Optical instrument performance detection system and method based on multi-source data fusion analysis

Abstract

The application discloses an optical instrument performance detection system and method based on multi-source data fusion analysis, and relates to the technical field of optical instrument detection, to solve the technical problem that in the prior art, the signal characteristics of environmental light interference and product defects are prone to confusion, and specifically, through the cooperative work of an instrument self-calibration module, a calibration evaluation unit, an environmental light interference detection unit and a health state prediction unit, a full-process intelligent detection system is constructed, the application of multi-source data fusion technology breaks through the limitation of traditional single data detection, multi-modal feature matching, time series data analysis, grid fine monitoring and other technologies improve the accuracy and comprehensiveness of detection, accurate differentiation of environmental light interference and product defects reduces misjudgment and missed judgment, and the proactive maintenance of the hardware health state reduces downtime loss.

Description

Technical Field

[0001] This invention relates to the field of optical instrument testing technology, specifically to an optical instrument performance testing system and method based on multi-source data fusion analysis. Background Technology

[0002] Optical inspection technology is currently widely used in manufacturing product quality inspection, precision instrument measurement and other fields. In particular, under the production mode of multiple varieties and small batches, frequent production changes have become the norm in the industry.

[0003] However, in existing technologies, the self-calibration performance is reduced due to the lag in calibration template updates or the degradation of instrument hardware. Calibration templates are mostly built based on single image data and their updates lag behind the product iteration speed, making it difficult to adapt to the testing requirements of new product models, resulting in the accumulation of testing deviations. In addition, the signal characteristics of ambient light interference and product defects are easily confused, and traditional detection methods lack a precise differentiation mechanism, often resulting in misjudgment and missed judgment. Ambient light interference and product defects are difficult to distinguish. Therefore, a solution is proposed. Summary of the Invention

[0004] The purpose of this invention is to solve the problems mentioned above by proposing an optical instrument performance testing system and method based on multi-source data fusion analysis.

[0005] The objective of this invention can be achieved through the following technical solution: an optical instrument performance testing system based on multi-source data fusion analysis, and a performance testing center, wherein the performance testing center has the following communication connections: The instrument self-calibration module performs self-calibration based on the product changeover stage of the optical instrument's application scenario. The calibration evaluation unit evaluates the self-calibration performance of optical instruments based on the analysis of the self-calibration process. After completing the self-calibration evaluation, the ambient light interference detection unit performs light interference detection during the use of the optical instrument; The health status prediction unit predicts the health status of optical instruments.

[0006] Furthermore, the instrument self-calibration module process is as follows: When optical instruments are used to acquire images in actual scenarios, they pre-store multi-source data templates (including high-resolution images, spectral features, and 3D point cloud data) of standard calibration boards of all different product models in the scenario. When changing products, the optical instruments automatically scan the new model standard calibration board at the current workstation and acquire its multi-source data in real time. Multimodal feature matching (including image feature point matching, spectral curve fitting, and flatness point cloud analysis) is performed between real-time scan data and pre-stored templates to calculate the deviation matrix between existing instrument parameters and ideal parameters; the deviation matrix is ​​then compared with a preset threshold. If the deviation matrix does not exceed the preset parameter threshold, the optical instrument is confirmed to be ready; if the deviation matrix exceeds the preset parameter threshold, the system automatically generates a set of optimal calibration parameter instructions and drives the optical instrument to perform adaptive calibration until the deviation enters the tolerance range; finally, a calibration report is generated, the calibration process is recorded, and all processes are marked as self-calibration processes.

[0007] Furthermore, the process of the instrument self-calibration evaluation unit is as follows: When the update speed of the pre-stored template is lagging, collect the proportion of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type and the average value of the data fluctuation span of the corresponding deviation parameters. If the proportion of the number of types corresponding to the deviation parameters in the deviation matrix of the current product type exceeds the quantity proportion threshold, or the average value of the data fluctuation span of the corresponding deviation parameters exceeds the average value of the fluctuation span threshold, it is inferred that the self-calibration performance is abnormal in the current production changeover stage. Specifically, if the template update is delayed, a changeover and re-inspection signal is generated and sent to the performance testing center. After receiving the changeover and re-inspection signal, the performance testing center performs product quality checks on products that have passed self-calibration and are executed, and stops production of the products if there are any abnormalities, and updates the corresponding pre-stored templates. For products that have not passed self-calibration, the pre-stored templates are updated and compared again. If there is no mismatch, the products are pre-processed and the product pass rate is tested. If the pass rate meets the requirements, the current product's multi-source data is included in the pre-stored template; otherwise, the product's multi-source data is adjusted and included in the pre-stored module. If the proportion of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type does not exceed the quantity proportion threshold, and the average value of the data fluctuation span of the corresponding deviation parameters does not exceed the average value of the fluctuation span threshold, then it is inferred that the self-calibration performance in the current production changeover stage is normal, the pre-stored module is updated in a timely manner, and the current product is continuously monitored.

[0008] Furthermore, when the pre-stored template update speed is in the timely stage, the numerical deviation fluctuation frequency of the fitting parameters in the deviation matrix corresponding to the current product type is obtained. If the numerical deviation fluctuation frequency is higher than the set frequency red line value, it is inferred that the self-calibration performance of the optical instrument has deteriorated, a hardware maintenance signal is generated and sent to the performance testing center. After receiving the signal, the performance testing center performs hardware maintenance on the optical instrument. If the frequency of the numerical deviation fluctuation is not higher than the set frequency red line value, it is inferred that the self-calibration performance of the optical instrument is stable, and a calibration execution signal is generated and sent to the performance testing center.

[0009] Furthermore, the process of the ambient light interference detection unit is as follows: The optical instrument is set as the main detection end, and optical sensors are set around the corresponding detection area of ​​the main detection end. The exposure of the main camera and the reading of the ambient light sensor are triggered synchronously at a fixed sampling period. The acquired image is divided into multiple non-overlapping grids, and the global average brightness of the entire image is calculated. Align the global average brightness sequence and ambient light intensity sequence of the current frame with timestamps to form a synchronized time-series data stream; set a time window and calculate the variance of the data sequence within the time window. If the calculated variance increases instantaneously, mark the corresponding time point within the time window as a change point; otherwise, mark it as a non-change point. Using mutation points as the detection entry point, the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are obtained. If the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are consistent, the product image to be detected acquired at the current time point is set as a global mutation. If the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are inconsistent, the product image to be detected acquired at the current time point is set as a local mutation.

[0010] Furthermore, when there is a localized mutation, if the corresponding grid brightness change continues, it is inferred that the image of the product to be tested has detected a product defect, generating a product abnormality signal and sending it to the performance testing center. When the performance testing center determines that the current ambient light interference test is qualified, it forwards it to the administrator and inspects and repairs the current tested product. When a global mutation occurs, the time taken for the grid brightness value to change within the current time window is recorded. If the change time is less than the set threshold and the grid brightness value returns to a fixed value after the change time is completed, it indicates that there is ambient light interference. An optical interference signal is generated and sent to the performance testing center.

[0011] Furthermore, the process of the health status prediction unit is as follows: During the continuous acquisition and testing of product images without interference during the continuous operation of the optical instrument, the actual range of change in image parameters of any grid in the acquired product image and the range of change in the displayed image are obtained, and the image parameter acquisition deviation is obtained by comparing the range of change; the increase in the number of adjacent grids with image parameter acquisition deviation in all grids is obtained, and the decrease in the area of ​​the corresponding grid in any grid without acquisition deviation is obtained.

[0012] Furthermore, if the increase in the number of adjacent grids with image parameter acquisition deviations in all grids exceeds the threshold for the increase in the number of grids, or if the decrease in the area of ​​the grid corresponding to the grid without acquisition deviation in any grid position exceeds the threshold for the decrease in area, it is inferred that the hardware status of the optical instrument during the continuous operation phase is in a sub-healthy state, and an instrument status correction signal is generated and sent to the performance testing center. If the increase in the number of adjacent grids with image parameter acquisition deviations within all grids does not exceed the threshold for the increase in the number of grids, and the decrease in the area of ​​the grid corresponding to the grid without acquisition deviations at any position does not exceed the threshold for the decrease in area, then it is inferred that the hardware status of the optical instrument is in a healthy state during the continuous operation phase, and a normal instrument status signal is generated and sent to the performance testing center.

[0013] Furthermore, after receiving the instrument status correction signal, the performance testing center maintains the optical instruments put into use and obtains the image acquisition parameters of the optical instruments when they are first put into use according to the work log, so as to analyze the current aging degree of the optical instruments and make targeted task allocation.

[0014] This invention also proposes a method for testing the performance of optical instruments based on multi-source data fusion analysis, the specific steps of which are as follows: Step 1: Instrument self-calibration. Perform self-calibration according to the product changeover stage of the optical instrument's application scenario. Step 2: Calibration Evaluation. Based on the analysis of the self-calibration process, the self-calibration performance of the optical instrument is evaluated. Step 3: After completing the self-calibration assessment, perform optical interference detection during the use of the optical instrument; Step 4: Health status prediction. Predict the health status of the optical instruments.

[0015] Compared with the prior art, the beneficial effects of the present invention are: The instrument self-calibration module's related steps break through the limitations of traditional single-data calibration by pre-storing multi-source data templates covering different product models, such as high-resolution images, spectral features, and 3D point clouds, providing comprehensive data source support for production changeover calibration. During production changeovers, the instrument automatically scans the new model standard calibration board and collects multi-source data in real time without manual intervention, significantly reducing the time and cost of calibration during production changeovers and improving the continuity of the production process. Multimodal feature matching technology (image feature point matching, spectral curve fitting, and flatness point cloud analysis) accurately captures the differences between instrument parameters and ideal values. Combined with the comparison of the deviation matrix and preset thresholds, the calibration judgment is more targeted. When the threshold is exceeded, the optimal calibration parameters are automatically generated and adaptive calibration is performed until the deviation enters the tolerance range, avoiding the subjectivity and instability of manual calibration. The generation of calibration reports enables the calibration process to be traceable and verifiable.

[0016] The calibration evaluation unit is designed with differentiated evaluation logic for two scenarios: template update lag and timely update. This effectively covers the core causes of self-calibration performance degradation. In the template update lag stage, by monitoring the proportion of deviation parameter types and the average data fluctuation range, the template failure problem can be accurately located, avoiding batch product failure due to untimely template updates. The subsequent performance testing center conducts quality checks on the tested products, handles production stoppages, and performs template updates and data adjustments for new models, forming a complete closed-loop management system. This significantly reduces the risk of non-conforming products leaving the market. During timely template updates, the instrument's self-calibration performance is judged by the frequency of numerical deviation fluctuations. This allows for early identification of hardware degradation trends, avoiding testing errors caused by hardware failures or sudden production line shutdowns, making instrument maintenance more proactive.

[0017] The ambient light interference detection unit ensures precise time alignment between image data and ambient light data by setting the main detection end and surrounding optical sensors, laying a solid foundation for interference judgment. It marks abrupt change points by calculating the variance of the time window and distinguishes between global and local abrupt changes by combining the interval duration and trend of grid brightness changes, successfully solving the pain point of difficulty in distinguishing between ambient light interference and product defects. When localized mutations occur, product defects are accurately identified and repair processes are triggered. When global mutations occur, intervention and control measures are taken to prevent and control ambient light mutations caused by forklifts passing by, doors and windows closing, etc. This ensures product quality, avoids the impact of environmental factors on detection efficiency, and improves the system's anti-interference capability.

[0018] The health status prediction unit acquires images of the tested products in an interference-free environment. By comparing the actual range of change in grid image parameters with the displayed range of change, it accurately obtains the image parameter acquisition deviation and objectively reflects the acquisition performance of the instrument hardware. By monitoring the increase in the number of adjacent deviation grids and the decrease in the area of ​​undevised grids, changes in hardware status are quantified into comparable indicators, avoiding ambiguous judgments about hardware status. After identifying sub-health conditions of the instruments in advance, the performance testing center analyzes the degree of aging by combining the acquired image parameters of the optical instruments when they are first put into use, and carries out targeted maintenance and task allocation to extend the service life of the instruments and reduce downtime losses caused by sudden failures. Attached Figure Description

[0019] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0020] Figure 1 This is a system principle block diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] Please see Figure 1 As shown, the optical instrument performance testing system based on multi-source data fusion analysis includes a performance testing center, which is connected to an instrument self-calibration module, a calibration evaluation unit, an ambient light interference detection unit, and a health status prediction unit. The performance testing center generates an instrument self-calibration signal and sends it to the instrument self-calibration module; after receiving the instrument self-calibration signal, the instrument self-calibration module performs self-calibration processing according to the product changeover stage of the optical instrument's application scenario. When optical instruments are used to acquire images in actual scenarios, they pre-store multi-source data templates (including high-resolution images, spectral features, and 3D point cloud data) of standard calibration boards of all different product models in the scenario. When changing products, the optical instruments automatically scan the new model standard calibration board at the current workstation and acquire its multi-source data in real time. Multimodal feature matching (such as image feature point matching, spectral curve fitting, and flatness point cloud analysis) is performed between real-time scan data and pre-stored templates to calculate the deviation matrix between existing instrument parameters and ideal parameters; the deviation matrix is ​​then compared with a preset threshold. If the deviation matrix does not exceed the preset parameter threshold, the optical instrument is confirmed to be ready; if the deviation matrix exceeds the preset parameter threshold, the system automatically generates a set of optimal calibration parameter instructions (such as adjusting the camera ISO, focal motor steps, and software distortion correction coefficients), and drives the optical instrument to perform adaptive calibration until the deviation enters the tolerance range; finally, a calibration report is generated, the calibration process is recorded, and all processes are marked as self-calibration processes. The performance testing center generates an instrument self-calibration evaluation signal and sends it to the instrument self-calibration evaluation unit. After receiving the instrument self-calibration evaluation signal, the instrument self-calibration evaluation unit evaluates the self-calibration performance of the optical instrument. When the pre-stored template update speed is lagging, the percentage of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type and the average data fluctuation span of the corresponding deviation parameters are collected. If the percentage of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type exceeds the percentage threshold, or the average data fluctuation span of the corresponding deviation parameters exceeds the average fluctuation span threshold, it is inferred that the self-calibration performance of the current production changeover stage is abnormal, specifically that the template update is lagging. A production changeover re-test signal is generated and sent to the performance testing center. After receiving the production changeover re-test signal, the performance testing center performs product quality checks on products that have passed self-calibration and are executed. If there is an abnormality, the product is stopped and the corresponding pre-stored template is updated. For products that have not passed self-calibration, the pre-stored template is updated and compared again. If there is no mismatch, the product is pre-processed and the product pass rate is tested. If the pass rate meets the requirements, the multi-source data of the current product is included in the pre-stored template. Otherwise, the multi-source data of the product is adjusted and included in the pre-stored module. If the proportion of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type does not exceed the proportion threshold, and the average value of the data fluctuation span of the corresponding deviation parameters does not exceed the average value of the fluctuation span threshold, then it is inferred that the self-calibration performance in the current production changeover stage is normal, the pre-stored module is updated in a timely manner, and the current product is continuously monitored. When the pre-stored template update speed is in the timely stage, the numerical deviation fluctuation frequency of the fitting parameters in the deviation matrix corresponding to the current product type is obtained. If the numerical deviation fluctuation frequency is higher than the set frequency red line value, it is inferred that the self-calibration performance of the optical instrument has deteriorated, a hardware maintenance signal is generated and sent to the performance testing center. After receiving the signal, the performance testing center performs hardware maintenance on the optical instrument. If the frequency of the numerical deviation does not exceed the set frequency red line value, it is inferred that the self-calibration performance of the optical instrument is stable, and a calibration execution signal is generated and sent to the performance testing center. After completing the self-calibration evaluation, the performance testing center generates an ambient light interference detection signal and sends it to the ambient light interference detection unit. After receiving the ambient light interference detection signal, the ambient light interference detection unit performs light interference detection during the use of the optical instrument. The optical instrument is set as the main detection end, and optical sensors are set around the corresponding detection area of ​​the main detection end. The exposure of the main camera and the reading of the ambient light sensor are synchronously triggered at a fixed sampling period (e.g., 500 frames per second). Each data packet carries a high-precision timestamp accurate to the microsecond level. A complete image of the product to be inspected is obtained through the main detection end, and the light intensity (Lux) and color temperature (K) values ​​of one or more data points are collected through the optical sensor. The acquired image is divided into multiple non-overlapping grids, and the global average brightness of the entire image is calculated. Align the global average brightness sequence and ambient light intensity sequence of the current frame with timestamps to form a synchronized time-series data stream; set a time window and calculate the variance of the data sequence within the time window. If the calculated variance increases instantaneously, mark the corresponding time point within the time window as a change point; otherwise, mark it as a non-change point. Using mutation points as the detection entry point, the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are obtained. If the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are consistent, the product image to be detected acquired at the current time point is set as a global mutation. If the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are inconsistent, the product image to be detected acquired at the current time point is set as a local mutation. When there is a local mutation, if the brightness change of the corresponding grid continues, it is inferred that the image of the product to be tested has detected a product defect, and an abnormal product signal is generated and sent to the performance testing center. When the performance testing center determines that the current ambient light interference test is qualified, it forwards it to the administrator and checks and repairs the current tested product. When a global sudden change occurs, the time taken for the grid brightness value to change within the current time window is recorded. If the change time is lower than the set threshold and the grid brightness value returns to a fixed value after the change time is completed, it indicates that there is ambient light interference. An optical interference signal is generated and sent to the performance testing center. After the performance testing platform determines the optical interference detection performance of the current optical instrument, it maintains the defective products and intervenes and controls the environment in the area where the current optical instrument is located to avoid environmental sudden changes that reduce optical detection efficiency, such as forklifts passing by or doors and windows closing, which cause instantaneous and drastic changes in ambient light in the area. The performance testing center generates a health status prediction signal and sends it to the health status prediction unit. After receiving the health status prediction signal, the health status prediction unit predicts the health status of the optical instrument. During the continuous acquisition and detection of product images without interference during the continuous operation of the optical instrument, the actual range of change of image parameters in any grid of the acquired product image and the range of change of the image displayed are obtained, and the image parameter acquisition deviation is obtained by comparing the range of change. The number of adjacent grids with image parameter acquisition deviations within all grids is increased by the span, and the area of ​​the corresponding grid without acquisition deviations in any grid position is decreased by the span. The increase in the number of adjacent grids with image parameter acquisition deviations within all grids, and the decrease in the area of ​​grids without acquisition deviations at any location, are compared with the thresholds for the increase in the number of grids and the decrease in the area of ​​grids, respectively. If the increase in the number of adjacent grids with image parameter acquisition deviations exceeds the threshold for the increase in the number of grids, or if the decrease in the area of ​​the grid corresponding to a grid without acquisition deviations at any position exceeds the threshold for the decrease in area, it is inferred that the hardware status of the optical instrument is in a sub-healthy state during the continuous operation phase. An instrument status correction signal is generated and sent to the performance testing center. After receiving the signal, the performance testing center maintains the optical instrument that has been put into use and obtains the image acquisition parameters acquired when the optical instrument was first put into use based on the work log in order to analyze the current aging degree of the optical instrument and make targeted task allocation. If the increase in the number of adjacent grids with image parameter acquisition deviations within all grids does not exceed the threshold for the increase in the number of grids, and the decrease in the area of ​​the grid corresponding to the grid without acquisition deviations at any position does not exceed the threshold for the decrease in area, then it is inferred that the hardware status of the optical instrument is in a healthy state during the continuous operation phase, and a normal instrument status signal is generated and sent to the performance testing center.

[0024] Please see Figure 2 As shown, this invention also proposes a method for testing the performance of optical instruments based on multi-source data fusion analysis, the specific steps of which are as follows: Step 1: Instrument self-calibration. Perform self-calibration according to the product changeover stage of the optical instrument's application scenario. Step 2: Calibration Evaluation. Based on the analysis of the self-calibration process, the self-calibration performance of the optical instrument is evaluated. Step 3: After completing the self-calibration assessment, perform optical interference detection during the use of the optical instrument; Step 4: Health status prediction. Predict the health status of the optical instruments.

[0025] Thresholds, preset values, preset ranges, etc. are set for result comparison and analysis to determine whether they are good or bad. The value of these thresholds is determined by a combination of large-scale model analysis of sample data and human experience. They can also be adjusted appropriately based on seasonal or common-sense influences. The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. An optical instrument performance detection system based on multi-source data fusion analysis, characterized in that, This includes a performance testing center, whose communication connections include: The instrument self-calibration module performs self-calibration based on the product changeover stage of the optical instrument's application scenario. The calibration evaluation unit evaluates the self-calibration performance of optical instruments based on the analysis of the self-calibration process. When the update speed of the pre-stored template is lagging, collect the proportion of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type and the average value of the data fluctuation span of the corresponding deviation parameters. If the proportion of the number of types corresponding to the deviation parameters in the deviation matrix of the current product type exceeds the quantity proportion threshold, or the average value of the data fluctuation span of the corresponding deviation parameters exceeds the average value of the fluctuation span threshold, it is inferred that the self-calibration performance is abnormal in the current production changeover stage. Specifically, if the template update is delayed, a change-of-production retest signal will be generated and sent to the performance testing center. After receiving the change-of-production retest signal, the performance testing center will conduct a product quality check on the products that have passed self-calibration and have been executed, and will stop production of the products if there are any abnormalities, and update the corresponding pre-stored template. Products that fail self-calibration will be re-pre-stored and compared using the template. If there is no matching type, the product will be pre-processed and the product pass rate will be tested. If the pass rate meets the requirements, the current product's multi-source data will be included in the pre-stored template. Otherwise, the product's multi-source data will be adjusted and included in the pre-stored module. If the proportion of the number of types corresponding to the deviation parameters in the deviation matrix corresponding to the current product type does not exceed the proportion threshold, and the average value of the data fluctuation span of the corresponding deviation parameters does not exceed the average value of the fluctuation span threshold, then it is inferred that the self-calibration performance in the current production changeover stage is normal, the pre-stored module is updated in a timely manner, and the current product is continuously monitored. When the pre-stored template update speed is in the timely stage, the numerical deviation fluctuation frequency of the fitting parameters in the deviation matrix corresponding to the current product type is obtained. If the numerical deviation fluctuation frequency is higher than the set frequency red line value, it is inferred that the self-calibration performance of the optical instrument has deteriorated, a hardware maintenance signal is generated and sent to the performance testing center. After receiving the signal, the performance testing center performs hardware maintenance on the optical instrument. If the frequency of the numerical deviation does not exceed the set frequency red line value, it is inferred that the self-calibration performance of the optical instrument is stable, and a calibration execution signal is generated and sent to the performance testing center. After completing the self-calibration evaluation, the ambient light interference detection unit performs light interference detection during the use of the optical instrument; The health status prediction unit predicts the health status of optical instruments.

2. The optical instrument performance detection system based on multi-source data fusion analysis according to claim 1, characterized in that, The instrument self-calibration module process is as follows: When optical instruments are used to acquire images in actual scenarios, they pre-store multi-source data templates of standard calibration boards of all different product models in the scenario, including high-resolution images, spectral features, and 3D point cloud data; when changing products, optical instruments automatically scan the new model standard calibration board at the current workstation and acquire its multi-source data in real time. Multimodal feature matching is performed between real-time scanning data and pre-stored templates, including image feature point matching, spectral curve fitting, and flatness point cloud analysis. The deviation matrix between existing instrument parameters and ideal parameters is calculated, and the deviation matrix is ​​compared with a preset threshold. If the deviation matrix does not exceed the preset parameter threshold, the optical instrument is confirmed to be ready; if the deviation matrix exceeds the preset parameter threshold, the system automatically generates a set of optimal calibration parameter instructions and drives the optical instrument to perform adaptive calibration until the deviation enters the tolerance range; finally, a calibration report is generated, the calibration process is recorded, and all processes are marked as self-calibration processes.

3. The optical instrument performance detection system based on multi-source data fusion analysis according to claim 1, characterized in that, The process of the ambient light interference detection unit is as follows: The optical instrument is set as the main detection end, and optical sensors are set around the corresponding detection area of ​​the main detection end. The exposure of the main camera and the reading of the ambient light sensor are triggered synchronously at a fixed sampling period. The acquired image is divided into multiple non-overlapping grids, and the global average brightness of the entire image is calculated. Align the global average brightness sequence of the current frame with the ambient light intensity sequence according to the timestamp to form a synchronized time-series data stream; Set a time window and calculate the variance of the data sequence within the time window. If the calculated variance increases instantaneously, mark the corresponding time point within the time window as a mutation point; otherwise, mark it as a non-mutation point. Using mutation points as the detection entry point, the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are obtained. If the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are consistent, the product image to be detected acquired at the current time point is set as a global mutation. If the brightness change interval and corresponding brightness change trend of the non-mutation point grid and mutation point grid in the current product image to be detected are inconsistent, the product image to be detected acquired at the current time point is set as a local mutation.

4. The optical instrument performance testing system based on multi-source data fusion analysis according to claim 3, characterized in that, When there is a local mutation, if the brightness change of the corresponding grid continues, it is inferred that the image of the product to be tested has detected a product defect, and an abnormal product signal is generated and sent to the performance testing center. When the performance testing center determines that the current ambient light interference test is qualified, it forwards it to the administrator and checks and repairs the current tested product. When a global mutation occurs, the time taken for the grid brightness value to change within the current time window is recorded. If the change time is less than the set threshold and the grid brightness value returns to a fixed value after the change time is completed, it indicates that there is ambient light interference. An optical interference signal is generated and sent to the performance testing center.

5. The optical instrument performance testing system based on multi-source data fusion analysis according to claim 1, characterized in that, The process of the health status prediction unit is as follows: During the continuous acquisition and testing of product images without interference during the continuous operation of the optical instrument, the actual range of change in image parameters of any grid in the acquired product image and the range of change in the displayed image are obtained, and the image parameter acquisition deviation is obtained by comparing the range of change; the increase in the number of adjacent grids with image parameter acquisition deviation in all grids is obtained, and the decrease in the area of ​​the corresponding grid in any grid without acquisition deviation is obtained.

6. The optical instrument performance testing system based on multi-source data fusion analysis according to claim 5, characterized in that, If the increase in the number of adjacent grids with image parameter acquisition deviations in all grids exceeds the threshold for the increase in the number of grids, or if the decrease in the area of ​​the grid corresponding to the grid without acquisition deviation in any grid position exceeds the threshold for the decrease in area, it is inferred that the hardware status of the optical instrument is in a sub-healthy state during the continuous operation phase, and an instrument status correction signal is generated and sent to the performance testing center. If the increase in the number of adjacent grids with image parameter acquisition deviations within all grids does not exceed the threshold for the increase in the number of grids, and the decrease in the area of ​​the grid corresponding to the grid without acquisition deviations at any position does not exceed the threshold for the decrease in area, then it is inferred that the hardware status of the optical instrument is in a healthy state during the continuous operation phase, and a normal instrument status signal is generated and sent to the performance testing center.

7. The optical instrument performance testing system based on multi-source data fusion analysis according to claim 6, characterized in that, After receiving the instrument status correction signal, the performance testing center maintains the optical instruments that have been put into use and obtains the image acquisition parameters of the optical instruments when they are first put into use according to the work log, so as to analyze the current aging degree of the optical instruments and make targeted task allocation.

8. A method for testing the performance of optical instruments based on multi-source data fusion analysis, characterized in that, The specific performance testing method for the optical instrument performance testing system based on multi-source data fusion analysis as described in any one of claims 1-7 is as follows: Step 1: Instrument self-calibration. Perform self-calibration according to the product changeover stage of the optical instrument's application scenario. Step 2: Calibration Evaluation. Based on the analysis of the self-calibration process, the self-calibration performance of the optical instrument is evaluated. Step 3: After completing the self-calibration assessment, perform optical interference detection during the use of the optical instrument; Step 4: Health status prediction. Predict the health status of the optical instruments.

Images