High-speed wire harness low-heat-loss welding method and system based on welding spot morphology recognition

Abstract

The invention relates to the technical field of high-speed wire harness welding, in particular to a high-speed wire harness low-heat-loss welding method and system based on welding spot morphology recognizing.The method comprises the steps that a target high-speed wire harness and wire harness welding window duration are determined, the target high-speed wire harness is welded, a controllable welding wire harness is obtained, welding spot shooting is conducted on the controllable welding wire harness, a wire harness morphology image is obtained, and the welding spot morphology image is obtained; and performing welding process evaluation on the wire harness morphology image to obtain a current welding process, regulating and controlling the original laser welding equipment according to the current welding process and the wire harness welding window duration to obtain target laser welding equipment, and returning to the welding step until a welding window ending instruction is received. The automation degree of the wire harness welding process can be improved, and the thermal damage risk of the wire harness is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high-speed wire harness welding, and in particular to a high-speed wire harness low-heat-loss welding method and system based on weld appearance identification. BACKGROUND

[0002] High-speed wire harness welding technology plays a key role in high-end fields such as electronic devices, automobile manufacturing, and aerospace, and its welding quality directly affects the stability of signal transmission, the reliability of device operation, and the safety of the overall system. With the improvement of industrial automation level, the control of heat input during welding process is increasingly strict to avoid material degradation or performance decline caused by overheating, thereby ensuring the long-term durability of high-speed wire harness in high-speed data transmission scenarios.

[0003] Traditional welding methods usually rely on operators to manually adjust welding current based on experience to achieve rough control of the weld. However, this method lacks real-time detection and feedback mechanism for weld appearance, which easily leads to mismatch between welding parameters and actual weld state, causing uneven heat input or excessive accumulation, and further causing defects such as damage to wire harness insulation layer or virtual welding of weld. SUMMARY

[0004] The present application provides a high-speed wire harness low-heat-loss welding method based on weld appearance identification and a computer readable storage medium, which aims to improve the automation level of wire harness welding process and reduce the risk of heat damage to wire harness.

[0005] To achieve the above-mentioned purpose, the present application provides a high-speed wire harness low-heat-loss welding method based on weld appearance identification, which comprises: determining the target high-speed wire harness and the wire harness welding window duration, and constructing a low-heat-loss welding device, wherein the low-heat-loss welding device comprises an original laser welding equipment and a weld appearance detection equipment; welding the target high-speed wire harness based on the original laser welding equipment to obtain a controllable welded wire harness, and using the weld appearance detection equipment to take pictures of the welds of the controllable welded wire harness to obtain a wire harness appearance image, wherein the wire harness appearance image is an image in YCbCr color space, and the wire harness appearance image includes a plurality of original wire harness pixel points; selecting seed points from the wire harness appearance image to obtain a target seed point group, wherein the target seed point group includes a plurality of target seed points; segmenting the weld area based on the target seed point group to obtain a plurality of wire harness segmentation areas, and evaluating the welding process according to the plurality of wire harness segmentation areas to obtain a current welding process; controlling the original laser welding equipment according to the current welding process and the wire harness welding window duration to obtain a target laser welding equipment; The target laser welding device and the controllable welding harness are taken as an original laser welding device and a target high-speed harness respectively, and the step of welding the target high-speed harness based on the original laser welding device is returned until a preset welding window end instruction is received, and the high-speed harness low-heat-loss welding based on the weld appearance recognition is completed.

[0006] Optionally, the seed point selection on the harness appearance image to obtain the target seed point group comprises: Original harness pixel points in the harness appearance image are sequentially extracted, and the extracted original harness pixel points are recorded as center harness pixel points: A set of adjacent harness pixel points of the center harness pixel points in the harness appearance image is determined, wherein the set of adjacent harness pixel points comprises a plurality of adjacent harness pixel points; The maximum distance value and the total neighborhood standard deviation are calculated according to the set of adjacent harness pixel points; The total neighborhood standard deviation and the maximum distance value corresponding to each original harness pixel point are respectively summarized to obtain a plurality of total neighborhood standard deviations and a plurality of maximum distance values; Pixel difference values are obtained by performing pixel difference evaluation on the plurality of total neighborhood standard deviations and the plurality of original harness pixel points; The target seed point group is selected in the harness appearance image according to the plurality of maximum distance values and the plurality of pixel difference values.

[0007] Optionally, the calculation of the maximum distance value and the total neighborhood standard deviation according to the set of adjacent harness pixel points comprises: Adjacent harness pixel points in the set of adjacent harness pixel points are sequentially extracted, and distance calculation is performed on the center harness pixel points based on the extracted adjacent harness pixel points to obtain channel distance values, and the channel distance values are summarized to obtain a plurality of channel distance values; The maximum distance value in the plurality of channel distance values is identified; A set of neighborhood channel standard deviations is constructed based on the set of adjacent harness pixel points, wherein the set of neighborhood channel standard deviations comprises a neighborhood Y channel standard deviation, a neighborhood Cb channel standard deviation and a neighborhood Cr channel standard deviation; The set of neighborhood channel standard deviations is summed to obtain the total neighborhood standard deviation.

[0008] Optionally, the pixel difference evaluation on the plurality of total neighborhood standard deviations and the plurality of original harness pixel points to obtain the plurality of pixel difference values comprises: A global maximum standard deviation is identified in the plurality of total neighborhood standard deviations, and the plurality of total neighborhood standard deviations are normalized by using the global maximum standard deviation to obtain a plurality of neighborhood normalized standard deviations; The plurality of original harness pixel points are recorded as a plurality of difference center pixel points, and the following operations are performed on each difference center pixel point in the plurality of difference center pixel points: obtaining a plurality of difference neighboring pixel points of the difference center pixel point, identifying a center total standard deviation corresponding to the difference center pixel point and a plurality of difference neighboring standard deviations corresponding to the plurality of difference neighboring pixel points in the plurality of neighborhood normalized standard deviations respectively; calculating a pixel difference value based on the center total standard deviation and the plurality of difference neighboring standard deviations; summarizing the pixel difference value to obtain a plurality of pixel difference values.

[0009] Optionally, the selecting the target seed point group from the wire bundle topographic image according to the plurality of maximum distance values and the plurality of pixel difference values comprises: determining a channel distance threshold and a pixel difference threshold according to the plurality of maximum distance values and the plurality of pixel difference values; marking a plurality of original wire bundle pixel points in the wire bundle topographic image as a plurality of candidate seed points; performing the following operation on each candidate seed point in the plurality of candidate seed points: determining a candidate distance value and a candidate difference value corresponding to the candidate seed point in the plurality of maximum distance values and the plurality of pixel difference values respectively; if the candidate distance value is greater than the channel distance threshold and the candidate difference value is less than the pixel difference threshold, marking the candidate seed point as an effective seed point; summarizing the effective seed points to obtain a plurality of effective seed points, performing connectivity identification on the plurality of effective seed points to obtain a plurality of adjacent seed point groups, wherein each adjacent seed point group comprises one or more adjacent seed points; extracting adjacent seed point groups from the plurality of adjacent seed point groups in sequence, obtaining a seed difference value group of the extracted adjacent seed point group, wherein the seed difference value group comprises a plurality of seed difference values, and each seed difference value corresponds to an adjacent seed point; determining a minimum difference value in the seed difference value group, and marking the adjacent seed point corresponding to the minimum difference value as a target seed point; summarizing the target seed points corresponding to each adjacent seed point group to obtain the target seed point group.

[0010] Optionally, the performing weld point region segmentation based on the target seed point group to obtain a plurality of wire bundle segmentation regions comprises: extracting target seed points from the target seed point group in sequence, and dividing an initial segmentation region of the extracted target seed point in the wire bundle topographic image; summarizing the initial segmentation region corresponding to each target seed point to obtain a plurality of initial segmentation regions; performing the following operation on each initial segmentation region in the plurality of initial segmentation regions: determining a region neighboring pixel point set of the initial segmentation region in the wire bundle topographic image, and dividing the region neighboring pixel point set to the initial segmentation region to obtain an updated segmentation region; record the updated segmentation region as an initial segmentation region, and return the step of determining a region adjacent pixel set of the initial segmentation region in the online beam topography image until the region adjacent pixel set is a preset empty set; When the region adjacent pixel set is an empty set, record the updated segmentation region as a beam segmentation region; aggregate the beam segmentation regions corresponding to the initial segmentation region to obtain a plurality of beam segmentation regions.

[0011] Optionally, the step of determining the region adjacent pixel set of the initial segmentation region in the online beam topography image comprises: determining an original adjacent pixel set of the initial segmentation region in the online beam topography image, wherein the original adjacent pixel set comprises a plurality of original adjacent pixels; calculating an initial segmentation channel value group based on the original adjacent pixel set; performing the following operations on each original adjacent pixel in the original adjacent pixel set: performing channel distance calculation on the original adjacent pixel based on the initial segmentation channel value group to obtain an original channel distance value; judging whether the original adjacent pixel has been segmented; if the original adjacent pixel has been segmented, obtaining a segmented region of the original adjacent pixel, and identifying a segmented distance value between the original adjacent pixel and the segmented region, if the segmented distance value is greater than the original channel distance value, recording the original adjacent pixel as a region adjacent pixel; if the original adjacent pixel has not been segmented, recording the original adjacent pixel as a region adjacent pixel; aggregating the region adjacent pixels to obtain the region adjacent pixel set.

[0012] Optionally, the step of performing welding process evaluation according to the plurality of beam segmentation regions to obtain a current welding process comprises: obtaining a welding completion template based on the target high-speed beam; performing template comparison on the plurality of beam segmentation regions using the welding completion template to obtain a plurality of template similarities, wherein the template similarities correspond one-to-one to the beam segmentation regions; identifying a completion similarity group in the plurality of template similarities according to a preset similarity threshold, wherein the completion similarity group comprises one or more completion similarities, or the completion similarity group is an empty set; determining a welding completion region group corresponding to the completion similarity group, and counting a welding completion area of the welding completion region group; calculating the current welding process according to the welding completion area and a preset current total welding area.

[0013] Optionally, the original laser welding equipment is regulated according to the current welding process and the wire harness welding window duration to obtain a target laser welding equipment, comprising: The target high-speed wire harness is welded to obtain a standard welding completed heat and a constraint welding current; The current welding window duration is recorded, and the remaining welding window duration is calculated according to the current welding window duration and the wire harness welding window duration; The remaining welding heat is determined based on the current welding process and the standard welding completed heat, and the adjusted welding current is calculated according to the remaining welding heat, a preset rated welding voltage and the remaining welding window duration; If the adjusted welding current is not greater than the constraint welding current, the adjusted welding current is recorded as a target welding current; If the adjusted welding current is greater than the constraint welding current, the constraint welding current is recorded as the target welding current; The original laser welding equipment is updated by using the target welding current to obtain a target laser welding equipment.

[0014] To achieve the above object, the application further provides a high-speed wire harness low-heat-loss welding system based on welding spot morphology identification, comprising: A high-speed wire harness determination module is configured to determine a target high-speed wire harness and a wire harness welding window duration, and construct a low-heat-loss welding device, wherein the low-heat-loss welding device comprises an original laser welding equipment and a welding spot morphology detection equipment; A wire harness image shooting module is configured to weld the target high-speed wire harness based on the original laser welding equipment to obtain a controllable welded wire harness, and shoot the welding spot of the controllable welded wire harness by using the welding spot morphology detection equipment to obtain a wire harness morphology image, wherein the wire harness morphology image is an image in a YCbCr color space, and the wire harness morphology image comprises a plurality of original wire harness pixel points; A welding process estimation module is configured to select seed points from the wire harness morphology image to obtain a target seed point group, wherein the target seed point group comprises a plurality of target seed points, perform welding spot region segmentation based on the target seed point group to obtain a plurality of wire harness segmentation regions, and perform welding process evaluation according to the plurality of wire harness segmentation regions to obtain a current welding process; A welding equipment regulation module is configured to regulate the original laser welding equipment according to the current welding process and the wire harness welding window duration to obtain a target laser welding equipment, and take the target laser welding equipment and the controllable welded wire harness as the original laser welding equipment and the target high-speed wire harness respectively, and return to the step of welding the target high-speed wire harness based on the original laser welding equipment until a preset welding window end instruction is received.

[0015] To solve the above problems, the application further provides an electronic device, comprising: A memory is configured to store at least one instruction; The processor executes instructions stored in the memory to implement the high-speed wire harness low-heat damage welding method based on solder joint topography recognition described above.

[0016] To solve the above problems, the application also provides a computer readable storage medium, which stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the high-speed wire harness low-heat damage welding method based on solder joint topography recognition described above.

[0017] To solve the problems described in the background art, first, the target high-speed wire harness and the wire harness welding window duration are determined. This step calculates the wire harness welding window duration through historical data. Compared with the method of relying on experience to set the welding time in the prior art, the welding process can be more accurately controlled, heat damage caused by improper duration is avoided, and the rationality of the welding parameters is improved. Then, the seed point selection is performed on the wire harness topography image to obtain a target seed point group. This step selects the seed point by comprehensively calculating the maximum distance value and the pixel difference value. Compared with the traditional single threshold method, the solder joint area boundary can be more accurately identified, the initial error of the region growing algorithm is reduced, and the segmentation accuracy is improved. Further, the solder joint region segmentation is performed based on the target seed point group to obtain a plurality of wire harness segmentation regions. The welding process evaluation is performed according to the plurality of wire harness segmentation regions to obtain the current welding process. This step uses the region growing algorithm to perform solder joint segmentation and combines template comparison to evaluate the welding process. Compared with the existing method relying on artificial visual inspection, the automation of the welding progress is realized, the intelligent level of the welding quality control is improved, and finally, the original laser welding equipment is regulated and controlled according to the current welding process and the wire harness welding window duration to obtain the target laser welding equipment. This step dynamically adjusts the welding current according to the remaining welding heat and the window duration. Compared with the fixed parameter welding technology, the welding quality can be guaranteed while being self-adapted, the risk of overheating is effectively reduced, and compared with the existing one-time welding method, the welding parameters are continuously corrected, and the stability of the entire welding process is enhanced. Therefore, the application can improve the automation degree of the wire harness welding process and reduce the risk of heat damage to the wire harness. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 The flowchart of the high-speed wire harness low-heat damage welding method based on solder joint topography recognition provided by an embodiment of the application is shown. Figure 2 The functional module diagram of the high-speed wire harness low-heat damage welding system based on solder joint topography recognition provided by an embodiment of the application is shown. Figure 3 The structural diagram of the electronic device for implementing the high-speed wire harness low-heat damage welding method based on solder joint topography recognition provided by an embodiment of the application is shown.

[0019] Explanation of reference signs: 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0020] The implementation, functional features and advantages of the present application will be further explained with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein merely exemplify the application and are not intended to limit the application.

[0022] Embodiments of the present application provide a high-speed wire harness low-heat-loss welding method based on weld point topography identification. The execution subject of the high-speed wire harness low-heat-loss welding method based on weld point topography identification includes but is not limited to at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiments of the present application. In other words, the high-speed wire harness low-heat-loss welding method based on weld point topography identification can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0023] Referring to Figure 1 Fig. 1 shows a flowchart of the high-speed wire harness low-heat-loss welding method based on weld point topography identification provided by an embodiment of the present application. In this embodiment, the high-speed wire harness low-heat-loss welding method based on weld point topography identification includes: S1, determining a target high-speed wire harness and a wire harness welding window duration, and constructing a low-heat-loss welding device, wherein the low-heat-loss welding device includes an original laser welding device and a weld point topography detection device.

[0024] It can be understood that the target high-speed wire harness refers to a wire harness that needs to be welded. The wire harness welding window duration refers to the duration of welding a certain welding point in the target high-speed wire harness. In order to ensure that other target high-speed wire harnesses can be welded in time, the duration of welding a certain welding point of the target high-speed wire harness by the subsequent low-heat-loss welding device does not exceed the wire harness welding window duration. The setting method of the wire harness welding window duration is as follows: record the historical welding durations of a plurality of historical high-speed wire harnesses in the past period or in the laboratory environment, wherein the model of the historical high-speed wire harness is the same as that of the target high-speed wire harness, the historical welding duration refers to the average duration of welding a certain welding point when the historical high-speed wire harness is welded, wherein the welding equipment when the historical high-speed wire harness is welded is the same as the low-heat-loss welding device, and the voltage in the welding process is constant, and the current is adjusted in real time by relevant operating personnel to ensure that the historical high-speed wire harness will not be overwelded in the welding process. Then, the average value and the standard deviation of the plurality of historical welding durations are calculated, and the duration value obtained by adding the average value to the plurality of standard deviations is the wire harness welding window duration. The low-heat-loss welding device refers to a device for welding the target high-speed wire harness, wherein the original laser welding equipment refers to a device for generating and controlling a laser beam to melt the metal terminals of the target high-speed wire harness to realize wire harness connection, for example: a laser welding machine, and the welding point appearance detection device refers to a device for collecting the surface image of the welding point after the formation of the target high-speed wire harness, for example: an industrial visual detection system equipped with a high-resolution industrial camera and a specific light source (such as a ring-shaped LED lamp).

[0025] S2, welding the target high-speed wire harness based on the original laser welding equipment to obtain a controllable welded wire harness, and using the welding point appearance detection device to shoot the welding point of the controllable welded wire harness to obtain a wire harness appearance image, wherein the wire harness appearance image is an image in YCbCr color space, and the wire harness appearance image includes a plurality of original wire harness pixel points.

[0026] It can be understood that the controllable welding wire harness refers to the target high-speed wire harness being welded, wherein the target high-speed wire harness is welded based on the original laser welding equipment: the original laser welding equipment is controlled through the preset welding parameters (such as initial current, initial voltage) to perform laser irradiation melting on the specified welding point of the target high-speed wire harness, wherein the initial voltage in the welding parameter is the rated welding voltage of the original laser welding equipment, and the initial current can be set as the average current in the welding process of the plurality of historical high-speed wire harnesses. The wire harness topography image refers to a digital image collected from the welding point area of the controllable welding wire harness, wherein the welding point shooting of the controllable welding wire harness by the welding point topography detection equipment refers to: the welding point area surface image is collected by the welding point topography detection equipment at a fixed working distance perpendicular to the welding point plane, and the image is converted into YCbCr color space format. The converted image is the wire harness topography image. The original wire harness pixel point refers to a pixel point in the wire harness topography image, wherein the purpose of converting into YCbCr color space format is to separate the brightness information (Y) of the image from the color information (Cb and Cr) to facilitate the calculation of the neighborhood total standard deviation and the pixel difference value in the subsequent process.

[0027] S3, seed point selection is performed on the wire harness topography image to obtain a target seed point group, wherein the target seed point group includes a plurality of target seed points.

[0028] It can be understood that the target seed point group refers to a set composed of a plurality of target seed points, wherein the target seed point refers to a pixel point used for subsequent welding point area segmentation.

[0029] In detail, the seed point selection on the wire harness topography image to obtain the target seed point group includes: extracting original wire harness pixel points in sequence from the plurality of original wire harness pixel points in the wire harness topography image, and recording the extracted original wire harness pixel points as center wire harness pixel points: determining a set of adjacent wire harness pixel points of the center wire harness pixel points in the wire harness topography image, wherein the set of adjacent wire harness pixel points includes a plurality of adjacent wire harness pixel points; calculating a maximum distance value and a neighborhood total standard deviation according to the set of adjacent wire harness pixel points; respectively collecting the neighborhood total standard deviation and the maximum distance value corresponding to each original wire harness pixel point to obtain a plurality of neighborhood total standard deviations and a plurality of maximum distance values; performing pixel difference evaluation according to the plurality of neighborhood total standard deviations and the plurality of original wire harness pixel points to obtain a plurality of pixel difference values; selecting the target seed point group in the wire harness topography image according to the plurality of maximum distance values and the plurality of pixel difference values.

[0030] It is explained that the adjacent beam pixel point set refers to a set of a plurality of original beam pixel points adjacent to the center beam pixel point in position, for example, taking the pixel points in the 8 directions (i.e. a 3x3 pixel matrix) around the center beam pixel point as the center to form a set as the adjacent beam pixel point set. The maximum distance value can represent the maximum difference degree of the center beam pixel point and its adjacent region in color and brightness features, so the greater the maximum distance value, the more prominent the feature of the center pixel point, that is, the greater the probability of the original beam pixel point corresponding to the maximum distance value as the target seed point. The above maximum distance value and the total neighborhood standard deviation will be described in detail in subsequent embodiments. The pixel difference value refers to a value quantifying the difference in regional uniformity between the center beam pixel point and the adjacent beam pixel point set. The smaller the pixel difference value, the closer the topography of the center beam pixel point to the topography of its surrounding region (i.e. the adjacent beam pixel point set), the more likely the center beam pixel point is located inside a uniform and continuous weld point region, and the greater the probability of the original beam pixel point corresponding to the pixel difference value as the target seed point.

[0031] In detail, the maximum distance value and the total neighborhood standard deviation are calculated according to the adjacent beam pixel point set, including: extracting adjacent beam pixel points in the adjacent beam pixel point set in turn, calculating the distance of the center beam pixel point based on the extracted adjacent beam pixel points to obtain a channel distance value, and summarizing the channel distance values to obtain a plurality of channel distance values; identifying the maximum distance value in the plurality of channel distance values; constructing a neighborhood channel standard deviation group based on the adjacent beam pixel point set, wherein the neighborhood channel standard deviation group includes a neighborhood Y channel standard deviation, a neighborhood Cb channel standard deviation and a neighborhood Cr channel standard deviation; summing the neighborhood channel standard deviation group to obtain a total neighborhood standard deviation.

[0032] It can be understood that the channel distance value refers to the distance value of the center beam pixel point and the adjacent beam pixel point in the pixel value, and the calculation method of the channel distance value is: wherein, the channel distance value is represented by d, , and Y, Cb and Cr respectively represent the Y channel value, the Cb channel value and the Cr channel value corresponding to the center beam pixel point, , and respectively represent Y channel values, Cb channel values and Cr channel values corresponding to adjacent bundle pixel points. The identifying the maximum distance value in the plurality of channel distance values means taking the distance value with the largest value in the plurality of channel distance values as the maximum distance value. The neighborhood channel standard deviation set means a set composed of a neighborhood Y channel standard deviation, a neighborhood Cb channel standard deviation and a neighborhood Cr channel standard deviation, wherein the neighborhood Y channel standard deviation means a standard deviation of Y channel values corresponding to all adjacent bundle pixel points in the set of adjacent bundle pixel points, the neighborhood Cb channel standard deviation means a standard deviation of Cb channel values corresponding to all adjacent bundle pixel points in the set of adjacent bundle pixel points, and the neighborhood Cr channel standard deviation means a standard deviation of Cr channel values corresponding to all adjacent bundle pixel points in the set of adjacent bundle pixel points. The above-mentioned neighborhood total standard deviation is a sum of the neighborhood Y channel standard deviation, the neighborhood Cb channel standard deviation and the neighborhood Cr channel standard deviation.

[0033] In detail, the pixel difference evaluation according to the plurality of neighborhood total standard deviations and the plurality of original bundle pixel points to obtain a plurality of pixel difference values comprises: identifying a global maximum standard deviation in the plurality of neighborhood total standard deviations, and normalizing the plurality of neighborhood total standard deviations by using the global maximum standard deviation to obtain a plurality of neighborhood normalized standard deviations; taking the plurality of original bundle pixel points as a plurality of difference center pixel points, and performing the following operations on each difference center pixel point in the plurality of difference center pixel points: obtaining a plurality of difference adjacent pixel points of the difference center pixel point, identifying a center total standard deviation corresponding to the difference center pixel point and a plurality of difference adjacent standard deviations corresponding to the plurality of difference adjacent pixel points in the plurality of neighborhood normalized standard deviations, respectively; calculating a pixel difference value based on the center total standard deviation and the plurality of difference adjacent standard deviations; summarizing the pixel difference values to obtain the plurality of pixel difference values.

[0034] It needs to be explained that the global maximum standard deviation means the neighborhood total standard deviation with the largest value in the plurality of neighborhood total standard deviations. The neighborhood normalized standard deviation means the neighborhood total standard deviation after normalization, wherein the normalization of the plurality of neighborhood total standard deviations by using the global maximum standard deviation means dividing each neighborhood total standard deviation in the plurality of neighborhood total standard deviations by the global maximum standard deviation, and the plurality of ratios obtained are the plurality of neighborhood normalized standard deviations. The difference adjacent pixel point means an adjacent bundle pixel point corresponding to the difference center pixel point. The center total standard deviation means the neighborhood normalized standard deviation corresponding to the difference center pixel point. The difference adjacent standard deviation means the neighborhood normalized standard deviation corresponding to the difference adjacent pixel point. The formula for calculating the pixel difference value based on the center total standard deviation and the plurality of difference adjacent standard deviations is: wherein, represents the pixel difference value, representing a number of difference adjacent standard deviations in the plurality of difference adjacent standard deviations, representing a center total standard deviation, representing taking an absolute value, representing a first difference adjacent standard deviation in the plurality of difference adjacent standard deviations. representing a first difference adjacent standard deviation in the plurality of difference adjacent standard deviations.

[0035] In detail, the selecting the target seed point group from the beam topography image according to the plurality of maximum distance values and the plurality of pixel difference values comprises: determining a channel distance threshold value and a pixel difference threshold value according to the plurality of maximum distance values and the plurality of pixel difference values; taking a plurality of original beam pixel points in the beam topography image as a plurality of candidate seed points; performing the following operations on each candidate seed point in the plurality of candidate seed points: determining a candidate distance value and a candidate difference value corresponding to the candidate seed point from the plurality of maximum distance values and the plurality of pixel difference values, respectively; if the candidate distance value is greater than the channel distance threshold value and the candidate difference value is less than the pixel difference threshold value, taking the candidate seed point as a valid seed point; collecting the valid seed points to obtain a plurality of valid seed points, performing connectivity recognition on the plurality of valid seed points, and obtaining a plurality of adjacent seed point groups, wherein each adjacent seed point group comprises one or more adjacent seed points; extracting adjacent seed point groups from the plurality of adjacent seed point groups in sequence, and obtaining a seed difference value group of the extracted adjacent seed point group, wherein the seed difference value group comprises a plurality of seed difference values, and each seed difference value corresponds to an adjacent seed point; determining a minimum difference value in the seed difference value group, and taking an adjacent seed point corresponding to the minimum difference value as a target seed point; collecting target seed points corresponding to each adjacent seed point group to obtain the target seed point group.

[0036] The channel distance threshold refers to a critical distance value for determining whether a candidate seed point has a sufficient distance value to become a seed point. The channel distance threshold is determined in the following manner: taking the average of all maximum distance values and multiplying the average by a safety factor (such as 1.2) to obtain the channel distance threshold. The pixel difference threshold refers to a critical difference value for determining whether the region where the candidate seed point is located is uniform enough. The pixel difference threshold is determined in the following manner: taking the average of all pixel difference values and multiplying the average by a safety factor (such as 0.8) to obtain the pixel difference threshold. The candidate distance value refers to the maximum distance value corresponding to the candidate seed point. The candidate difference value refers to the pixel difference value corresponding to the candidate seed point. When the candidate seed point simultaneously satisfies the conditions that the candidate distance value is greater than the channel distance threshold and the candidate difference value is less than the pixel difference threshold, it indicates that the candidate seed point not only locates in a boundary region with significant features, but also has a good uniformity in the local region where it is located, which meets the basic conditions for being a region growing starting point (i.e., a target seed point). Therefore, the candidate seed point can be recorded as an effective seed point. The adjacent seed point group refers to a set including one or more adjacent seed points. All adjacent seed points in the same adjacent seed point group are adjacent to each other, that is, these adjacent seed points can be divided into the same adjacent beam pixel point set. Since only one target seed point is needed as a growing starting point for an adjacent weld point region (a region included in an adjacent beam pixel point set) in subsequent region segmentation, if multiple target seed points exist in an adjacent weld point region, it may cause over-segmentation or calculation redundancy. Therefore, one target seed point needs to be selected from the same adjacent seed point group. If the adjacent seed point group only includes one adjacent seed point, it indicates that the adjacent seed point is independent in space, that is, the adjacent seed point is the target seed point.

[0037] Further, the specific steps of identifying the connectivity of the plurality of effective seed points are as follows: effective seed points are extracted from the plurality of effective seed points in sequence, and the extracted effective seed point is recorded as a seed point to be identified. If there is an effective seed point adjacent to the seed point to be identified in the plurality of effective seed points, the seed point to be identified and all effective seed points adjacent to the seed point to be identified are merged, and the set obtained after the merging is an adjacent seed point group. If there is no effective seed point adjacent to the seed point to be identified in the plurality of effective seed points, the seed point to be identified is recorded as an adjacent seed point group. The adjacent seed point groups corresponding to each seed point to be identified are summarized to obtain a plurality of adjacent seed point groups. It should be noted that if an effective seed point has been divided into a certain adjacent seed point group, the effective seed point will not be considered in subsequent adjacent seed point group discrimination. The seed point difference value group refers to a set composed of a plurality of seed point difference values, wherein the seed point difference value refers to the pixel difference value corresponding to a certain adjacent seed point in the adjacent seed point group. The minimum difference value refers to the seed difference value with the smallest value in the seed difference value group.

[0038] S4, performing solder point region segmentation based on the target seed point group to obtain a plurality of wire bundle segmentation regions, and performing welding process evaluation according to the plurality of wire bundle segmentation regions to obtain a current welding process.

[0039] It should be explained that the wire bundle segmentation region refers to an image region obtained after solder point region segmentation, and each wire bundle segmentation region represents a connected region that is continuous and uniform in color and texture features and is extended from a single target seed point through a region growing algorithm. The current welding process refers to the degree of completion of the current welding.

[0040] In detail, the solder point region segmentation based on the target seed point group to obtain a plurality of wire bundle segmentation regions includes: extracting a target seed point from the target seed point group in sequence, and dividing an initial segmentation region of the extracted target seed point in the wire bundle topography image; summarizing the initial segmentation region corresponding to each target seed point to obtain a plurality of initial segmentation regions; performing the following operations on each initial segmentation region in the plurality of initial segmentation regions: determining a regionally adjacent pixel point set of the initial segmentation region in the wire bundle topography image, dividing the regionally adjacent pixel point set to the initial segmentation region to obtain an updated segmentation region; regarding the updated segmentation region as the initial segmentation region, and returning to the step of determining the regionally adjacent pixel point set of the initial segmentation region in the wire bundle topography image until the regionally adjacent pixel point set is a preset empty set; when the regionally adjacent pixel point set is the empty set, regarding the updated segmentation region as the wire bundle segmentation region; summarizing the wire bundle segmentation region corresponding to the initial segmentation region to obtain a plurality of wire bundle segmentation regions.

[0041] It should be explained that the initial segmentation region refers to a region formed by the adjacent wire bundle pixel point set of the target seed point. The regionally adjacent pixel point set refers to a set of all original wire bundle pixel points that are adjacent to the initial segmentation region and have similar color features. The updated segmentation region refers to the initial segmentation region after segmentation, wherein dividing the regionally adjacent pixel point set to the initial segmentation region refers to supplementing the regionally adjacent pixel point set to the initial segmentation region, and the initial segmentation region after the supplement is the updated segmentation region. When the regionally adjacent pixel point set is the empty set, it indicates that the initial segmentation region has been expanded to its boundary, and there is no longer an original wire bundle pixel point in the wire bundle topography image that is directly adjacent to the initial segmentation region and has sufficient similarity. At this time, the updated segmentation region is a region that has completed segmentation, and the updated segmentation region can be regarded as an on-line segmentation region.

[0042] In detail, the region-adjacent pixel set of the initial segmentation region determined in the online beam profile image comprises: determining an original-adjacent pixel set of the initial segmentation region in the online beam profile image, wherein the original-adjacent pixel set comprises a plurality of original-adjacent pixels; calculating an initial segmentation channel value group based on the original-adjacent pixel set; performing the following operations on each original-adjacent pixel in the original-adjacent pixel set: performing channel distance calculation on the original-adjacent pixel based on the initial segmentation channel value group to obtain an original channel distance value; judging whether the original-adjacent pixel is segmented; if the original-adjacent pixel is segmented, obtaining a segmented region of the original-adjacent pixel and identifying a segmented distance value between the original-adjacent pixel and the segmented region, and if the segmented distance value is greater than the original channel distance value, recording the original-adjacent pixel as a region-adjacent pixel; if the original-adjacent pixel is not segmented, recording the original-adjacent pixel as a region-adjacent pixel; collecting the region-adjacent pixels to obtain a region-adjacent pixel set.

[0043] It needs to be explained that the original-adjacent pixel set refers to a set of all pixels in the initial segmentation region. The initial segmentation channel value group refers to a set of an initial segmentation Y channel value, an initial segmentation Cb channel value and an initial segmentation Cr channel value, wherein the initial segmentation Y channel value refers to the average value of the Y channel values of all original-adjacent pixels in the original-adjacent pixel set, the initial segmentation Cb channel value refers to the average value of the Cb channel values of all original-adjacent pixels in the original-adjacent pixel set, and the initial segmentation Cr channel value refers to the average value of the Cr channel values of all original-adjacent pixels in the original-adjacent pixel set. The original channel distance value refers to a numerical value of the degree of similarity of color features between the original-adjacent pixel and the initial segmentation region. The greater the original channel distance value, the lower the degree of similarity of color features between the original-adjacent pixel and the initial segmentation region. The calculation method of the original channel distance value is as follows: obtaining an original channel value group of the original-adjacent pixel, the original channel value group comprising an original Y channel value, an original Cb channel value and an original Cr channel value, calculating the Euclidean distance between the original channel value group and the initial segmentation channel value group, and the calculation formula of the Euclidean distance is the same as the calculation method of the channel distance value. The Euclidean distance is the original channel distance value.

[0044] Further, the judging whether the original adjacent pixel point is segmented or not refers to judging whether the original adjacent pixel point has been divided into an initial segmentation region corresponding to another target seed point, if the original adjacent pixel point is divided into the initial segmentation region corresponding to another target seed point, it indicates that the original adjacent pixel point has been segmented. The segmented region refers to an initial segmentation region where the original adjacent pixel point is located after being segmented. The segmented distance value refers to an original channel distance value between the original adjacent pixel point and the segmented region, if the segmented distance value is greater than the original channel distance value, it indicates that the similarity between the original adjacent pixel point and the initial segmentation region is higher than the similarity between the original adjacent pixel point and the segmented region, and the original adjacent pixel point can be re-segmented to the initial segmentation region.

[0045] In detail, the welding progress evaluation according to the plurality of wire harness segmentation regions to obtain the current welding progress comprises: obtaining a welding completion template based on the target high-speed wire harness; comparing the plurality of wire harness segmentation regions with the welding completion template to obtain a plurality of template similarities, wherein the template similarity corresponds to the wire harness segmentation region one by one; identifying a completion similarity group from the plurality of template similarities according to a preset similarity threshold, wherein the completion similarity group includes one or more completion similarities, or the completion similarity group is an empty set; determining a welding completion region group corresponding to the completion similarity group, and counting a welding completion area of the welding completion region group; calculating the current welding progress according to the welding completion area and a preset total welding area.

[0046] It needs to be explained that the welding completion template refers to a standard image of a welding point region which has completed welding, and the welding completion template is obtained by selecting a wire harness sample which is completely same as the target high-speed wire harness in model and specification and has completed welding, then shooting an image of the welding point region of the wire harness sample, and taking the image as the welding completion template. The template similarity refers to a value quantifying the similarity between the welding completion template and a certain wire harness segmentation region, the greater the template similarity is, the greater the probability that the corresponding wire harness segmentation region has completed welding is, the above template comparison is a process of obtaining the template similarity, and the template similarity is obtained by calculating the cosine similarity between the feature vectors of the welding completion template and the wire harness segmentation region, or calculating the matching degree of the color histogram and the texture feature. The similarity threshold refers to the minimum similarity standard for determining whether a wire harness segmentation region can be recognized as having completed welding, and the similarity threshold is set by obtaining a large number of template similarities of qualified welding points and unqualified welding points in history, and taking the minimum value of the template similarities of all qualified welding points as the similarity threshold.

[0047] Further, the complete similarity group refers to a set of template similarities greater than a similarity threshold value, and when the template similarity is greater than the similarity threshold value, it indicates that the wire bundle segmentation area corresponding to the template similarity is complete. The welding completion area group refers to a set of wire bundle segmentation areas corresponding to each complete similarity in the complete similarity group. The welding completion area refers to the sum of the areas of all welding completion areas in the welding completion area group. The current welding total area refers to the total area of the wire bundle topographic image. The calculation method of the current welding progress is as follows: the value obtained by dividing the welding completion area by the current welding total area is the current welding progress.

[0048] S5, according to the current welding progress and the wire bundle welding window duration, the original laser welding equipment is regulated to obtain the target laser welding equipment.

[0049] It can be understood that the target laser welding equipment refers to the original laser welding equipment after regulation.

[0050] In detail, the regulation of the original laser welding equipment according to the current welding progress and the wire bundle welding window duration to obtain the target laser welding equipment comprises: performing welding test on the target high-speed wire bundle to obtain standard welding completion heat and constraint welding current; record the current welding window duration, and calculate the remaining welding window duration according to the current welding window duration and the wire bundle welding window duration; determine the remaining welding heat based on the current welding progress and the standard welding completion heat, and calculate the adjusted welding current according to the remaining welding heat, the preset rated welding voltage and the remaining welding window duration; if the adjusted welding current is not greater than the constraint welding current, the adjusted welding current is recorded as the target welding current; if the adjusted welding current is greater than the constraint welding current, the constraint welding current is recorded as the target welding current; update the original laser welding equipment with the target welding current to obtain the target laser welding equipment.

[0051] The standard welding completion heat refers to the total heat required for completing the welding of one welding point of the target high-speed wire harness. The standard welding completion heat is obtained in the following manner: a plurality of test wire harnesses of the same model as the target high-speed wire harness are obtained, and welding point welding is performed on each of the plurality of test wire harnesses, wherein the welding point welding step is the same as the welding step described above for the target high-speed wire harness, and different test currents are used for welding during the welding of the welding points of different test wire harnesses, and these test currents show an increasing trend. Finally, the total time length for each test wire harness to be welded is calculated, and the total time length, the rated welding voltage, and the test current corresponding to the test wire harness are multiplied to obtain a value, which is the welding completion heat. The welding completion heats corresponding to each test wire harness are summarized to obtain a plurality of welding completion heats, and the average value of the plurality of welding completion heats is taken as the standard welding completion heat. The constraint welding current refers to the upper limit of the safe current set to ensure the welding quality. Exceeding this current may cause damage to the insulation layer of the target high-speed wire harness or overheating and deformation of the metal terminal. When the current of the original laser welding equipment exceeds the constraint welding current, the target high-speed wire harness is at risk of being damaged due to overheating. The constraint welding current is obtained in the following manner: a preliminary current test is performed on the target high-speed wire harness.

[0052] Further, the current welding window time length refers to the time length during which the welding of the target high-speed wire harness has been performed. The remaining welding window time length refers to the difference between the current welding window time length and the wire harness welding window time length. The remaining welding heat refers to the heat required for completing the welding of the target high-speed wire harness. The remaining welding heat is calculated in the following manner: wherein, represents the remaining welding heat, represents the standard welding completion heat, represents the current welding progress. The adjusted welding current refers to the current value set for the original laser welding equipment to ensure that the welding is completed within the remaining welding window time length. The adjusted welding current is calculated in the following manner: wherein, represents the adjusted welding current, represents the rated welding voltage, which refers to the rated working voltage of the original laser welding equipment.

[0053] It can be understood that if the adjusted welding current is not greater than the constraint welding current, it indicates that the calculated adjusted welding current is within a safe range and can complete the remaining welding task without damaging the target high-speed wire harness, and at this time the adjusted welding current can be recorded as the target welding current. If the adjusted welding current is greater than the constraint welding current, it indicates that there is a risk of overheating if the adjusted welding current is directly used, and at this time the constraint welding current needs to be recorded as the target welding current. The above updating of the original laser welding equipment using the target welding current means that the current output parameter of the original laser welding equipment is adjusted to the value of the target welding current.

[0054] S6, the target laser welding equipment and the controllable welding wire harness are respectively taken as the original laser welding equipment and the target high-speed wire harness, and the step of welding the target high-speed wire harness based on the original laser welding equipment is returned until a preset welding window end instruction is received, and the high-speed wire harness low-heat-loss welding based on the weld appearance recognition is completed.

[0055] It can be understood that the welding window end instruction refers to an instruction for ending welding automatically issued by a control system, wherein the control system refers to a central processing unit for coordinating the entire welding process. When the welding time of the target high-speed wire harness here is equal to the wire harness welding window time, the control system generates a welding window end instruction. After receiving the welding window end instruction, in order to ensure that the welding process of other wire harnesses on the production line is not blocked, the target high-speed wire harness at this time needs to be transferred to an artificial re-inspection station, and the artificial re-inspection station needs to perform artificial detection on the target high-speed wire harness. If the artificial detection finds that the target high-speed wire harness has not completed welding, a related person needs to perform repair welding.

[0056] The present application is to solve the problems in the background art. First, the target high-speed wire harness and the wire harness welding window duration are determined. This step calculates the wire harness welding window duration through historical data. Compared with the method of setting welding time relying on experience in the prior art, the welding process can be more accurately controlled, heat damage caused by improper duration is avoided, and the rationality of the welding parameters is improved. Then, seed point selection is performed on the wire harness topographic image to obtain a target seed point group. This step selects seed points by comprehensively calculating the maximum distance value and the pixel difference value. Compared with the traditional single threshold method, the weld area boundary can be more accurately identified, the initial error of the region growing algorithm is reduced, and the segmentation accuracy is improved. Further, the present application performs weld area segmentation based on the target seed point group to obtain multiple wire harness segmentation areas. The welding process is evaluated according to the multiple wire harness segmentation areas to obtain the current welding process. This step uses the region growing algorithm to perform weld segmentation and combines template comparison to evaluate the welding process. Compared with the existing method relying on artificial visual inspection, the welding progress is automatically and real-time monitored, and the intelligent level of welding quality control is improved. Finally, the original laser welding equipment is regulated and controlled according to the current welding process and the wire harness welding window duration to obtain the target laser welding equipment. This step dynamically adjusts the welding current according to the remaining welding heat and the window duration. Compared with the fixed parameter welding technology, the welding quality can be ensured while being self-adaptively adjusted, the risk of overheating is effectively reduced, and the stability of the entire welding process is ensured compared with the existing one-time welding method. Therefore, the present application can improve the automation level of the wire harness welding process and reduce the risk of heat damage to the wire harness.

[0057] As Figure 2 shown, it is a functional module diagram of the high-speed wire harness low-heat damage welding system based on weld topography identification provided by an embodiment of the present application.

[0058] The high-speed wire harness low-heat damage welding system based on weld topography identification 100 can be installed in an electronic device. According to the functions implemented, the high-speed wire harness low-heat damage welding system based on weld topography identification 100 can include a high-speed wire harness determination module 101, a wire harness image shooting module 102, a welding process estimation module 103, and a welding equipment regulation module 104. The modules of the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, which are stored in the memory of the electronic device. The high-speed wire harness determination module 101 is used to determine the target high-speed wire harness and the wire harness welding window duration, and to construct a low-heat damage welding device. The low-heat damage welding device includes an original laser welding equipment and a weld topography detection equipment. The wire harness image shooting module 102 is configured to weld a target high-speed wire harness based on an original laser welding device to obtain a controllable welded wire harness, and to shoot a welding point of the controllable welded wire harness by using a welding point topography detection device to obtain a wire harness topography image, wherein the wire harness topography image is an image in a YCbCr color space, and the wire harness topography image includes a plurality of original wire harness pixel points. The welding process estimation module 103 is configured to select seed points from the wire harness topography image to obtain a target seed point group, wherein the target seed point group includes a plurality of target seed points, to perform welding point region segmentation based on the target seed point group to obtain a plurality of wire harness segmentation regions, and to perform welding process evaluation according to the plurality of wire harness segmentation regions to obtain a current welding process. The welding device regulation module 104 is configured to regulate the original laser welding device according to the current welding process and a wire harness welding window time length to obtain a target laser welding device, and to use the target laser welding device and the controllable welded wire harness as the original laser welding device and the target high-speed wire harness, respectively, and to return to the step of welding the target high-speed wire harness based on the original laser welding device until a preset welding window end instruction is received.

[0059] In detail, the modules in the high-speed wire harness low-heat-loss welding system 100 based on welding point topography recognition in the embodiment of the present application use the same technical means as the high-speed wire harness low-heat-loss welding method based on welding point topography recognition in the above Figure 1 , and can produce the same technical effects, which will not be described here.

[0060] As shown in Figure 3 , it is a structural schematic diagram of an electronic device for implementing the high-speed wire harness low-heat-loss welding method based on welding point topography recognition according to an embodiment of the present application.

[0061] The electronic device 1 can include a processor 10, a memory 11 and a bus 12, and can further include a computer program stored in the memory 11 and executable on the processor 10, such as a high-speed wire harness low-heat-loss welding method based on welding point topography recognition program.

[0062] The memory 11 includes at least one type of readable storage medium, such as flash memory, mobile hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 can also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 1. Further, the memory 11 includes both an internal storage unit and an external storage device of the electronic device 1. The memory 11 can be used to store application software and various data installed on the electronic device 1, such as the code of the high-speed wire harness low-heat-loss welding method program based on solder joint topography recognition, and can also be used to temporarily store data that has been output or will be output.

[0063] The processor 10 can be composed of an integrated circuit in some embodiments, such as a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more combinations of central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 10 is the control core of the electronic device, which connects various components of the entire electronic device through various interfaces and lines, executes programs or modules stored in the memory 11 (such as the high-speed wire harness low-heat-loss welding method program based on solder joint topography recognition, etc.), and calls data stored in the memory 11, to perform various functions and process data of the electronic device 1.

[0064] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0065] Figure 3 Only the electronic device with components is shown, and those skilled in the art can understand that, Figure 3The illustrated structure does not constitute a limitation on the electronic device 1, and can include fewer or more components than illustrated, or combine certain components, or different component arrangements.

[0066] For example, although not shown, the electronic device 1 can also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, so that functions such as charge management, discharge management, and power consumption management can be achieved through the power management device. The power supply can also include one or more direct current or alternating current power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and any other components. The electronic device 1 can also include various sensors, Bluetooth modules, Wi-Fi modules, and the like, which are not described here.

[0067] Further, the electronic device 1 can also include a network interface, which can optionally include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), and is typically used to establish a communication connection between the electronic device 1 and other electronic devices.

[0068] Optionally, the electronic device 1 can also include a user interface, which can be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally a standard wired interface, a wireless interface. Optionally, in some embodiments, the display can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. The display can also be appropriately referred to as a display screen or a display unit, and is used to display information processed in the electronic device 1 and to display a visualized user interface.

[0069] The program of the high-speed wire harness low-heat-loss welding method based on solder joint topography recognition stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, which, when executed in the processor 10, can achieve: determining a target high-speed wire harness and a wire harness welding window duration, and constructing a low-heat-loss welding device, wherein the low-heat-loss welding device includes an original laser welding device and a solder joint topography detection device; welding the target high-speed wire harness based on the original laser welding device to obtain a controllable welded wire harness, and using the solder joint topography detection device to take a solder joint image of the controllable welded wire harness to obtain a wire harness topography image, wherein the wire harness topography image is an image in the YCbCr color space, and the wire harness topography image includes multiple original wire harness pixel points; selecting a target seed point group from the wire harness topography image, wherein the target seed point group includes multiple target seed points; based on the target seed point group, weld point region segmentation is performed to obtain a plurality of wire harness segmentation regions, and welding process evaluation is performed according to the plurality of wire harness segmentation regions to obtain a current welding process; According to the current welding process and the wire harness welding window time length, the original laser welding equipment is regulated to obtain a target laser welding equipment; The target laser welding equipment and the controllable welding wire harness are respectively taken as the original laser welding equipment and the target high-speed wire harness, and the step of welding the target high-speed wire harness based on the original laser welding equipment is returned until a preset welding window end instruction is received, and the high-speed wire harness low-heat-loss welding based on weld point morphology identification is completed.

[0070] Specifically, the specific implementation method of the processor 10 to the above instructions can refer to Figures 1 to 3 The description of related steps in the corresponding embodiments will not be repeated here.

[0071] Further, the modules / units integrated in the electronic device 1, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. The computer readable storage medium can be volatile or non-volatile. For example, the computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory).

[0072] The application also provides a computer readable storage medium, the readable storage medium stores a computer program, when the computer program is executed by the processor of the electronic device, the computer program can realize: determine the target high-speed wire harness and the wire harness welding window time length, and construct a low-heat-loss welding device, wherein the low-heat-loss welding device comprises an original laser welding equipment and a weld point morphology detection equipment; weld the target high-speed wire harness based on the original laser welding equipment to obtain a controllable welding wire harness, and use the weld point morphology detection equipment to take a picture of the controllable welding wire harness to obtain a wire harness morphology image, wherein the wire harness morphology image is an image in YCbCr color space, and the wire harness morphology image comprises a plurality of original wire harness pixel points; seed point selection is performed on the wire harness morphology image to obtain a target seed point group, wherein the target seed point group comprises a plurality of target seed points; based on the target seed point group, weld point region segmentation is performed to obtain a plurality of wire harness segmentation regions, and welding process evaluation is performed according to the plurality of wire harness segmentation regions to obtain a current welding process; According to the current welding process and the wire harness welding window time length, the original laser welding equipment is regulated to obtain a target laser welding equipment; The target laser welding device and the controllable welding harness are taken as an original laser welding device and a target high-speed harness respectively, and the step of welding the target high-speed harness based on the original laser welding device is returned until a preset welding window end instruction is received, and the high-speed harness low-heat-loss welding based on the welding spot topography recognition is completed.

[0073] In several embodiments provided by the present application, it should be understood that the disclosed devices, systems and methods can be implemented in other manners. For example, the embodiments of the system described above are merely schematic, and the actual implementation can be divided into other forms.

[0074] The modules described as separated components can or can not be physically separated, and the components displayed as modules can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiments.

[0075] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0076] It is obvious for those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0077] Finally, it should be noted that the above embodiments are merely used to illustrate the technical solutions of the present application but not limit the present application, and although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application.

Claims

1. A method for low-heat-loss welding of high-speed wire harnesses based on solder joint morphology recognition, characterized in that, The method includes: The target high-speed wire harness and the wire harness welding window duration were determined, and a low heat loss welding device was constructed. The low heat loss welding device includes: original laser welding equipment and weld joint morphology detection equipment. The target high-speed wire bundle is welded using the original laser welding equipment to obtain a controllable welded wire bundle. The weld point morphology detection equipment is used to photograph the weld points of the controllable welded wire bundle to obtain a wire bundle morphology image. The wire bundle morphology image is an image in the YCbCr color space and includes multiple original wire bundle pixels. Seed points are selected from the wire harness topography image to obtain a target seed point group, wherein the target seed point group includes multiple target seed points; The solder joint area is segmented based on the target seed point group to obtain multiple wire harness segmentation areas. The welding process is evaluated based on the multiple wire harness segmentation areas to obtain the current welding process. The original laser welding equipment is adjusted according to the current welding process and the welding window duration of the wire harness to obtain the target laser welding equipment; The target laser welding equipment and the controllable welding wire bundle are respectively used as the original laser welding equipment and the target high-speed wire bundle. The process of welding the target high-speed wire bundle based on the original laser welding equipment is repeated until the preset welding window end command is received, thus completing the low heat loss welding of the high-speed wire bundle based on the weld point morphology recognition.

2. The high-speed wire harness low heat loss welding method based on solder joint morphology recognition as described in claim 1, characterized in that, The seed point selection of the wire harness topography image to obtain the target seed point group includes: Extract original wire bundle pixels sequentially from multiple original wire bundle pixels in the wire bundle topography image, and denote the extracted original wire bundle pixels as the center wire bundle pixel: In the wire bundle topography image, determine the set of neighboring wire bundle pixels of the central wire bundle pixel, wherein the set of neighboring wire bundle pixels includes multiple neighboring wire bundle pixels; Calculate the maximum distance value and the total standard deviation of the neighborhood based on the pixel set of adjacent line bundles; The total standard deviation and maximum distance value of the neighborhood corresponding to each original line bundle pixel are summarized to obtain multiple total standard deviations and multiple maximum distance values; Pixel difference is evaluated based on the total standard deviation of multiple neighborhoods and multiple original line bundle pixels to obtain multiple pixel difference values; The target seed point group is selected from the bundle topography image based on multiple maximum distance values ​​and multiple pixel difference values.

3. The high-speed wire harness low-heat-loss welding method based on solder joint morphology recognition as described in claim 2, characterized in that, The calculation of the maximum distance value and the total standard deviation of the neighborhood based on the adjacent line bundle pixel set includes: Adjacent bundle pixels are extracted sequentially from the adjacent bundle pixel set. The distance between the extracted adjacent bundle pixels and the central bundle pixel is calculated to obtain the channel distance value. The channel distance values ​​are then summarized to obtain multiple channel distance values. Identify the maximum distance value among multiple channel distance values; The neighborhood channel standard deviation set is constructed based on the adjacent line bundle pixel set. The neighborhood channel standard deviation set includes: neighborhood Y channel standard deviation, neighborhood Cb channel standard deviation and neighborhood Cr channel standard deviation. The total standard deviation of the neighborhood channels is obtained by summing the standard deviations of the neighborhood channels.

4. The high-speed wire harness low heat loss welding method based on solder joint morphology recognition as described in claim 3, characterized in that, The pixel difference evaluation is performed based on the total standard deviation of multiple neighborhoods and multiple original line bundle pixels to obtain multiple pixel difference values, including: Identify the global maximum standard deviation among multiple neighborhood total standard deviations, and use the global maximum standard deviation to normalize the multiple neighborhood total standard deviations to obtain the multiple neighborhood normalized standard deviations. The original line bundle pixels are designated as multiple difference center pixels. The following operation is performed on each of the multiple difference center pixels: Obtain multiple neighboring pixels of the difference center pixel, and identify the central total standard deviation corresponding to the difference center pixel and the multiple neighboring standard deviations corresponding to the multiple neighboring pixels in the multiple neighborhood normalized standard deviations respectively; Pixel difference values ​​are calculated based on the central total standard deviation and multiple adjacent standard deviations of the differences. Summarize the pixel difference values ​​to obtain multiple pixel difference values.

5. The high-speed wire harness low heat loss welding method based on solder joint morphology recognition as described in claim 4, characterized in that, The step of selecting a target seed point group from the bundle topography image based on multiple maximum distance values ​​and multiple pixel difference values ​​includes: The channel distance threshold and pixel difference threshold are determined based on multiple maximum distance values ​​and multiple pixel difference values. Multiple original wire bundle pixels in the wire bundle topography image are recorded as multiple candidate seed points; For each of the multiple candidate seed points, perform the following operation: Among multiple maximum distance values ​​and multiple pixel difference values, the candidate distance value and candidate difference value corresponding to the candidate seed point are determined respectively; If the candidate distance value is greater than the channel distance threshold and the candidate difference value is less than the pixel difference threshold, then the candidate seed point is recorded as a valid seed point. Summarize the valid seed points to obtain multiple valid seed points. Perform connectivity identification on the multiple valid seed points to obtain multiple adjacent seed point groups, where each adjacent seed point group includes one or more adjacent seed points. In a series of adjacent seed point groups, adjacent seed point groups are extracted sequentially to obtain the seed difference value group of the extracted adjacent seed point groups. The seed difference value group includes multiple seed difference values, and each seed difference value corresponds one-to-one with an adjacent seed point. Determine the minimum difference value in the seed difference value group, and record the adjacent seed point corresponding to the minimum difference value as the target seed point; Summarize the target seed points corresponding to each adjacent seed point group to obtain the target seed point group.

6. The high-speed wire harness low heat loss welding method based on solder joint morphology recognition as described in claim 5, characterized in that, The solder joint region segmentation based on the target seed point group yields multiple wire harness segmentation regions, including: Target seed points are extracted sequentially from the target seed point group, and the initial segmentation region of the extracted target seed points is divided in the bundle topography image; By summing up the initial segmentation regions corresponding to each target seed point, multiple initial segmentation regions are obtained; For each of the multiple initial segmentation regions, perform the following operation: In the wire bundle topography image, determine the set of neighboring pixels of the initial segmentation region, and segment the set of neighboring pixels into the initial segmentation region to obtain the updated segmentation region; The updated segmented region is recorded as the initial segmented region, and the steps of determining the set of adjacent pixels of the initial segmented region in the wire bundle topography image are returned until the set of adjacent pixels of the region is a preset empty set. When the set of adjacent pixels in a region is empty, the segmented region will be updated and recorded as a wire harness segmented region. By summing the wire harness segmentation regions corresponding to the initial segmentation regions, multiple wire harness segmentation regions are obtained.

7. The high-speed wire harness low-heat-loss welding method based on solder joint morphology recognition as described in claim 6, characterized in that, The method for determining the set of adjacent pixels of the initial segmented region in the wire bundle topography image includes: In the wire bundle topography image, determine the original neighboring pixel set of the initial segmentation region, wherein the original neighboring pixel set includes multiple original neighboring pixels; Calculate the initial segmentation channel value group based on the original set of adjacent pixels; For each of the original neighboring pixels in the original set of neighboring pixels, perform the following operation: Based on the initial segmented channel value group, the channel distance between the original adjacent pixels is calculated to obtain the original channel distance value; Determine whether the original adjacent pixels have been segmented; If the original adjacent pixels have been segmented, the segmented region of the original adjacent pixels is obtained, and the segmented distance value between the original adjacent pixels and the segmented region is identified. If the segmented distance value is greater than the original channel distance value, the original adjacent pixels are recorded as region adjacent pixels. If the original adjacent pixels are not segmented, then the original adjacent pixels are recorded as the region adjacent pixels; Summarize the adjacent pixels of the region to obtain the set of adjacent pixels of the region.

8. The high-speed wire harness low-heat-loss welding method based on solder joint morphology recognition as described in claim 7, characterized in that, The step of evaluating the welding process based on multiple wire harness segmentation regions to obtain the current welding process includes: Obtain the welding completion template based on the target high-speed wire harness; The welding completed template was used to compare the templates of multiple wire harness segmentation areas to obtain multiple template similarities. The template similarity corresponds one-to-one with the wire harness segmentation area. Based on a preset similarity threshold, complete similarity groups are identified among multiple template similarities. A complete similarity group may include one or more complete similarities, or the complete similarity group may be an empty set. Determine the welding completion area group corresponding to the completion similarity group, and count the welding completion area of ​​the welding completion area group; The current welding progress is calculated based on the completed welding area and the preset total current welding area.

9. The high-speed wire harness low-heat-loss welding method based on solder joint morphology recognition as described in claim 8, characterized in that, The process of adjusting the original laser welding equipment according to the current welding process and the wire harness welding window duration to obtain the target laser welding equipment includes: Welding tests were conducted on the target high-speed wire harness to obtain the standard welding completion heat and constrained welding current. Record the current welding window duration and calculate the remaining welding window duration based on the current welding window duration and the wire harness welding window duration; The remaining welding heat is determined based on the current welding process and the standard welding completion heat. The welding current is then calculated and adjusted according to the remaining welding heat, the preset rated welding voltage, and the remaining welding window duration. If the adjusted welding current is not greater than the constrained welding current, then the adjusted welding current is recorded as the target welding current; If the adjusted welding current is greater than the constrained welding current, then the constrained welding current is recorded as the target welding current; The original laser welding equipment was updated using the target welding current to obtain the target laser welding equipment.

10. A high-speed wire harness low-heat-loss welding system based on solder joint morphology recognition, characterized in that, The system includes: The high-speed wire harness determination module is used to determine the target high-speed wire harness and the wire harness welding window duration, and to construct a low heat loss welding device. The low heat loss welding device includes: original laser welding equipment and weld joint morphology detection equipment. The wire harness image capture module is used to weld a target high-speed wire harness based on the original laser welding equipment to obtain a controllable welded wire harness. The weld point morphology detection equipment is used to capture the weld points of the controllable welded wire harness to obtain a wire harness morphology image. The wire harness morphology image is an image in the YCbCr color space and includes multiple original wire harness pixels. The welding process estimation module is used to select seed points from the wire harness topography image to obtain a target seed point group, wherein the target seed point group includes multiple target seed points. Based on the target seed point group, the welding point region is segmented to obtain multiple wire harness segmentation regions. The welding process is evaluated based on the multiple wire harness segmentation regions to obtain the current welding process. The welding equipment control module is used to adjust the original laser welding equipment according to the current welding process and the welding window duration of the wire harness to obtain the target laser welding equipment. The target laser welding equipment and the controllable welding wire harness are respectively used as the original laser welding equipment and the target high-speed wire harness, and the process returns to the step of welding the target high-speed wire harness based on the original laser welding equipment until a preset welding window end command is received.

Images