Multi-mode FPC defect detection method and system, medium and product

Abstract

The invention relates to the technical field of defect detection, in particular to a multi-mode FPC defect detection method and system, a medium and a product, and the method comprises the steps: obtaining optical data, thermal imaging data and circuit on-off data corresponding to a to-be-detected FPC; determining a surface feature parameter corresponding to each detection position on the to-be-detected FPC from the optical data; determining a temperature characteristic parameter corresponding to each detection position on the to-be-detected FPC from the thermal imaging data; determining a circuit characteristic parameter corresponding to each detection position on the to-be-detected FPC from the circuit on-off data; and obtaining a defect type and a defect detection score corresponding to each detection position on the to-be-detected FPC based on the surface feature parameter, the temperature feature parameter and the circuit feature parameter corresponding to each detection position on the to-be-detected FPC and a preset parameter weight. The defect detection precision can be conveniently improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of defect detection, in particular to a multi-modal FPC defect detection method, system, medium and product. BACKGROUND

[0002] Flexible Printed Circuit (FPC) is an indispensable core component in modern detection systems, with its lightness, thinness, bendability, high integration and other characteristics, it is widely used in smart phones, wearable devices, automotive electronics, medical devices and other fields. With the rapid development of 5G communication, Internet of Things, artificial intelligence and other technologies, the complexity and performance requirements of FPC are constantly improving, however, this technology upgrade also brings higher manufacturing difficulty and reliability challenges.

[0003] The manufacturing process of FPC involves photoetching, etching, lamination, drilling, electroplating and other precise processes, any slight deviation in any link may cause defects such as wire breakage, short circuit, failure, etc. Currently, optical cameras or X-ray detection technology are commonly used for FPC defect detection in the industry, however, these traditional methods are limited by the limitations of single detection mode, and cannot fully capture various types of FPC defects, especially in complex environments, single mode detection often leads to low defect detection accuracy, if the FPC with defects flows into the market due to low defect detection accuracy and other problems, it may seriously affect the service life and stability of the product. SUMMARY

[0004] In order to improve the defect detection accuracy, the present application provides a multi-modal FPC defect detection method, system, medium and product.

[0005] In a first aspect, the present application provides a multi-modal FPC defect detection method, which adopts the following technical solution: A multi-modal FPC defect detection method, comprising: obtaining optical data, thermal imaging data and circuit on-off data corresponding to a to-be-detected FPC; determining surface feature parameters corresponding to each detection position on the to-be-detected FPC from the optical data; determining temperature feature parameters corresponding to each detection position on the to-be-detected FPC from the thermal imaging data; determining circuit feature parameters corresponding to each detection position on the to-be-detected FPC from the circuit on-off data; based on the surface feature parameters, temperature feature parameters, circuit feature parameters corresponding to each detection position on the to-be-detected FPC and a preset parameter weight, obtaining a defect type and a defect detection score corresponding to each detection position on the to-be-detected FPC.

[0006] By adopting the technical scheme, the optical data, the thermal imaging data and the circuit on-off data of the FPC to be detected are analyzed at the same time, so that the advantages of different detection modes can be fully utilized by means of multi-modal data fusion, and mutual supplement and verification, wherein the optical data is analyzed to facilitate understanding of the intuitive information of the surface of each detection position, the thermal imaging data is analyzed to facilitate detection of the local temperature abnormality of each detection position caused by defects, and the circuit on-off data is analyzed to facilitate direct reflection of the circuit connection of each detection position. By comprehensively analyzing these multi-modal data, the defects of the FPC to be detected can be more accurately recognized and positioned, the miss detection and false detection rates can be effectively reduced even in a complex environment, and the reliability of defect detection is enhanced, so that the quality of the FPC products flowing into the market can meet the requirements, and the service life and stability of the products are ensured.

[0007] In a possible implementation manner, the defect type and the defect detection score corresponding to the detection position are obtained based on the surface feature parameter, the temperature feature parameter, the circuit feature parameter corresponding to the detection position and the preset parameter weight, and include: The surface feature parameter, the temperature feature parameter and the circuit feature parameter corresponding to the detection position are normalized to obtain a surface feature value, a temperature feature value and a circuit feature value; The surface feature value, the temperature feature value and the circuit feature value are determined to correspond to a surface feature coordinate value, a temperature feature coordinate value and a circuit feature coordinate value according to the preset parameter weight; A three-dimensional display feature is determined according to the surface feature coordinate value, the temperature feature coordinate value and the circuit feature coordinate value, and the defect type and the defect detection score corresponding to the detection position are determined based on the three-dimensional display feature and a preset peripheral coil.

[0008] By adopting the technical scheme, the surface feature parameter, the temperature feature parameter and the circuit feature parameter corresponding to the detection position are normalized, so that the influence of different feature parameters caused by the difference in dimension or value range can be eliminated, and the feature parameters can be compared or fused in the same numerical scale. The surface feature value, the temperature feature value and the circuit feature value are determined to correspond to the surface feature coordinate value, the temperature feature coordinate value and the circuit feature coordinate value according to the preset parameter weight, so that the importance of different feature parameters in defect detection can be reflected by introducing the preset parameter weight, and the key feature parameters can occupy more prominent positions in the feature coordinate system in defect detection by applying the preset parameter weight to the conversion of the feature values, so that the influence of the feature parameters on the defect detection result can be better reflected. Finally, the defect type and the defect detection score corresponding to the detection position can be determined intuitively and accurately by the three-dimensional display feature and the preset peripheral coil.

[0009] In a possible implementation manner, the method further includes: determining, according to the surface feature parameter corresponding to the detection position and a preset peripheral interval mapping relationship, a peripheral interval corresponding to the detection position, the preset peripheral interval mapping relationship being a corresponding relationship between a surface feature parameter and a peripheral interval; determining a final preset peripheral loop based on the peripheral interval, and superimposing the preset peripheral loop and the stereoscopic display feature to obtain a loop superimposed display feature map; identifying a target peripheral line from the loop superimposed display feature map, and determining a defect detection score corresponding to the detection position according to the target peripheral line and a preset defect score mapping relationship, wherein a peripheral line surrounding the stereoscopic display feature is determined as the target peripheral line, and the preset defect score mapping relationship is a corresponding relationship between a detection position and a defect detection score; identifying a defect detection score corresponding to each feature vertex in the stereoscopic display feature from the loop superimposed display feature map, and determining a defect type corresponding to the detection position according to the defect detection score corresponding to each feature vertex and a preset defect type mapping relationship, the preset defect type mapping relationship being a corresponding relationship between a score combination of defect detection scores corresponding to respective feature vertices and a defect type.

[0010] By using the above technical solution, because different surface feature parameters have different influences on defect detection results, the corresponding peripheral interval is dynamically determined based on the surface feature parameter, so that the adaptability in the defect detection process is improved, thereby avoiding missed detection or false detection of some defects due to unreasonable setting of the peripheral interval. By superimposing the preset peripheral loop on the stereoscopic display feature, the preset peripheral loop and the stereoscopic display feature are intuitively combined together, so that the target peripheral line is intuitively and accurately determined from the preset peripheral loop, that is, the defect detection score corresponding to the detection position is accurately determined. Because each feature vertex in the stereoscopic display feature contains rich defect information, by identifying the defect detection score corresponding to each feature vertex and comprehensively analyzing by using the preset defect type mapping relationship, the defect type of the detection position is more accurately determined.

[0011] In a possible implementation manner, the circuit feature parameter corresponding to any detection position on the FPC to be detected is determined from the circuit on-off data, including: identifying, from the circuit on-off data, an on-off state, an on-off position, and a time domain feature parameter corresponding to the detection position; when the time domain feature parameter includes a preset time domain feature, determining an influence position and an influence parameter value based on the preset time domain feature; obtaining an actual parameter value corresponding to the influence position, and when a parameter difference between the actual parameter value and the influence parameter value is lower than a preset difference threshold, optimizing the time domain characteristic parameter according to the influence parameter value to obtain an optimized time domain characteristic parameter; determining a circuit characteristic parameter corresponding to the detection position based on the on-off state, the on-off position, and the optimized time domain characteristic parameter.

[0012] By using the above technical solution, the on-off state, the on-off position, and the time domain characteristic parameter are used as the circuit characteristic parameter in the defect detection process, which facilitates improving the comprehensiveness in the defect detection process. In addition, when the time domain characteristic parameter includes a preset time domain characteristic that may have a short circuit hidden danger, the situation that may occur at the influence position is predicted, and the actual situation that occurs at the influence position is collected for reverse verification. When the verification is passed, the time domain characteristic parameter can be optimized. The optimized time domain characteristic parameter facilitates more accurately reflecting the dynamic performance and potential problems of the detection position, thereby facilitating improving the accuracy in the defect detection.

[0013] In a possible implementation manner, when the surface characteristic parameter of the detection position includes a preset surface characteristic, the method further includes: determining the detection position including the preset surface characteristic as an abnormal detection position; statistically obtaining a historical abnormal frequency and a historical failure probability corresponding to the abnormal detection position from historical detection records; adjusting a preset parameter weight corresponding to the abnormal detection position based on the historical abnormal frequency and the historical failure probability.

[0014] By using the above technical solution, the detection position including the preset surface characteristic is determined as the abnormal detection position, which can quickly and accurately lock the area that may have defects. By combining the historical data with the current detection result, the influence of the current abnormal detection position on the reliability of the to-be-detected FPC can be more accurately evaluated. By dynamically adjusting the parameter weight through multiple parameters, the adaptability of the detection operation to various complex situations can be improved, thereby facilitating improving the accuracy in determining the defect detection score.

[0015] In a possible implementation manner, the method further includes: performing cluster analysis on a defect type and a defect detection score corresponding to each detection position on each to-be-detected FPC in a preset time period; judging whether similar defect groups exist on different to-be-detected FPCs based on the cluster analysis result, the similar defect groups including at least two same detection positions from different to-be-detected FPCs, and a defect score difference between defect detection scores corresponding to the at least two same detection positions being lower than a preset score threshold. If yes, process feedback information is generated based on the similar defect group.

[0016] By adopting the technical scheme, the defect types and defect detection scores of each detection position on each FPC to be detected in a period of time are analyzed by clustering, potential correlations possibly existing on different FPCs to be detected are effectively mined, process feedback information is generated through the potential correlations, and process improvement clues are provided for relevant staff.

[0017] In a second aspect, the present application provides a detection system, which adopts the following technical scheme: A detection system, comprising: at least one processor; a memory; at least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one application program is configured to execute the above-mentioned multi-modal FPC defect detection method.

[0018] In a third aspect, the present application provides a computer-readable storage medium, which adopts the following technical scheme: A computer-readable storage medium, comprising a computer program capable of being loaded and executed by a processor to execute the above-mentioned multi-modal FPC defect detection method.

[0019] In a fourth aspect, the present application provides a computer program product, which adopts the following technical scheme: A computer program product, comprising a computer program, wherein the computer program is executed by a processor to implement the above-mentioned multi-modal FPC defect detection method.

[0020] In summary, the present application has at least one of the following beneficial technical effects: By simultaneously analyzing the optical data, thermal imaging data and circuit on-off data of the FPC to be detected, the advantages of different detection modes are fully utilized by means of multi-modal data fusion, and are supplemented and verified with each other. By analyzing the optical data, the intuitive information of the surface of each detection position is obtained. By analyzing the thermal imaging data, the local temperature abnormality of each detection position caused by defects is detected. By analyzing the circuit on-off data, the circuit connection of each detection position is directly reflected. By comprehensively analyzing these multi-modal data, the defects of the FPC to be detected are more accurately recognized and positioned. Even in a complex environment, the false negative rate and the false positive rate are effectively reduced, the reliability of defect detection is enhanced, the quality of FPC products flowing into the market is ensured to meet the requirements, and the service life and stability of the products are ensured.

[0021] The corresponding peripheral interval is dynamically determined through the surface feature parameter, so as to improve the adaptability in the defect detection process, thereby avoiding the missed detection or false detection of some defects due to unreasonable setting of the peripheral interval. The preset peripheral coil is superimposed on the three-dimensional display feature, so as to intuitively combine the preset peripheral coil and the three-dimensional display feature together, thereby intuitively and accurately determining the target peripheral line from the preset peripheral coil, that is, accurately determining the defect detection score corresponding to the detection position. Since each feature vertex in the three-dimensional display feature contains rich defect information, by identifying the defect detection score corresponding to each feature vertex and comprehensively analyzing by using the preset defect type mapping relationship, the defect type of the detection position can be more accurately determined. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 is a flowchart of a multi-modal FPC defect detection method in the embodiment of the present application; Figure 2 is a flowchart of determining a defect type and a defect detection score in the embodiment of the present application; Figure 3 is a structural diagram of a detection system in the embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be further described in detail in combination with the accompanying drawings. Figures 1 to 3 The present application is further described in detail.

[0024] Those skilled in the art can make modifications to the embodiments of the present application without creative contribution after reading the present specification, but as long as the modifications are within the scope of the claims of the present application, they are protected by the patent law.

[0025] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0026] It should be noted that in the optional embodiments of the present application, the data related to the object information and the like needs to be obtained with the permission or consent of the object when the embodiments of the present application are applied to specific products or technologies, and the collection, use and processing of the related data need to comply with the relevant laws, regulations and standards of the country and region. That is, if the embodiments of the present application involve data related to the object, the data needs to be obtained with the authorization and consent of the object, the authorization and consent of the relevant department, and in compliance with the relevant laws, regulations and standards of the country and region. If the embodiments involve personal information, the consent of the individual needs to be obtained for the acquisition of all personal information, and the individual consent needs to be obtained for sensitive information. The embodiments also need to be implemented with the authorization and consent of the object.

[0027] Specifically, the embodiments of the present application provide a multi-modal FPC defect detection method, which is executed by a detection system. The detection system can be a server or a terminal device. The server can be a standalone physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected through wired or wireless communication, and the embodiments of the present application do not limit this.

[0028] Reference Figure 1 , Figure 1 is a flowchart of a multi-modal FPC defect detection method in the embodiments of the present application. The method comprises steps S110-S150, wherein: Step S110: Obtain optical data, thermal imaging data and circuit on-off data corresponding to the FPC to be detected.

[0029] Specifically, the FPC to be inspected is a flexible circuit board that requires defect detection. When acquiring optical data, thermal imaging data, and circuit continuity data of the FPC to be inspected, the FPC can be placed on an inspection platform. The optical data can be acquired by an image acquisition device set around the inspection platform and then uploaded to the inspection system. The image acquisition device can be an optical microscope or a high-resolution camera. To improve the clarity of the optical data, lighting equipment can be used in conjunction with the image acquisition device during image acquisition to avoid shadows or reflections. The thermal imaging data can be acquired by an infrared thermal imager set around the inspection platform and then uploaded to the inspection system. To ensure the accuracy of the thermal imaging data, an infrared imager with high sensitivity, high resolution, and a wide temperature measurement range can be selected. The circuit continuity data can be acquired by a circuit tester set around the inspection platform and then uploaded to the inspection system. The circuit tester can be a flying probe tester, an in-circuit tester, etc. The specific image acquisition device, infrared imager, and circuit tester are not specifically limited in this embodiment.

[0030] Since analyzing optical data facilitates the discovery of surface and structural defects in the FPC under test, analyzing thermal imaging data facilitates the location of thermal anomalies and potential faults in the FPC under test, and analyzing circuit continuity data facilitates the verification of the electrical functional integrity of the FPC under test, it is necessary to adopt a multimodal data fusion approach to comprehensively analyze optical data, thermal imaging data, and circuit continuity data during the defect detection process in order to improve the comprehensiveness and accuracy of defect detection.

[0031] Step S120: Determine the surface feature parameters corresponding to each detection position on the FPC to be inspected from the optical data.

[0032] Specifically, the detection positions on the FPC to be detected can be divided into regions in advance by relevant staff according to the size of the FPC to be detected, and each region is regarded as a detection position. The specific region division manner is not limited in the embodiments of the present application. The surface feature parameters corresponding to each detection position can be identified from the optical data based on a preset feature recognition algorithm. The preset feature recognition algorithm can be an edge detection algorithm, and the specific preset feature recognition algorithm is not limited in the embodiments of the present application. For any detection position, the line boundary corresponding to the detection position can be identified according to the preset edge detection algorithm, and the image texture corresponding to the detection position can be identified through a preset texture recognition algorithm. The image texture recognition result is quantified to obtain the roughness corresponding to the detection region. The linear defects, such as wire edge height or crack length, corresponding to the detection position can also be identified through a preset morphological operation. The surface feature parameters include but are not limited to image texture and crack damage. The preset texture recognition algorithm can be a gray level co-occurrence matrix or a local binary pattern algorithm, and the preset morphological operation can be a dilation operation or an erosion operation. The specific preset texture recognition algorithm and the specific preset morphological operation are not limited in the embodiments of the present application. In this way, the surface feature parameters corresponding to each detection position can be determined.

[0033] Step S130: determining the temperature feature parameters corresponding to each detection position on the FPC to be detected from the thermal imaging data.

[0034] Specifically, the temperature feature parameters include but are not limited to average temperature, extreme temperature, temperature gradient, and hot spot distribution information. For any detection position, the average temperature reflecting the overall thermal state of the divided region corresponding to the detection position can be obtained by calculating the temperature mean value of all pixel points at the detection position. The potential thermal abnormal source in the divided region corresponding to the detection position can be located by identifying the temperature extreme points of each pixel point in the divided region corresponding to the detection position. The temperature gradient can be obtained by calculating the spatial variation rate of the temperature in the divided region corresponding to the detection position, which facilitates the evaluation of the uniformity of heat conduction. The local high-temperature region, i.e., the hot spot, in the divided region corresponding to the detection position can be identified through a preset threshold segmentation algorithm. The hot spot distribution information can be obtained by calculating the area, shape and temperature difference of the surrounding area of the hot spot, which facilitates the quantification of the dispersion degree of the temperature in the divided region corresponding to the detection position, thereby facilitating the evaluation of the consistency of the heat distribution. In this way, the temperature feature parameters corresponding to each detection position can be obtained.

[0035] Step S140: determining the circuit feature parameters corresponding to each detection position on the FPC to be detected from the circuit on-off data.

[0036] Specifically, the circuit characteristic parameters can include, but are not limited to, on-off states, on-off positions, and time domain characteristic parameters. For any detection position, the on-off state corresponding to the detection position, i.e., 0 / 1 state, can be determined from the circuit characteristic parameters through feature recognition. The on-off position corresponds to the detection position. The time domain characteristic parameters include response time and time domain waveform characteristics. The response time can be determined by recording the delay of the circuit response corresponding to the detection position after applying an excitation signal, such as a pulse voltage, which can be used to evaluate the signal transmission efficiency. By extracting the rising edge / falling edge time, overshoot amplitude, ringing period, and other time domain waveform characteristics in the circuit on-off data, the signal integrity can be evaluated. In this way, the circuit characteristic parameters corresponding to each detection position can be determined.

[0037] Step S150: Based on the surface characteristic parameters, temperature characteristic parameters, circuit characteristic parameters corresponding to each detection position on the FPC to be detected, and the preset parameter weight, the defect type and defect detection score corresponding to each detection position on the FPC to be detected are obtained.

[0038] Specifically, after determining the surface characteristic parameters, temperature characteristic parameters, and circuit characteristic parameters corresponding to each detection position, the multi-dimensional parameters can be comprehensively analyzed based on the preset parameter weight, and finally the defects corresponding to each detection position can be quantitatively processed. The defect types include, but are not limited to, short circuit, open circuit, etc. The preset parameter weight can be 3:4:3, i.e., the surface characteristic parameter weight is 0.3, the temperature characteristic parameter weight is 0.4, and the circuit characteristic parameter weight is 0.3. The preset parameter weight can be dynamically adjusted by the relevant technical personnel according to the actual detection requirements. For example, in the scene where thermal abnormalities are focused on, the weight of the temperature characteristic parameter can be increased.

[0039] For the embodiments of the present application, by simultaneously analyzing the optical data, thermal imaging data, and circuit on-off data of the FPC to be detected, the advantages of different detection modes can be fully utilized through multi-modal data fusion, and they can be supplemented and verified. By analyzing the optical data, the intuitive information of the surface of each detection position can be obtained. By analyzing the thermal imaging data, the local temperature abnormality caused by defects at each detection position can be detected. By analyzing the circuit on-off data, the circuit connection of each detection position can be directly reflected. By comprehensively analyzing these multi-modal data, the defects of the FPC to be detected can be more accurately recognized and located. Even in complex environments, the false negative and false positive rates can be effectively reduced, the reliability of defect detection can be enhanced, the quality of FPC products flowing into the market can be ensured to meet the requirements, and the service life and stability of the products can be ensured.

[0040] Further, in order to intuitively and accurately determine the defect type and defect detection score corresponding to the detection position, when the surface feature parameter, temperature feature parameter, circuit feature parameter corresponding to the detection position and the preset parameter weight are used to obtain the defect type and defect detection score corresponding to the detection position, the following steps can be specifically included: The surface feature parameter, temperature feature parameter and circuit feature parameter corresponding to the detection position are normalized to obtain surface feature values, temperature feature values and circuit feature values; the surface feature coordinate values, temperature feature coordinate values and circuit feature coordinate values corresponding to the surface feature values, temperature feature values and circuit feature values are determined according to the preset parameter weight; the three-dimensional display feature is determined according to the surface feature coordinate values, temperature feature coordinate values and circuit feature coordinate values, and the defect type and defect detection score corresponding to the detection position are determined based on the three-dimensional display feature and the preset peripheral coil.

[0041] Specifically, the surface feature parameter, temperature feature parameter and circuit feature parameter can be normalized according to Min-Max standardization or Z-score standardization, so as to eliminate the influence of the dimensional difference between different feature parameters on subsequent defect detection analysis. The way of normalizing the plurality of surface feature parameters is not specifically limited in the embodiments of the present application, as long as the surface feature values, temperature feature values and circuit feature values obtained after normalization belong to the same dimension, and each feature parameter can be compared or fused under the same numerical scale.

[0042] The three normalized feature values obtained after normalization of the detection position can be surface feature values S, temperature feature values T and circuit feature values C. When the preset parameter weight is 3:4:3, the three-dimensional coordinate values corresponding to the detection position can be obtained by multiplying each normalized feature value by the corresponding weight. The surface feature coordinate value X = 0.3 * S. The temperature feature coordinate value Y = 0.4 * T. The circuit feature coordinate value Z = 0.3 * C.

[0043] By introducing the preset parameter weight, the importance of different feature parameters in defect detection can be reflected. By applying it to the conversion of feature values, the key feature parameters can occupy a more prominent position in the feature coordinate system in defect detection, so as to better reflect the influence of each feature parameter on the defect detection result. Finally, through the three-dimensional display feature and the preset peripheral coil, the defect type and defect detection score corresponding to the detection position can be intuitively and accurately determined. The three-dimensional coordinate values corresponding to the detection position are introduced into the preset three-dimensional coordinate system to obtain the three-dimensional display feature corresponding to the detection position.

[0044] The preset peripheral coil includes a plurality of peripheral lines, each of which corresponds to a different defect detection score. By superimposing the preset peripheral coil and the stereoscopic display feature, the target peripheral line corresponding to the stereoscopic display feature can be determined intuitively, and the defect detection score corresponding to the target peripheral line is determined as the defect detection score corresponding to the detection position. The defect detection score corresponding to each peripheral line in the preset peripheral coil can be determined by relevant staff according to historical experimental data and uploaded to the detection system. Since each feature vertex in the stereoscopic display feature contains rich defect information, the defect type of the detection position can be accurately determined by identifying the defect detection score corresponding to each feature vertex. The defect type corresponding to the combination of the defect detection scores corresponding to different feature vertices is different.

[0045] Further, the application provides a specific process for determining the defect type and defect detection score corresponding to the detection position based on the stereoscopic display feature and the preset peripheral line, including steps S210-S240, as shown in Figure 2 Step S210: determining the peripheral interval corresponding to the detection position according to the surface feature parameter corresponding to the detection position and the preset peripheral interval mapping relationship. The preset peripheral interval mapping relationship is the corresponding relationship between the surface feature parameter and the peripheral interval.

[0046] Specifically, since the surface feature parameter can reflect the damage degree of the detection position, and the intuitive damage condition has a greater impact on the defect detection result, it is necessary to analyze the surface feature parameter corresponding to the detection position and determine the peripheral interval corresponding to the detection position based on this. The preset peripheral interval mapping relationship is the corresponding relationship between the surface feature parameter and the peripheral interval. Based on this mapping relationship, the peripheral interval corresponding to any surface feature parameter can be determined. The specific content of the mapping relationship is not limited in the embodiments of the application and can be determined by relevant staff according to historical experimental data and uploaded to the detection system.

[0047] Step S220: determining the final preset peripheral coil based on the peripheral interval, and superimposing the preset peripheral coil and the stereoscopic display feature to obtain a coil superimposed display feature map.

[0048] ​Specifically, the interval between each of the peripheral coils in the initial preset peripheral coil can be adjusted according to the determined peripheral interval to obtain a final preset peripheral coil. Finally, the final preset peripheral coil can be combined with the stereoscopic display feature through a preset feature superposition method, and the contact condition of each feature vertex in the stereoscopic display feature and the final preset peripheral coil can be directly observed through the coil superposition display feature map. The preset feature superposition method can be Boolean operation, and the specific feature superposition algorithm is not limited in the embodiments of the present application. In addition, the contact condition between the final peripheral coil and the stereoscopic display feature can also be directly observed by setting the transparency or color of the coil. The specific transparency or color is not limited in the embodiments of the present application, and can be set by a relevant technical personnel according to actual needs.

[0049] Step S230: identifying a target peripheral line from the coil superposition display feature map, and determining a defect detection score corresponding to the detection position according to a preset defect score mapping relationship, wherein the peripheral line surrounding the stereoscopic display feature is determined as the target peripheral line, and the preset defect score mapping relationship is the corresponding relationship between the detection position and the defect detection score.

[0050] Specifically, since the stereoscopic display feature can include multiple feature vertices, for example, the stereoscopic display feature includes three feature vertices, namely vertex a, vertex b and vertex c, wherein vertex a is in contact with peripheral line 1, vertex b is in contact with peripheral line 2, and vertex c is in contact with peripheral line 3, wherein the diameter corresponding to peripheral line 1 is greater than the diameter corresponding to peripheral line 2, and is also greater than the diameter corresponding to peripheral line 3, that is, peripheral line 1 can surround the entire stereoscopic display feature. At this time, peripheral line 1 can be determined as the target peripheral line, and the defect detection score corresponding to the target peripheral line can be determined according to the preset defect score mapping relationship, wherein the preset defect score mapping relationship includes the defect detection scores corresponding to various target peripheral lines, and the specific content is not limited in the embodiments of the present application, and can be determined by a relevant staff according to historical experimental data and uploaded to the detection system.

[0051] Step S240: identifying the defect detection score corresponding to each feature vertex in the stereoscopic display feature from the coil superposition display feature map, and determining the defect type corresponding to the detection position according to the defect detection score corresponding to each feature vertex and a preset defect type mapping relationship, wherein the preset defect type mapping relationship is the corresponding relationship between the score combination of the defect detection scores corresponding to each feature vertex and the defect type.

[0052] Specifically, when determining the defect type corresponding to the detection position, each feature vertex can be identified from the coil display feature map to contact or the closest peripheral line, a score combination is generated according to the identification result, and finally the defect type corresponding to the score combination is determined based on the preset defect type mapping relationship. The preset defect type mapping relationship contains the defect type corresponding to various score combinations, and the specific content of the preset defect type mapping relationship is not limited in the embodiments of the present application, which can be determined by relevant staff according to historical experimental data and uploaded to the detection system.

[0053] For the embodiments of the present application, since different surface feature parameters have different influences on the defect detection result, the corresponding peripheral interval is dynamically determined by the surface feature parameter, which facilitates to improve the adaptability in the defect detection process, thereby avoiding missed detection or false detection of some defects due to unreasonable setting of the peripheral interval. By superimposing the preset peripheral coil on the three-dimensional display feature, the preset peripheral coil and the three-dimensional display feature can be intuitively combined together, so that the target peripheral line can be determined from the preset peripheral coil, that is, the defect detection score corresponding to the detection position can be accurately determined. Since each feature vertex in the three-dimensional display feature contains rich defect information, by identifying the defect detection score corresponding to each feature vertex and comprehensively analyzing by using the preset defect type mapping relationship, the defect type of the detection position can be more accurately determined.

[0054] Further, in order to facilitate the improvement of the accuracy of defect detection, the circuit feature parameter corresponding to any detection position on the FPC to be detected is determined from the circuit on-off data, which can specifically include: The on-off state, on-off position, and time domain feature parameter corresponding to the detection position are identified from the circuit on-off data. When the time domain feature parameter contains a preset time domain feature, the influence position and influence parameter value are determined based on the preset time domain feature. The actual parameter value corresponding to the influence position is obtained, and when the parameter difference between the actual parameter value and the influence parameter value is lower than the preset difference threshold, the time domain feature parameter is optimized according to the influence parameter value to obtain the optimized time domain feature parameter. The circuit feature parameter corresponding to the detection position is determined based on the on-off state, on-off position, and optimized time domain feature parameter.

[0055] Specifically, the on-off state, on-off position and time domain characteristic parameter corresponding to the detection position can be determined from the circuit on-off data by a preset characteristic identification algorithm, and the specific preset characteristic identification algorithm is not specifically limited in the embodiments of the present application. The preset time domain characteristic can be determined by relevant personnel according to historical experimental data and uploaded to the detection system. When the time domain characteristic parameter corresponding to the detection position contains the preset time domain characteristic, it indicates that the detection position may have a short circuit risk, for example, the rise time is lower than the preset rise threshold, the resistance fluctuation rate is higher than the preset fluctuation rate threshold, the overshoot amplitude is higher than the preset overshoot threshold, and the stable time is lower than the preset time length. At this time, it can only be concluded that there is a short circuit risk, but it is not directly determined that there is a short circuit at the detection position. Among them, the specific preset rise threshold, preset fluctuation rate threshold, preset overshoot threshold and preset time length are not specifically limited in the embodiments of the present application, and can be determined by relevant personnel according to historical experimental data and uploaded to the detection system.

[0056] In the embodiments of the present application, a verification method is provided, that is, after it is determined that the detection position has a short circuit risk, the position and the influence degree that may be affected after the detection position has a short circuit can be predicted according to the preset time domain characteristic contained in the detection position, that is, the influence position and the influence parameter value, and then the actual parameter value of the influence position is obtained. Finally, by comparing the influence parameter value with the actual parameter value, it is judged whether the verification is passed. When the verification is passed, it can be determined that there is a short circuit at the detection position. When the parameter difference between the actual parameter value and the influence parameter value is lower than the preset difference threshold, it can be determined that the verification is passed. The specific preset difference threshold is not specifically limited in the embodiments of the present application, and can be determined by relevant personnel according to historical experimental data. For example, if the time domain characteristic of detection position A meets the short circuit risk, and the deviation between the time domain characteristic of downstream node detection position B and the predicted value is >20%, the short circuit of detection position A is verified. The influence parameter can be voltage, current and other parameters, as long as it can indicate whether the influence position is in a stable running state. The specific influence parameter is not specifically limited in the embodiments of the present application, and can be determined by relevant personnel according to historical experimental data and uploaded to the detection system.

[0057] After the verification is passed, the time domain characteristic parameter corresponding to the detection position can be optimized and adjusted based on the influence parameter value, for example, the time domain characteristic parameter value is increased to obtain the optimized time domain characteristic parameter. The optimized time domain characteristic parameter can more accurately reflect the dynamic performance and potential problems of the detection position, thereby facilitating the improvement of the precision of defect detection. The specific optimization and adjustment method or optimization and adjustment amount is not specifically limited in the embodiments of the present application, and can be determined by relevant personnel according to historical experimental data and uploaded to the detection system.

[0058] Further, in order to facilitate the accuracy of determining the defect detection score, when the surface feature parameter of the detection position contains the preset surface feature, the method can specifically include: determining the detection position containing the preset surface feature as an abnormal detection position; counting the historical abnormal frequency and the historical failure probability corresponding to the abnormal detection position from the historical detection record; and adjusting the preset parameter weight corresponding to the abnormal detection position based on the historical abnormal frequency and the historical failure probability.

[0059] Specifically, the preset surface feature can be a typical scratch, a typical crack, or a typical pit that is prone to cause failure, and the specific preset surface feature is not limited in the embodiments of the present application and can be determined by relevant staff according to historical experimental data and uploaded to the detection system. When the surface feature parameter corresponding to the detection position contains the preset surface feature, the detection position can be determined as an abnormal detection position. The historical detection record is the defect detection score and the defect detection type of each detection position on all FPCs to be detected within a period of time before the current time. The similar historical detection position corresponding to the abnormal detection position is located from the historical detection data, and the frequency of abnormal defects and the probability of actual failure of the similar historical detection position within the historical time period are obtained, wherein the presence of the preset surface feature does not represent that the abnormal detection position actually fails, but has a high probability of failure.

[0060] After the historical abnormal frequency and the historical abnormal failure probability are determined, the preset weight adjustment parameter mapping relationship can be used to determine the weight adjustment value corresponding to the abnormal detection position, wherein the preset weight adjustment parameter mapping relationship is a parameter combination of the historical abnormal frequency and the historical abnormal failure probability, and the corresponding relationship between the weight adjustment value, and the specific content of the preset weight adjustment parameter mapping relationship is not limited in the embodiments of the present application and can be determined by relevant staff according to historical experimental data and uploaded to the detection system.

[0061] Further, in order to facilitate the provision of process improvement clues for relevant staff, the method provided in the embodiments of the present application further includes: performing cluster analysis on the defect type and the defect detection score of each detection position on each FPC to be detected within a preset time period; determining whether there is a similar defect group on different FPCs to be detected based on the cluster analysis result, the similar defect group containing at least two same detection positions from different FPCs to be detected, and the defect score difference between the defect detection scores corresponding to the at least two same detection positions being lower than a preset score threshold; and if so, generating process feedback information based on the similar defect group.

[0062] Specifically, the preset time period is a period of time before the current time, and the duration corresponding to the preset time period can be 24 hours or 48 hours, and the specific duration is not limited in the embodiments of the present application. When performing clustering analysis on the defect types and defect detection scores corresponding to each detection position, each detection position can be grouped first, and then each detection position is clustered individually. The clustering parameter K-Means can be set according to actual detection requirements, for example, the number of clusters K=3 can be determined according to historical clustering data. After clustering analysis according to the set clustering parameter, the clustering labels corresponding to each detection position can be obtained, such as cluster 0, cluster 1, cluster 2, and the center point of the typical defect feature representing each cluster. The specific clustering process is not limited in the embodiments of the present application.

[0063] Based on the clustering analysis result, the same detection positions located on different FPCs to be detected can be identified, and the defect score difference between the defect detection scores corresponding to the same detection positions is not less than the preset score threshold. For example, the defect detection score corresponding to the detection position 1a located on the FPC1 to be detected is 85, the defect detection score corresponding to the detection position 2a located on the FPC2 to be detected is 82, and the defect score difference between the two is 3, which is lower than the preset score threshold 5. At this time, the detection position 1a located on the FPC1 to be detected and the detection position 2a located on the FPC2 to be detected can be determined as a similar defect group. By performing clustering analysis on the defect types and defect detection scores corresponding to each detection position on each FPC to be detected within a period of time, potential correlations that can exist on different FPCs to be detected can be effectively mined. Process feedback information is generated through these potential correlations, which facilitates providing process improvement clues for relevant personnel.

[0064] An embodiment of the present application provides a detection system, as shown in Figure 3 , as shown in Figure 3 The detection system 300 shown in the figure includes a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the detection system 300 can also include a transceiver 304. It should be noted that the transceiver 304 is not limited to one in actual application, and the structure of the detection system 300 does not constitute a limitation on the embodiments of the present application.

[0065] The processor 301 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in connection with the disclosure. The processor 301 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0066] The bus 302 can include a path for transmitting information between the above-mentioned components. The bus 302 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 302 can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 3 Only one line is used in the middle, but it does not mean that there is only one bus or one type of bus.

[0067] The memory 303 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, an optical disk storage (including a compact disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and capable of being accessed by a computer, but not limited to this.

[0068] The memory 303 is configured to store application program codes for implementing the solutions of the present application, and the processor 301 is configured to control the execution of the application program codes stored in the memory 303. The processor 301 is configured to execute the application program codes stored in the memory 303 to implement the content shown in the foregoing method embodiments.

[0069] The detection system includes, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Personal Computers), PMPs (Portable Multimedia Players), vehicle terminals (for example, vehicle navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. The detection system can also be a server or the like. Figure 3 The detection system shown is only an example, and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0070] The embodiments of the present application provide a computer readable storage medium, which stores a computer program. When the computer program is run on a computer, the computer can execute the corresponding content in the foregoing method embodiments.

[0071] The embodiments of the present application provide a computer program product, which includes a computer program. When the computer program is executed by a processor, the method in any of the foregoing embodiments is implemented.

[0072] It should be understood that, although each step in the flowchart of the accompanying drawings is shown in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other sequences. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed in rotation or alternation with at least part of other steps or sub-steps or stages of other steps.

[0073] The above only describes some embodiments of the present application. It should be noted that, for those skilled in the art, without departing from the principles of the present application, several improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A multi-modal FPC defect detection method, characterized by, The method comprises the following steps: obtaining optical data, thermal imaging data and circuit on-off data corresponding to the FPC to be detected; determining surface feature parameters corresponding to each detection position on the FPC to be detected from the optical data; determining temperature feature parameters corresponding to each detection position on the FPC to be detected from the thermal imaging data; determining circuit feature parameters corresponding to each detection position on the FPC to be detected from the circuit on-off data; based on the surface feature parameters, temperature feature parameters, circuit feature parameters corresponding to each detection position on the FPC to be detected and preset parameter weights, obtaining the defect type and defect detection score corresponding to each detection position on the FPC to be detected.

2. The multi-modal FPC defect detection method of claim 1, wherein, Based on the surface feature parameters, temperature feature parameters, circuit feature parameters corresponding to each detection position and the preset parameter weights, the defect type and defect detection score corresponding to each detection position are obtained, which comprises: normalizing the surface feature parameters, temperature feature parameters and circuit feature parameters corresponding to the detection position to obtain surface feature values, temperature feature values and circuit feature values; determining surface feature coordinate values, temperature feature coordinate values and circuit feature coordinate values corresponding to the surface feature values, temperature feature values and circuit feature values according to the preset parameter weights; determining a three-dimensional display feature according to the surface feature coordinate values, temperature feature coordinate values and circuit feature coordinate values, and determining the defect type and defect detection score corresponding to the detection position based on the three-dimensional display feature and a preset peripheral coil.

3. The multi-modal FPC defect detection method of claim 2, wherein, The determination of the defect type and defect detection score corresponding to the detection position based on the three-dimensional display feature and the preset peripheral coil further comprises: determining a peripheral interval corresponding to the detection position according to the surface feature parameters corresponding to the detection position and a preset peripheral interval mapping relationship, wherein the preset peripheral interval mapping relationship is a corresponding relationship between surface feature parameters and peripheral intervals; determining a final preset peripheral coil based on the peripheral interval, and superimposing the preset peripheral coil and the three-dimensional display feature to obtain a coil superimposed display feature map; identifying a target peripheral line from the coil superimposed display feature map, and determining the defect detection score corresponding to the detection position according to the target peripheral line and a preset defect score mapping relationship, wherein the peripheral line surrounding the three-dimensional display feature is determined as the target peripheral line, and the preset defect score mapping relationship is a corresponding relationship between detection positions and defect detection scores; identifying the defect detection score corresponding to each feature vertex in the three-dimensional display feature from the coil superimposed display feature map, and determining the defect type corresponding to the detection position according to the defect detection score corresponding to each feature vertex and a preset defect type mapping relationship, wherein the preset defect type mapping relationship is a corresponding relationship between the score combination of the defect detection scores corresponding to each feature vertex and the defect type.

4. The multi-modal FPC defect detection method of claim 1, wherein, The determination of the circuit feature parameters corresponding to any detection position on the FPC to be detected from the circuit on-off data comprises: identify, from the circuit on-off data, an on-off state, an on-off position, and a time-domain characteristic parameter corresponding to the detection position; when the time-domain characteristic parameter contains a preset time-domain characteristic, determine an influence position and an influence parameter value based on the preset time-domain characteristic; obtain an actual parameter value corresponding to the influence position, and when a parameter difference between the actual parameter value and the influence parameter value is lower than a preset difference threshold, optimize the time-domain characteristic parameter according to the influence parameter value to obtain an optimized time-domain characteristic parameter; determine a circuit characteristic parameter corresponding to the detection position based on the on-off state, the on-off position, and the optimized time-domain characteristic parameter.

5. The multi-modal FPC defect detection method of claim 1, wherein, when the surface characteristic parameter of the detection position contains a preset surface characteristic, further comprising: determine the detection position containing the preset surface characteristic as an abnormal detection position; statistically obtain a historical abnormal frequency and a historical failure probability corresponding to the abnormal detection position from historical detection records; adjust a preset parameter weight corresponding to the abnormal detection position based on the historical abnormal frequency and the historical failure probability.

6. The multi-modal FPC defect detection method of claim 1, wherein, further comprising: perform clustering analysis on a defect type and a defect detection score corresponding to each detection position on each FPC to be detected within a preset time period; determine whether there is a similar defect group on different FPCs to be detected based on the clustering analysis result, the similar defect group containing at least two same detection positions from different FPCs to be detected, and a defect score difference between defect detection scores corresponding to the at least two same detection positions being lower than a preset score threshold; if yes, generate process feedback information based on the similar defect group.

7. A detection system characterized by, The detection system comprises: at least one processor; a memory; at least one application program, wherein the at least one application program is stored in the memory and is configured to be executed by the at least one processor, and the at least one application program is configured to execute the multi-modal FPC defect detection method of any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, comprising: a computer program stored in the memory and capable of being loaded and executed by the processor to implement the multi-modal FPC defect detection method of any one of claims 1-6.

9. A computer program product, characterised in that, comprising a computer program, which, when executed by the processor, implements the steps of the multi-modal FPC defect detection method of any one of claims 1-6.

Images