction and feedback control of the focusing process of the detection camera.

[0095] (3) Hardware acceleration: The current high-precision AOI inspection system has very strict requirements on workpiece positioning. Therefore, the system can only operate in a closed space. In order to achieve high-precision positioning, the entire system is huge and cannot be directly compatible with the production line. In the present invention, both scanning automatic path planning and imaging automatic focusing require high-speed signal processing. The present invention parallelizes this part of the processing, realizes high-speed processing by FPGA, and improves the inspection efficiency to a certain extent through hardware acceleration.

[0096] (4) Fast algorithm. Design and use fast processing algorithms to reduce the single detection time. At the same time, in terms of control principle, the present invention adopts a method of common feedforward and feedback to speed up the process of automatic path planning and imaging focusing.

[0097] (5) The automated optical detection method of the present invention does not require scanning and imaging of the entire system, but only detects the portion that needs to be imaged based on the image features, thereby reducing the scanning and imaging time and achieving high-speed detection.

[0098] In general, the fast automated optical inspection system and method of the present invention realizes high-speed and high-precision optical inspection at low cost, and its spatial resolution can reach the micron level. The resolution and detection speed of the image are greatly improved. At the same time, the system uses automated path planning to automatically adjust the planned path during detection, which improves the detection accuracy compared to traditional fixed path planning. The present invention can be applied to module-based automated path planning, and can also be applied to image acquisition methods based on robotic arms and other 3D path planning. The spatial resolution and temporal resolution of image acquisition are greatly improved. Specifically, the present invention can be applied to detection scenarios such as FPC line detection, glass panel quality detection, PCB visual quality detection, and mobile phone screen detection.

[0099] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A rapid automated optical detection system, characterized in that: The device comprises an image processing module, a motion platform, a first image acquisition device, a second image acquisition device and a third image acquisition device, wherein the image processing module is connected to the motion platform, the first image acquisition device, the second image acquisition device and the third image acquisition device respectively; wherein the first image acquisition device and the second image acquisition device are arranged on the motion platform, and the third image acquisition device is arranged at a fixed position which is not located on the motion platform; The first image acquisition device is used to acquire a local image of the workpiece to be inspected; The second image acquisition device is used to acquire image information of the workpiece to be inspected; The third image acquisition device is used to acquire a global image of the workpiece to be inspected; The motion platform is used to control the first image acquisition device and the second image acquisition device to move together according to the received movement information; The image processing module is used to sequentially combine the global image and the local image to plan the detection path, generate movement information according to the planned detection path and send it to the motion platform, and perform optical detection on the workpiece to be detected according to the image information; The image processing module is also used to extract edge features of the global image and the local image respectively after preprocessing the global image and the local image; Converting the edge features into motion space position information; Acquire the current coordinate position of the second image acquisition device according to the motion space position information; In combination with a preset algorithm, the next coordinate position of the second image acquisition device is obtained according to the current coordinate position; the image processing module is also used to move the positions of the first image acquisition device and the second image acquisition device according to the next coordinate position after the second image acquisition device has acquired the image information of the current coordinate position; Get the local image at the current coordinate position; The second image acquisition device acquires image information of the workpiece to be inspected each time it moves to a position; After each acquisition of the image information is completed, the second image acquisition device is moved to the next coordinate position; and the predicted next coordinate position of the second image acquisition device is corrected based on the local image.

2. A rapid automated optical inspection system according to claim 1, characterized in that: The second image acquisition device adopts a high-resolution industrial camera.

3. A rapid automated optical inspection system according to claim 1, characterized in that: The motion platform includes an X-axis moving mechanism, a Y-axis moving mechanism and a Z-axis moving mechanism.

4. A rapid automated optical inspection system according to claim 1, characterized in that: The first image acquisition device and the second image acquisition device are both provided with a light source.

5. A rapid automated optical detection method, characterized in that: The following steps are involved: S1, collecting a local image of the workpiece to be detected by a first image acquisition device, collecting a global image of the workpiece to be detected by a third image acquisition device, and combining the global image and the local image in sequence to plan a detection path; After preprocessing the global image and the local image, the edge features of the global image and the local image are extracted respectively; Converting the edge features into motion space position information; Acquire the current coordinate position of the second image acquisition device according to the motion space position information; In combination with a preset algorithm, the next coordinate position of the second image acquisition device is obtained according to the current coordinate position; S2, controlling the positions of the first image acquisition device and the second image acquisition device according to the planned detection path, and moving the positions of the first image acquisition device and the second image acquisition device according to the next coordinate position after the second image acquisition device completes acquiring the image information of the current coordinate position; Get the local image at the current coordinate position; The second image acquisition device acquires image information of the workpiece to be inspected each time it moves to a position; After completing the acquisition of the image information each time, moving the second image acquisition device to the next coordinate position; Correcting the predicted next coordinate position of the second image acquisition device based on the local image; S3. Perform optical inspection on the workpiece to be inspected according to the image information.

6. A rapid automated optical detection method as claimed in claim 5, characterized in that: The step S3 specifically comprises the following steps: After image information corresponding to each coordinate position is stitched together, a stitched image is obtained; It is determined whether the stitched image meets the preset requirements. If so, feature extraction is performed on the stitched image, and whether the workpiece to be inspected has defects is analyzed based on the features; otherwise, the process returns to step S1.

7. A rapid automated optical detection method as claimed in claim 5, characterized in that: The method further includes an automatic focusing step, wherein the automatic focusing step is specifically: Acquire a grayscale characteristic value of the image according to the image information, and determine whether the second image acquisition device is in a focused state according to the grayscale characteristic value; When it is detected that the second image acquisition device is in a non-focusing state, the position of the second image acquisition device in the Z-axis direction is adjusted according to the focusing state to focus the second image acquisition device.

8. A rapid automated optical detection method as claimed in claim 5, characterized in that: The step of position correction is also included, and the step of position correction is specifically as follows: The image information collected at the current position is combined with the preset coordinate position to analyze and determine whether a position deviation occurs, and when a position deviation is detected, the position of the second image collection device is adjusted according to the position deviation.

Images