Product quality detection method, system and equipment applied to endoscope camera module

Abstract

The invention discloses a product quality detection method, system and equipment applied to an endoscope camera module, and relates to the technical field of camera module quality detection. The method comprises the following steps: clamping a target endoscope camera module group to a camera module group test tool, electrifying the target endoscope camera module group, and clamping an image test card to a position right above the target endoscope camera module group; the method comprises the following steps: controlling a camera module test tool to perform imaging detection operation on a target endoscope camera module through an upper computer, and receiving camera module detection imaging information based on a decoding deck; performing signal filtering to obtain an available camera module detection image, and transmitting the available camera module detection image to an upper computer; performing imaging defect analysis to obtain defect feature data of the camera module; and performing quality evaluation and judgment on the defect feature data of the camera module to obtain a quality detection result of the camera module. The technical problem of low detection efficiency of the endoscope camera module in the prior art is solved, and the technical effect of improving the detection efficiency is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of camera module quality inspection, and in particular to a product quality inspection method, system and equipment applied to an endoscope camera module. Background Art

[0002] With the widespread application of endoscope technology, the quality of endoscope camera modules, as an important component of endoscope equipment, directly affects the clarity and accuracy of endoscopic imaging. However, in the prior art, the quality inspection method for endoscope camera modules mainly relies on manual inspection or simple automated inspection methods. This inspection method has the following problems: on the one hand, manual inspection is inefficient and easily affected by subjective factors, resulting in instability of the inspection results; on the other hand, the existing automated inspection system can usually only identify more obvious defects, and has insufficient detection capabilities for complex image quality problems (such as blur, distortion, color deviation, etc.). Summary of the invention

[0003] The present application provides a product quality inspection method, system and equipment for an endoscope camera module, which solves the technical problem of low inspection efficiency of an endoscope camera module in the prior art.

[0004] In view of the above problems, the present application provides a product quality inspection method, system and equipment for endoscope camera modules.

[0005] In a first aspect of the present application, a product quality detection method for an endoscope camera module is provided, the method comprising: Obtain a camera module test fixture, clamp the target endoscope camera module onto the camera module test fixture in a preset direction, power on the target endoscope camera module, and clamp the image test card directly above the target endoscope camera module; control the camera module test fixture through a host computer to perform imaging detection operations on the target endoscope camera module, and receive camera module detection imaging information based on a decoder board; perform signal filtering on the camera module detection imaging information based on the decoder board to obtain a usable camera module detection image, and transmit the usable camera module detection image signal to the host computer; obtain an image defect detection algorithm network through the host computer, perform imaging defect analysis on the usable camera module detection image based on the image defect detection algorithm network, and obtain camera module defect feature data; build a product quality determiner, perform quality assessment and determination on the camera module defect feature data based on the product quality determiner, and obtain a camera module quality detection result.

[0006] A second aspect of the present application provides a product quality detection system for an endoscope camera module, the system comprising: Installation module: obtain the camera module test fixture, clamp the target endoscope camera module to the camera module test fixture according to the preset direction, and power on the target endoscope camera module, and at the same time clamp the image test card just above the target endoscope camera module; Imaging detection module: control the camera module test fixture to perform imaging detection operations on the target endoscope camera module through the host computer, and receive the imaging information detected by the camera module based on the decoder board; Signal filtering module: filter the imaging information detected by the camera module based on the decoder board Signal filtering is used to obtain an available camera module detection image, and the available camera module detection image signal is transmitted to the host computer; imaging defect analysis module: the image defect detection algorithm network is obtained through the host computer, and imaging defect analysis is performed on the available camera module detection image based on the image defect detection algorithm network to obtain camera module defect feature data; quality assessment module: a product quality determiner is built, and quality assessment and determination is performed on the camera module defect feature data based on the product quality determiner to obtain camera module quality detection results.

[0007] The third aspect of the present application provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the product quality inspection method for an endoscope camera module provided in the present application.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, obtain the camera module test fixture, clamp the target endoscope camera module to the camera module test fixture in a preset direction, power on the target endoscope camera module, and clamp the image test card directly above the target endoscope camera module. Then, the host computer controls the camera module test fixture to perform imaging detection operations on the target endoscope camera module, and receives the camera module detection imaging information based on the decoder board. Further, based on the decoder board, the camera module detection imaging information is signal filtered to obtain an available camera module detection image, and the available camera module detection image signal is transmitted to the host computer. Then, the image defect detection algorithm network is obtained through the host computer, and the imaging defect analysis of the available camera module detection image is performed based on the image defect detection algorithm network to obtain the camera module defect feature data. Finally, a product quality determiner is built, and the camera module defect feature data is quality evaluated and determined based on the product quality determiner to obtain the camera module quality detection result. The invention solves the technical problem of low detection efficiency of the endoscope camera module in the prior art and achieves the technical effect of improving the detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic flow chart of a product quality inspection method for an endoscope camera module provided in an embodiment of the present application; Figure 2 A first structural schematic diagram of a camera module testing tool provided in an embodiment of the present application; Figure 3 A second structural schematic diagram of the camera module testing tooling provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of a module test seat in a camera module test tooling provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a product quality inspection system for an endoscope camera module provided in an embodiment of the present application; Figure 6 This is a schematic diagram of the structure of an exemplary electronic device of the present application.

[0011] Explanation of the accompanying drawings: installation module 11, imaging detection module 12, signal filtering module 13, imaging defect analysis module 14, quality assessment module 15, processor 21, memory 22, input device 23, output device 24, module test seat 31, rotating platform 32, X-axis adjustment knob 33, Y-axis adjustment knob 34, Z-axis adjustment handwheel 35 and first interface 36, second interface 37 and third interface 38. DETAILED DESCRIPTION

[0012] The present application solves the technical problem of low detection efficiency of endoscope camera modules in the prior art by providing a product quality detection method, system and equipment applied to endoscope camera modules.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or are inherent to these processes, methods, products or devices.

[0015] Embodiment 1, as Figure 1 As shown, the present application provides a product quality detection method applied to an endoscope camera module, wherein the method comprises: Obtain the camera module test fixture, clamp the target endoscope camera module onto the camera module test fixture in a preset direction, power on the target endoscope camera module, and clamp the image test card directly above the target endoscope camera module.

[0016] The camera module test fixture is a device used to detect the camera module, which is used to fix the camera module and provide a stable test environment. First, the target endoscope camera module to be tested is installed on the camera module test fixture according to the preset direction, wherein the preset direction refers to a correct installation angle or position specified according to the working state and design characteristics of the camera module to ensure the accuracy and consistency of the test data; then, the target endoscope camera module is powered on to put it into working state; at the same time, the image test card is installed directly above the target endoscope camera module and kept parallel to the optical axis of the camera module to ensure that the image captured by the module can completely cover the content of the test card, wherein the image test card is a board with a standardized pattern, which is used to detect the imaging quality of the camera module. Through this standardized clamping and test preparation, detection errors caused by installation deviations or irregular operations can be effectively avoided.

[0017] Furthermore, if Figure 2 , Figure 3 , Figure 4 As shown, the camera module testing tooling includes a module testing seat 31, a rotating platform 32, an X-axis adjustment knob 33, a Y-axis adjustment knob 34, a Z-axis adjustment handwheel 35, and a first interface 36, a second interface 37 and a third interface 38.

[0018] The camera module test tooling includes a module test seat 31, a rotating platform 32, an X-axis adjustment knob 33, a Y-axis adjustment knob 34, a Z-axis adjustment handwheel 35, and a first interface 36, a second interface 37, and a third interface 38; the module test seat 31 is used to fix the target endoscope camera module to ensure its stability and reliability during the test process; the rotating platform 32 supports multi-axis angle adjustment of the camera module to facilitate alignment with the test target; the X-axis adjustment knob 33 and the Y-axis adjustment knob 34 respectively realize precise fine-tuning of the camera module in the left and right (X-axis direction) and up and down (Y-axis direction) to ensure alignment with the center position of the test card; the Z-axis adjustment handwheel 35 is used to adjust the distance between the camera module and the test card to ensure imaging clarity; the first interface 36 is a USB Type A interface for connecting external devices such as a power supply or a host computer; the second interface 37 is a Lemo interface (connected to the module test seat) for signal and power connection with the camera module; the third interface 38 is a USB Type C interface (connected to a computer) is used to transmit the test signal of the camera module to the host computer for processing and analysis.

[0019] Furthermore, the step of clamping the target endoscope camera module onto the camera module testing tooling according to a preset direction includes: The target endoscope camera module is clamped and fixed on the module test seat according to the preset direction, and the position of the target endoscope camera module is adjusted using the X-axis adjustment knob, Y-axis adjustment knob and Z-axis adjustment handwheel; based on the rotating platform, the angle of the target endoscope camera module is adjusted according to the camera module test scenario, and the first interface, the second interface and the third interface are connected to external devices in sequence.

[0020] Specifically, first, clamp and fix the target endoscope camera module to the module test seat in the preset direction to ensure that the position of the camera module is stable and reliable; then, use the X-axis adjustment knob, Y-axis adjustment knob and Z-axis adjustment handwheel to accurately adjust the position of the target endoscope camera module to ensure that the center of the camera module is aligned with the test card or other target position in the test scene; after completing the position adjustment, adjust the angle of the camera module based on the test scene requirements through the rotating platform to ensure that the viewing angle of the camera module matches the requirements of the test scene; at the same time, connect the first interface (USB Type A interface), the second interface (Lemo interface) and the third interface (USB Type C interface) to external devices in turn to realize the power supply, signal transmission and data processing support of the upper computer of the camera module. Through this clamping and adjustment process, the test accuracy and stability of the endoscope camera module are ensured, laying the foundation for subsequent quality inspection.

[0021] The camera module testing tooling is controlled by a host computer to perform imaging detection operations on the target endoscope camera module, and the camera module detection imaging information is received based on a decoding board.

[0022] The host computer is a computer or a main control device, which is the control core of the entire test process; the decoding board is used to receive and process the original imaging information transmitted by the camera module. The imaging data captured by the camera module (usually transmitted in the form of digital signals) will be passed to the decoding board through the interface.

[0023] The host computer controls the camera module test fixture to perform imaging detection operations on the target endoscope camera module, including starting the camera module, adjusting the position of the test fixture, and setting relevant detection parameters, so that the camera module performs imaging operations on the image test card or other test targets. At the same time, based on the decoder board receiving the detection imaging information transmitted by the camera module, the decoder board performs preliminary processing on the received imaging data, including signal decoding, correction and format conversion, to ensure the integrity and availability of the data. The processed imaging information is transmitted to the host computer to provide reliable basic data support for subsequent image defect analysis and camera module quality assessment.

[0024] Based on the decoding board, the imaging information detected by the camera module is filtered to obtain a usable camera module detection image, and the usable camera module detection image signal is transmitted to the host computer.

[0025] The decoder board extracts signals from the imaging information detected by the camera module, and applies a variety of filtering algorithms (such as low-pass filtering, Gaussian filtering or median filtering) to process the noise, interference and other factors that affect the imaging quality in the data, so as to remove high-frequency noise in the imaging signal, reduce background interference, and enhance the edge details of the image, thereby ensuring the clarity and accuracy of the signal. After filtering, the obtained image signal is converted into a usable camera module detection image, which has a good signal-to-noise ratio and high-quality imaging effect, and can reflect the real imaging performance of the camera module. Subsequently, the decoder board sends these usable detection image signals to the host computer through a standardized transmission interface (such as USB Type A or USB Type C) to ensure the integrity and stability of the signal.

[0026] Furthermore, the method of obtaining an image that can be detected by a camera module includes: Based on the decoding board, the camera module detection imaging information is subjected to signal type recognition and signal decoding conversion to obtain a camera module decoded detection image; the camera module decoded detection image is subjected to color channel conversion and signal enhancement to obtain a camera module enhanced detection image; the camera module enhanced detection image is subjected to noise recognition to obtain image multi-scale noise information; based on the image multi-scale noise information, the camera module enhanced detection image is filtered to obtain a usable camera module detection image.

[0027] Specifically, the camera module detection imaging information is identified by the signal type based on the decoding board, and the signal transmission format is analyzed to determine whether it is an analog signal or a digital signal. The signal is decoded and format converted based on the identification result, and the original imaging signal is converted into a standardized image data format, thereby obtaining a camera module decoded detection image; then, the decoded camera module detection image is converted into a color channel conversion operation, and the color space of the image is converted from the RGB format to a YCbCr or HSV format suitable for subsequent processing to optimize the color restoration effect of the image. At the same time, the image is enhanced by adjusting parameters such as brightness, contrast and color saturation to obtain the camera module enhanced detection image. image; next, the noise information in the camera module enhanced detection image is identified, and the multi-scale analysis method is used to extract multi-scale noise features such as high-frequency noise, low-frequency interference and random noise in the image to form multi-scale noise information of the image; finally, the camera module enhanced detection image is filtered based on the extracted multi-scale noise information of the image, and combined with algorithms such as Gaussian filtering, adaptive median filtering or bilateral filtering, denoising operations are gradually performed on noises of different types and scales, while retaining the key details and edge information in the image, and finally a usable camera module detection image with high signal-to-noise ratio and complete details is obtained, providing high-quality data support for subsequent image defect detection and camera module quality assessment.

[0028] Furthermore, the obtaining of the image detected by the camera module includes: Obtain a list of filtering algorithms, and match the type characteristics of the multi-scale noise information of the image with the filtering algorithm list to obtain a multi-scale noise matching algorithm set; perform filtering parameter analysis on the multi-scale noise information of the image based on the multi-scale noise matching algorithm set to obtain a multi-scale noise filter; use the multi-scale noise filter to filter the camera module enhanced detection image to obtain the available camera module detection image.

[0029] Specifically, a filtering algorithm list containing multiple filtering algorithms is obtained, and the filtering algorithm list contains filtering algorithms designed for different types of noise, such as Gaussian filtering, adaptive median filtering, bilateral filtering, etc., which are used to process multiple noise types such as high-frequency noise, low-frequency interference and random noise; based on the extracted multi-scale noise information of the image, the type characteristics of each noise (such as the spatial distribution, intensity and frequency range of the noise) are analyzed, and these characteristics are matched one by one with the algorithms in the filtering algorithm list, so as to generate a multi-scale noise matching algorithm set, in which each noise type corresponds to one or more suitable filtering algorithms; based on the multi-scale noise matching algorithm set, the multi-scale noise information of the image is filtered Wave parameter analysis, specifically including determining the key filtering parameters of each matching algorithm, such as filter window size, smoothing coefficient or edge retention strength, etc., to meet the processing requirements of different noise characteristics. Through this process, multiple multi-scale noise filters are generated, and each multi-scale noise filter is optimized for specific noise types and characteristics; the generated multi-scale noise filters are used to filter the camera module enhanced detection image layer by layer. During the processing, each multi-scale noise filter is only effective for its corresponding noise type, thereby achieving hierarchical removal of high-frequency noise, low-frequency interference and random noise, while trying to retain the key details and edge information in the image to avoid excessive smoothing or detail loss. Through the above process, a usable camera module detection image is finally obtained. The usable camera module detection image has a high signal-to-noise ratio and complete detail quality, providing highly reliable data support for subsequent defect detection and quality assessment.

[0030] An image defect detection algorithm network is obtained through the host computer, and imaging defect analysis is performed on the available camera module detection image based on the image defect detection algorithm network to obtain camera module defect feature data.

[0031] The host computer calls the image defect detection algorithm network from pre-loaded or external storage. The network is usually based on deep learning or machine learning technology, such as convolutional neural network (CNN) or target detection-based algorithm models (such as YOLO, Faster R-CNN), which is specifically used to analyze the imaging defects of the camera module. The host computer inputs the available camera module detection image transmitted from the decoding board into the image defect detection algorithm network. The image defect detection algorithm network performs feature extraction and pattern analysis on the available camera module detection image to identify possible imaging defects. After the analysis is completed, the camera module defect feature data is output. The camera module defect feature data includes defect type (such as blur, distortion or color reproduction abnormality, etc.), defect location (specific coordinates or area range of the defect in the image), and defect degree (severity of the defect).

[0032] Furthermore, the obtaining of the image defect detection algorithm network through the host computer includes: According to the application standard of the endoscope camera module, a camera module imaging defect label library is constructed; a camera module defect image data set is collected and acquired, and the camera module defect image data set is type-labeled based on the camera module imaging defect label library to obtain a camera module defect image sample set; image defect detection requirements are acquired, and according to the image defect detection requirements, an image detection algorithm network structure is determined; the camera module defect image sample set is trained and optimized using the image detection algorithm network structure to obtain an image defect detection algorithm network and store it in the host computer.

[0033] Specifically, according to the application standards of endoscope camera modules (such as industry standards or imaging quality requirements in specific application scenarios), a camera module imaging defect label library is constructed. The camera module imaging defect label library is used to define various imaging defect types that may exist in the camera module and their standardized descriptions, such as blur, color cast, distortion, dark corners, noise or bad pixels, and set standardized descriptions and classification rules for each defect type, including feature parameters, position markers and severity levels; a large number of image data sets containing camera module defects are obtained by collecting actual captured images, generating simulated defect images, etc.; based on the constructed camera module imaging defect label library, the defect image data set is type-labeled, and label information such as defect type, defect area coordinates and severity is added to each image, thereby forming a camera module defect image sample set to provide high-quality data support for subsequent algorithm training; according to the image defect detection requirements of the camera module, the actual application scenario requirements such as the defect type to be detected, the detection accuracy requirements and the detection speed are analyzed, and a suitable image detection algorithm network structure is selected or designed. The image detection algorithm network structure can adopt a mature deep learning model architecture (such as YOLO, Faster R-CNN or a custom model based on a convolutional neural network) or adjust and optimize it in combination with actual needs; after determining the network structure, the camera module defect image sample set is input into the network for multiple rounds of training and tuning. In the training stage, the network weights are optimized through large-scale image data in the sample set to improve the network's ability to recognize various imaging defects; in the tuning stage, the network's detection accuracy and robustness are further improved by adjusting the learning rate, optimizing the loss function, and introducing data enhancement technology, while reducing the false alarm rate and missed alarm rate. Finally, the trained and tuned image defect detection algorithm network is saved to the host computer as the core tool for subsequent camera module defect detection tasks. Through this process, the obtained algorithm network has the ability to accurately detect camera module imaging defects, can quickly identify the defect type in the image, locate the defect area, and evaluate the degree of the defect, thereby providing efficient and reliable technical support for the quality assessment of the endoscope camera module.

[0034] Furthermore, the obtaining of the camera module defect feature data includes: Based on the image defect detection algorithm network, the available camera module detection image is marked for defect identification to obtain a camera module defect area set and a camera module defect type set; edge segmentation and integral calculation are performed on the camera module defect area set to obtain a camera module defect area area set; an image positioning coordinate system is constructed, and the camera module defect area set is positioned in turn based on the image positioning coordinate system to obtain a camera module defect area position set; based on the camera module defect type set, the camera module defect area set and the camera module defect area position set, the camera module defect feature data is obtained.

[0035] Specifically, based on the image defect detection algorithm network, defect recognition marking is performed on the available camera module detection images. The image defect detection algorithm network accurately identifies the defect types in the image (such as blur, color cast, distortion, dark corners, noise, etc.) by analyzing the characteristics of the pixels in the image, and generates a corresponding marking area for each defect, forming a camera module defect area set and a camera module defect type set. The camera module defect area set defines the boundary range of all detected defects in the image, and the camera module defect type set classifies these defects; then, edge segmentation and integral calculation are performed on each defect area in the camera module defect area set. The purpose of edge segmentation is to extract the boundary line of each defect area through an algorithm (such as Canny edge detection or Sobel operator) to ensure the accurate outline of the defect area; the integral calculation obtains the defect area of ​​each defect by summing the area of ​​the pixel points in the defect area. Area set, the camera module defect area set contains the area sizes of all defect areas, which is used for subsequent quality assessment; then, an image positioning coordinate system is constructed, with the upper left corner of the image as the origin, and the two-dimensional coordinate system of the X-axis and the Y-axis is defined; based on the image positioning coordinate system, each defect area in the camera module defect area set is positioned, that is, the spatial position data such as the center point coordinates and the boundary vertex coordinates of the defect area are calculated to form a camera module defect area position set, which is used to clarify the specific distribution position of each defect in the image; the camera module defect type set, camera module defect area set and camera module defect area position set obtained above are combined to generate complete camera module defect feature data, which includes detailed information such as the type, area, and position of each defect, providing comprehensive data support for quality inspection of the camera module, and can be used for subsequent quality judgment and optimization analysis.

[0036] A product quality determiner is built, and quality assessment and determination is performed on the defect feature data of the camera module based on the product quality determiner to obtain a camera module quality inspection result.

[0037] The product quality determiner is a rule-based evaluation tool used to analyze defect feature data; the camera module defect feature data is input into the product quality determiner for quality evaluation and judgment, so as to determine whether the camera module meets the quality standards, and finally output the camera module quality inspection results.

[0038] Furthermore, the construction of the product quality determiner includes: Acquire product quality judgment factors, wherein the product quality judgment factors include a camera module defect type set, a defect region area, and a defect region position; construct a product quality judgment coordinate system according to the product quality judgment factors; divide the product quality judgment coordinate system into qualified product regions based on the product qualification standard of the target endoscope camera module to obtain qualified product region logic lines; and build the product quality determiner based on the product quality judgment coordinate system and the qualified product region logic lines.

[0039] Specifically, product quality judgment factors are obtained, which include the defect type set, defect area area and defect area position of the camera module. These factors respectively describe the nature, size and spatial distribution of the defects, and are the core indicators for quality judgment. According to the product quality judgment factors, a product quality judgment coordinate system is constructed, in which the defect type is used as a classification dimension to distinguish different defect categories (such as blur, color cast, distortion, etc.); the defect area is used as a size dimension to quantify the impact range of the defect; the defect area position is used as a spatial dimension to describe the distribution position of the defect in the image; through the combination of these three-dimensional data, a product quality judgment coordinate system is established that can comprehensively describe the defect characteristics of the camera module.

[0040] According to the product qualification standard of the target endoscope camera module, the product quality judgment coordinate system is divided into qualified product areas. Specifically, the defect types allowed in the qualification standard, the maximum threshold of the defect area area, and the restriction conditions of the defect location are analyzed, and the qualified product area logic line in the coordinate system is drawn using these standards to divide the coordinate system into qualified areas and unqualified areas. For example, some minor defects (such as edge noise) may fall into the qualified area, while serious defects (such as blur or color deviation in the center area of ​​the image) will fall into the unqualified area. Finally, based on the product quality judgment coordinate system and the qualified product area logic line, a product quality determiner is built to enable it to automatically analyze the defect feature data of the camera module. In specific implementation, the defect feature data (defect type, area and location) is mapped to the product quality judgment coordinate system, and the judgment result is quickly obtained by judging whether the data point is located in the qualified product area. At the same time, the determiner can output the judgment basis, including information such as the defect exceeding the standard type and the unqualified location, to achieve efficient evaluation and intuitive feedback on the product quality of the camera module.

[0041] In summary, the embodiments of the present application have at least the following technical effects: First, obtain the camera module test fixture, clamp the target endoscope camera module to the camera module test fixture in a preset direction, power on the target endoscope camera module, and clamp the image test card directly above the target endoscope camera module. Then, the host computer controls the camera module test fixture to perform imaging detection operations on the target endoscope camera module, and receives the camera module detection imaging information based on the decoder board. Further, based on the decoder board, the camera module detection imaging information is signal filtered to obtain an available camera module detection image, and the available camera module detection image signal is transmitted to the host computer. Then, the image defect detection algorithm network is obtained through the host computer, and the imaging defect analysis of the available camera module detection image is performed based on the image defect detection algorithm network to obtain the camera module defect feature data. Finally, a product quality determiner is built, and the camera module defect feature data is quality evaluated and determined based on the product quality determiner to obtain the camera module quality detection result. The invention solves the technical problem of low detection efficiency of the endoscope camera module in the prior art and achieves the technical effect of improving the detection efficiency.

[0042] Embodiment 2 is based on the same inventive concept as the product quality inspection method applied to the endoscope camera module in the above embodiment. Figure 5 As shown, the present application provides a product quality inspection system for an endoscope camera module, wherein the system comprises: Installation module 11: obtain the camera module test fixture, clamp the target endoscope camera module to the camera module test fixture according to the preset direction, and power on the target endoscope camera module, and at the same time clamp the image test card just above the target endoscope camera module; Imaging detection module 12: control the camera module test fixture to perform imaging detection operations on the target endoscope camera module through the host computer, and receive the imaging information detected by the camera module based on the decoding board; Signal filtering module 13: filter the imaging information detected by the camera module based on the decoding board The row signal is filtered to obtain an available camera module detection image, and the available camera module detection image signal is transmitted to the host computer; the imaging defect analysis module 14: obtains the image defect detection algorithm network through the host computer, performs imaging defect analysis on the available camera module detection image based on the image defect detection algorithm network, and obtains the camera module defect feature data; the quality assessment module 15: builds a product quality determiner, performs quality assessment and determination on the camera module defect feature data based on the product quality determiner, and obtains the camera module quality detection result.

[0043] Furthermore, the signal filtering module 13 is used to perform the following method: Based on the decoding board, the camera module detection imaging information is subjected to signal type recognition and signal decoding conversion to obtain a camera module decoded detection image; the camera module decoded detection image is subjected to color channel conversion and signal enhancement to obtain a camera module enhanced detection image; the camera module enhanced detection image is subjected to noise recognition to obtain image multi-scale noise information; based on the image multi-scale noise information, the camera module enhanced detection image is filtered to obtain a usable camera module detection image.

[0044] Furthermore, the signal filtering module 13 is used to perform the following method: Obtain a list of filtering algorithms, and match the type characteristics of the multi-scale noise information of the image with the filtering algorithm list to obtain a multi-scale noise matching algorithm set; perform filtering parameter analysis on the multi-scale noise information of the image based on the multi-scale noise matching algorithm set to obtain a multi-scale noise filter; use the multi-scale noise filter to filter the camera module enhanced detection image to obtain the available camera module detection image.

[0045] Furthermore, the imaging defect analysis module 14 is used to perform the following method: According to the application standard of the endoscope camera module, a camera module imaging defect label library is constructed; a camera module defect image data set is collected and acquired, and the camera module defect image data set is type-labeled based on the camera module imaging defect label library to obtain a camera module defect image sample set; image defect detection requirements are acquired, and according to the image defect detection requirements, an image detection algorithm network structure is determined; the camera module defect image sample set is trained and optimized using the image detection algorithm network structure to obtain an image defect detection algorithm network and store it in the host computer.

[0046] Furthermore, the imaging defect analysis module 14 is used to perform the following method: Based on the image defect detection algorithm network, the available camera module detection image is marked for defect identification to obtain a camera module defect area set and a camera module defect type set; edge segmentation and integral calculation are performed on the camera module defect area set to obtain a camera module defect area area set; an image positioning coordinate system is constructed, and the camera module defect area set is positioned in turn based on the image positioning coordinate system to obtain a camera module defect area position set; based on the camera module defect type set, the camera module defect area set and the camera module defect area position set, the camera module defect feature data is obtained.

[0047] Furthermore, the quality assessment module 15 is used to perform the following method: Acquire product quality judgment factors, wherein the product quality judgment factors include a camera module defect type set, a defect region area, and a defect region position; construct a product quality judgment coordinate system according to the product quality judgment factors; divide the product quality judgment coordinate system into qualified product regions based on the product qualification standard of the target endoscope camera module to obtain qualified product region logic lines; and build the product quality determiner based on the product quality judgment coordinate system and the qualified product region logic lines.

[0048] Furthermore, the installation module 11 is used to execute the following method: The camera module testing tooling includes a module testing seat, a rotating platform, an X-axis adjustment knob, a Y-axis adjustment knob, a Z-axis adjustment handwheel, and a first interface, a second interface, and a third interface.

[0049] Furthermore, the installation module 11 is used to execute the following method: The target endoscope camera module is clamped and fixed on the module test seat according to the preset direction, and the position of the target endoscope camera module is adjusted using the X-axis adjustment knob, Y-axis adjustment knob and Z-axis adjustment handwheel; based on the rotating platform, the angle of the target endoscope camera module is adjusted according to the camera module test scenario, and the first interface, the second interface and the third interface are connected to external devices in sequence.

[0050] Embodiment 3, Figure 6 The schematic diagram of the structure of the electronic device provided in the third embodiment of the present invention shows a block diagram of an exemplary electronic device suitable for implementing the implementation mode of the present invention. Figure 6 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention. Figure 6 As shown, the electronic device includes a processor 21, a memory 22, an input device 23 and an output device 24; the number of processors 21 in the electronic device can be one or more. Figure 6 Taking a processor 21 as an example, the processor 21, memory 22, input device 23 and output device 24 in the electronic device can be connected through a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0051] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0052] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0053] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A product quality inspection method for an endoscope camera module, characterized in that: The method comprises: Obtain a camera module test fixture, and clamp the target endoscope camera module onto the camera module test fixture in a preset direction, wherein the preset direction refers to a specified installation angle or position according to the working state and design characteristics of the camera module, and power on the target endoscope camera module, and at the same time clamp the image test card just above the target endoscope camera module; The camera module test fixture is controlled by a host computer to perform imaging detection operations on the target endoscope camera module, and the camera module detection imaging information is received based on a decoder board; Based on the decoding board, the camera module detection imaging information is subjected to signal filtering to obtain an available camera module detection image, and the available camera module detection image signal is transmitted to the host computer; Acquire an image defect detection algorithm network through the host computer, perform imaging defect analysis on the available camera module detection image based on the image defect detection algorithm network, and obtain camera module defect feature data; A product quality determiner is built, and quality assessment and determination is performed on the defect feature data of the camera module based on the product quality determiner to obtain a camera module quality inspection result.

2. The product quality inspection method for an endoscope camera module according to claim 1, characterized in that: The method of obtaining an image that can be detected by a camera module includes: Based on the decoding board, the camera module detection imaging information is subjected to signal type recognition and signal decoding conversion to obtain a camera module decoding detection image; Performing color channel conversion and signal enhancement on the camera module decoded detection image to obtain a camera module enhanced detection image; Performing noise recognition on the camera module enhanced detection image to obtain multi-scale noise information of the image; The camera module enhanced detection image is filtered based on the multi-scale noise information of the image to obtain a usable camera module detection image.

3. The product quality inspection method for an endoscope camera module according to claim 2, characterized in that: The method of obtaining an image detected by a camera module includes: Acquire a filtering algorithm list, and match the filtering algorithm list with the type characteristics of the multi-scale noise information of the image respectively to obtain a multi-scale noise matching algorithm set; Based on the multi-scale noise matching algorithm set, filtering parameters are analyzed for the multi-scale noise information of the image to obtain a multi-scale noise filter; The multi-scale noise filter is used to filter the camera module enhanced detection image to obtain the usable camera module detection image.

4. The product quality inspection method for an endoscope camera module according to claim 3, characterized in that: The obtaining of the image defect detection algorithm network through the host computer includes: According to the endoscope camera module application standards, build a camera module imaging defect label library; Acquire a camera module defect image dataset, perform type annotation on the camera module defect image dataset based on the camera module imaging defect label library, and obtain a camera module defect image sample set; Obtaining image defect detection requirements, and determining an image detection algorithm network structure according to the image defect detection requirements; The image detection algorithm network structure is used to train and optimize the camera module defect image sample set to obtain an image defect detection algorithm network and store it in the host computer.

5. The product quality inspection method for an endoscope camera module according to claim 1, characterized in that: The obtaining of camera module defect feature data includes: Based on the image defect detection algorithm network, the available camera module detection image is marked for defect identification to obtain a camera module defect area set and a camera module defect type set; Performing edge segmentation and integral calculation on the camera module defect region set respectively to obtain a camera module defect region area set; Constructing an image positioning coordinate system, and positioning the camera module defect area set in sequence based on the image positioning coordinate system to obtain a camera module defect area position set; Based on the camera module defect type set, the camera module defect region area set and the camera module defect region position set, the camera module defect feature data is obtained.

6. The product quality inspection method for an endoscope camera module according to claim 1, characterized in that: The construction of the product quality determiner includes: Acquire product quality determination factors, wherein the product quality determination factors include a camera module defect type set, a defect area, and a defect area position; Constructing a product quality determination coordinate system according to the product quality determination factors; Based on the product qualification standard of the target endoscope camera module, the product quality determination coordinate system is divided into qualified product areas to obtain qualified product area logic lines; The product quality determiner is constructed based on the product quality determination coordinate system and the qualified product area logic line.

7. The product quality inspection method for an endoscope camera module according to claim 6, characterized in that: The camera module testing tooling includes a module testing seat, a rotating platform, an X-axis adjustment knob, a Y-axis adjustment knob, a Z-axis adjustment handwheel, and a first interface, a second interface, and a third interface.

8. The product quality inspection method for an endoscope camera module according to claim 7, characterized in that: The step of clamping the target endoscope camera module onto the camera module testing fixture according to a preset direction includes: Clamping and fixing the target endoscope camera module to the module test seat according to the preset direction, and using the X-axis adjustment knob, Y-axis adjustment knob and Z-axis adjustment handwheel to adjust the position of the target endoscope camera module; Based on the rotating platform, the angle of the target endoscope camera module is adjusted according to the camera module test scenario, and the first interface, the second interface and the third interface are connected to external devices in sequence.

9. A product quality inspection system for an endoscope camera module, characterized in that: For implementing the product quality detection method for an endoscope camera module according to any one of claims 1 to 8, the system comprises: Installation module: obtain the camera module test fixture, clamp the target endoscope camera module onto the camera module test fixture in a preset direction, power on the target endoscope camera module, and clamp the image test card just above the target endoscope camera module; Imaging detection module: controlling the camera module test fixture to perform imaging detection operations on the target endoscope camera module through the host computer, and receiving the camera module detection imaging information based on the decoding board; Signal filtering module: performing signal filtering on the imaging information detected by the camera module based on the decoding board to obtain an available camera module detection image, and transmitting the available camera module detection image signal to the host computer; Imaging defect analysis module: obtaining an image defect detection algorithm network through the host computer, performing imaging defect analysis on the available camera module detection image based on the image defect detection algorithm network, and obtaining camera module defect feature data; Quality assessment module: Build a product quality determiner, perform quality assessment and determination on the camera module defect feature data based on the product quality determiner, and obtain the camera module quality detection result.

10. An electronic device, characterized in that: The electronic device comprises: A memory for storing executable instructions; The processor is used to implement the product quality detection method applied to the endoscope camera module according to any one of claims 1 to 8 when executing the executable instructions stored in the memory.

Images