CN119035863B - A visual inspection method and system for automobile welding production line - Google Patents
A visual inspection method and system for automobile welding production line Download PDFInfo
- Publication number
- CN119035863B CN119035863B CN202411537129.6A CN202411537129A CN119035863B CN 119035863 B CN119035863 B CN 119035863B CN 202411537129 A CN202411537129 A CN 202411537129A CN 119035863 B CN119035863 B CN 119035863B
- Authority
- CN
- China
- Prior art keywords
- welding
- weld
- quality
- image
- abnormal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003466 welding Methods 0.000 title claims abstract description 358
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 111
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000011179 visual inspection Methods 0.000 title claims abstract description 23
- 230000002159 abnormal effect Effects 0.000 claims abstract description 123
- 238000009826 distribution Methods 0.000 claims abstract description 60
- 230000000007 visual effect Effects 0.000 claims abstract description 49
- 238000004458 analytical method Methods 0.000 claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 36
- 238000005516 engineering process Methods 0.000 claims abstract description 22
- 230000011218 segmentation Effects 0.000 claims abstract description 18
- 238000012800 visualization Methods 0.000 claims abstract description 5
- 230000008859 change Effects 0.000 claims description 19
- 230000005856 abnormality Effects 0.000 claims description 15
- 238000003708 edge detection Methods 0.000 claims description 10
- 230000001965 increasing effect Effects 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 238000013441 quality evaluation Methods 0.000 claims description 4
- 230000004807 localization Effects 0.000 claims 2
- 238000001514 detection method Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 13
- 230000007547 defect Effects 0.000 description 7
- 238000003908 quality control method Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000007689 inspection Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K31/00—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups
- B23K31/12—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups relating to investigating the properties, e.g. the weldability, of materials
- B23K31/125—Weld quality monitoring
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K31/00—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups
- B23K31/02—Processes relevant to this subclass, specially adapted for particular articles or purposes, but not covered by only one of the preceding main groups relating to soldering or welding
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/49—Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8854—Grading and classifying of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Mechanical Engineering (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
本发明涉及视觉检测技术领域，具体为一种汽车焊装生产线视觉检测方法及系统，包括以下步骤：基于汽车焊装生产线视频数据，对原始图像序列进行帧分割，逐帧提取关键视觉信息，利用图像处理技术调整对比度和亮度，生成序列化图像帧。本发明中，通过利用图像处理技术，如对比度和亮度调整，能够生成更清晰的序列化图像帧，提高焊缝边界的识别准确性，执行图像二值化处理不仅明确焊缝与背景的分界，还增强图像中焊缝边界的可视化，使异常温度分布的识别更为精确，通过像素强度分析识别内部不均匀性，系统能够更有效地标记和追踪异常点，提升对异常温度焊缝的定位精度，不仅优化生产效率，也提高焊接质量的整体标准。
The present invention relates to the field of visual inspection technology, specifically to a visual inspection method and system for an automobile welding production line, comprising the following steps: based on the video data of the automobile welding production line, performing frame segmentation on the original image sequence, extracting key visual information frame by frame, adjusting contrast and brightness using image processing technology, and generating serialized image frames. In the present invention, by using image processing technology, such as contrast and brightness adjustment, a clearer serialized image frame can be generated, the recognition accuracy of the weld boundary can be improved, and the image binarization processing is performed to not only clarify the boundary between the weld and the background, but also enhance the visualization of the weld boundary in the image, making the recognition of abnormal temperature distribution more accurate, and identifying internal inhomogeneities through pixel intensity analysis. The system can more effectively mark and track abnormal points, improve the positioning accuracy of abnormal temperature welds, not only optimize production efficiency, but also improve the overall standard of welding quality.
Description
Technical Field
The invention relates to the technical field of visual detection, in particular to a visual detection method and system for an automobile welding production line.
Background
The visual inspection technology is a key technology applied to industrial automation and is widely applied to quality control and inspection in the manufacturing process. Mainly relying on high resolution cameras, image processing software and machine learning algorithms, by capturing images of the product or component and then performing real-time analysis to identify defects, measure dimensions or confirm assembly integrity. The visual detection system can improve detection speed and accuracy, reduce labor cost and improve overall production efficiency. With the advancement of artificial intelligence and machine learning techniques, the level of intelligence in visual inspection systems is increasing, enabling more complex inspection tasks to be handled and providing more accurate analysis results.
The visual inspection method of the automobile welding production line is characterized in that in the automobile manufacturing process, particularly in the welding and assembling links, a visual inspection technology is used for ensuring the quality and the safety of products. The welding point and the assembly quality are monitored by using a specially designed camera and an image processing system, and welding defects such as cracks, holes or uneven welding lines are detected in real time. The automatic production line has the main purposes of improving the automation level of a production line, reducing the outflow of defective products, ensuring the structural strength and durability of automobile parts, improving the safety and reliability of the products, and avoiding the follow-up high-cost reworking or recall by manufacturers.
The conventional visual inspection technology is widely applied to the field of industrial automation, plays a key role in quality control and inspection, but still shows some defects when processing complex or variable production scenes. For example, conventional visual detection systems are not robust enough in image processing in the presence of illumination changes or background disturbances, i.e. are not stable and reliable enough to face uncertainties and changes, resulting in frequent false identification or missed detection problems. Conventional systems require high performance processor support when analyzing high speed moving objects in real time, increasing costs. Even a small detection delay can lead to a decrease in production efficiency and an increase in cost on an automobile welding production line with a very high degree of automation. The prior art has limitations in early identification and accurate positioning of welding defects, and reduces the fault response speed and maintenance efficiency of the production line to some extent.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides a visual detection method and a visual detection system for an automobile welding production line.
In order to achieve the above purpose, the invention adopts the following technical scheme that the visual inspection method for the automobile welding production line comprises the following steps:
S1, based on video data of an automobile welding production line, carrying out frame segmentation on an original image sequence, extracting key visual information frame by frame, and adjusting contrast and brightness by utilizing an image processing technology to generate a serialized image frame;
S2, recognizing the weld joint boundary of the automobile welding by adopting the serialized image frame, and performing image binarization processing to distinguish the weld joint from the background so as to obtain a weld joint boundary image;
S3, analyzing the non-uniformity in the welding of the automobile by utilizing the pixel intensity through the weld boundary image, identifying abnormal temperature distribution, collecting abnormal characteristics, marking, tracking and positioning abnormal points, and generating a calibrated abnormal weld;
s4, adopting the calibration abnormal welding line, utilizing a time stamp index to analyze key time points in an automobile welding production line, and recording the occurrence time of the abnormality by comparing the quality change of the welding line in a differentiated time period to obtain a time positioning result;
S5, analyzing the quality of the automobile welding based on the time positioning result, classifying the welding quality according to the shape and the thermal distribution of the welding seam, classifying the automobile welding according to the quality index, and generating a welding quality grade;
And S6, adjusting and optimizing welding parameters by utilizing the welding quality grade, updating the current and the speed of a welding machine on a production line, and matching the differentiated welding requirements to obtain a production parameter adjustment scheme.
As a further scheme of the invention, the serialized image frames comprise images with adjusted contrast, images with adjusted frame brightness and key visual information of each frame, the weld boundary images comprise images with distinguished weld boundaries and images with contrasted backgrounds and weld seams, the calibrated abnormal weld seams comprise weld seam images with abnormal temperature distribution marks, intra-weld non-uniformity analysis images and abnormal point tracking positioning maps, the time positioning results comprise timestamp indexes of key time points and time period records of weld seam quality changes, the welding quality grades comprise weld seam shape classification result maps, thermal distribution analysis result maps and welding quality grading standards, and the production parameter adjustment schemes comprise updated current setting parameters, adjusted welding speed parameters and differentiated welding requirement matching schemes.
As a further scheme of the invention, based on the video data of the automobile welding production line, the original image sequence is subjected to frame segmentation, key visual information is extracted frame by frame, contrast and brightness are adjusted by utilizing an image processing technology, and the step of generating the serialized image frame specifically comprises the following steps:
S101, carrying out frame segmentation processing on video data of an automobile welding production line, extracting frames of a target number per second at fixed time intervals, verifying continuity and integrity of the data, and generating frame sequence images;
S102, extracting coordinates of a welding point in each frame and a contact point position of an automobile body by adopting the frame sequence image, distinguishing a key visual area through color filtering, and marking visual information to obtain a key visual frame sequence;
And S103, adjusting the contrast and brightness parameters of each frame of image through the key visual frame sequence, respectively increasing the brightness and contrast of the target quantity, optimizing the image quality and visual effect, and generating the serialized image frames.
As a further scheme of the invention, the serialized image frame is adopted to identify the weld boundary of the automobile welding, and the image binarization processing is performed to distinguish the weld from the background, so that the step of obtaining the weld boundary image is specifically as follows:
S201, adjusting the image to a uniform gray scale range according to the contrast difference between a welding line and a non-welding line area of an automobile welding production line by adopting the serialized image frame, and verifying the visualization of the welding line characteristics in the image to obtain a welding line salient image;
s202, based on the weld joint salient image, setting a gray threshold value to binarize the image, and separating a weld joint region from a non-weld joint region by optimizing the contrast degree of a weld joint boundary and a background to obtain a background distinguishing image;
S203, recognizing and extracting the position of the weld joint boundary through the background distinguishing image, optimizing the continuity and definition of the boundary line, evaluating and verifying the visual recognition effect of the boundary line, and generating a weld joint boundary image.
As a further scheme of the invention, the non-uniformity in the welding of the automobile is analyzed by utilizing the pixel intensity through the weld boundary image, the abnormal temperature distribution is identified, the abnormal characteristics are collected and marked, and the abnormal points are tracked and positioned, so that the step of generating the calibrated abnormal weld is specifically as follows:
s301, evaluating the non-uniformity inside the welding device by analyzing the intensity of pixels by adopting the weld boundary image, recording the area of the intensity change of the pixels, and marking the potential non-uniform area to obtain a non-uniform weld analysis chart;
s302, based on the uneven weld analysis chart, adopting an edge detection technology to identify an abnormal temperature region in an automobile welding production line, recording a region deviated from a standard weld temperature distribution, and processing the abnormal temperature distribution to obtain an abnormal temperature weld chart;
s303, marking the position of abnormal temperature distribution through the abnormal temperature weld map, tracking and positioning abnormal points, and correlating the mark with the real-time position of the abnormal region to generate a calibrated abnormal weld.
As a further aspect of the present invention, the formula of the edge detection technique is as follows:
Wherein, For the magnitude of the gradient,AndThe gradients of the image in the horizontal and vertical directions respectively,AndFor the weighting coefficients in the horizontal and vertical directions,As a weighting factor for the diagonal gradient,For the gradient in the diagonal direction of the image,Is the coefficient of the gaussian blur,Is the standard deviation of the two-dimensional image,Is the base of the natural index.
As a further scheme of the invention, the calibration abnormal weld joint is adopted, key time points in an automobile welding production line are analyzed by using a time stamp index, and the abnormal occurrence time is recorded by comparing the quality changes of the weld joint in a differentiated time period, so that the time positioning result is obtained specifically by the steps of:
S401, adopting the calibrated abnormal weld joints to allocate a time stamp to each abnormal weld joint, synchronizing the time stamp with a log of a welding production line, and identifying and extracting abnormal time to obtain an abnormal time stamp index map;
S402, analyzing the quality of the welding seam near the key time point based on the abnormal time stamp index map, comparing the quality change trend of the welding seam image identification of the differentiated time period, and analyzing the time interval of the welding seam quality change to obtain a time differentiated welding seam quality map;
S403, recording the time of each quality abnormality through the time-differentiated welding seam quality map, comparing the time with the history of an automobile welding production line, marking the occurrence time of the abnormality, and generating a time positioning result.
As a further scheme of the invention, based on the time positioning result, the quality of the automobile welding is analyzed, the welding quality is classified according to the shape and the heat distribution of the welding seam, the automobile welding is classified according to the quality index, and the step of generating the welding quality grade specifically comprises the following steps:
S501, analyzing the welding quality of a target time point by adopting the time positioning result, and evaluating the consistency with the standard quality by monitoring the shape and the heat distribution of the welding seam to obtain a welding seam shape and heat distribution analysis chart;
S502, classifying the welding seam by applying a preset quality evaluation index based on the welding seam shape and the thermal distribution analysis chart, wherein the welding seam comprises the integrity, the uniformity and the regularity of thermal distribution of the welding seam, and a classified welding quality chart is obtained;
And S503, grading the automobile welding according to the quality index through the classified welding quality map, sorting the priority of the quality, verifying that each grade accords with the associated quality standard, and generating a welding quality grade.
As a further scheme of the invention, the welding quality grade is utilized to adjust and optimize welding parameters, the current and the speed of a welding machine on a production line are updated, and the different welding requirements are matched, so that the steps of the production parameter adjustment scheme are specifically as follows:
S601, adopting the welding quality grades, analyzing welding quality data under the differential grades, identifying welding parameters including current and speed to be adjusted, and formulating adjustment requirements for each quality grade to obtain a welding parameter requirement analysis table;
s602, setting a current and speed adjustment scheme based on the welding parameter demand analysis table, and verifying that the welding parameters of each level are matched with quality standards according to the requirements of welding quality levels in an adjustment range to obtain a welding parameter adjustment scheme;
S603, updating the current and the speed of a welding machine on a production line through the welding parameter adjustment scheme, and generating a production parameter adjustment scheme by adjusting the setting of a differential welding station and matching the welding requirement changing in real time.
An automobile welding production line vision detection system for executing the above automobile welding production line vision detection method, the system comprising:
The frame segmentation module segments an original image sequence based on video data of an automobile welding production line, extracts visual information frame by frame, adjusts contrast and brightness of each frame and generates a serialized image frame;
The boundary recognition module recognizes a welding line and a background based on the serialized image frame, executes image binarization processing and generates a welding line boundary image;
the abnormal temperature analysis module analyzes pixel intensity, identifies abnormal temperature distribution, collects and marks abnormal characteristics, tracks and locates abnormal points by utilizing the weld boundary image, and generates a calibrated abnormal weld;
The timestamp index module analyzes key time points in an automobile welding production line through the calibration abnormal welding line, and records the occurrence time of the abnormality by comparing the quality change of the welding line to obtain a time positioning result;
And the welding quality classification module analyzes the shape and the heat distribution of the welding seam according to the time positioning result, classifies the welding quality, adjusts and optimizes welding parameters according to differentiated welding requirements, and generates a production parameter adjustment scheme.
Compared with the prior art, the invention has the advantages and positive effects that:
According to the invention, the frame segmentation and the frame-by-frame key visual information extraction are carried out on the video data of the automobile welding production line, and a clearer sequential image frame can be generated by utilizing an image processing technology such as contrast and brightness adjustment, so that the recognition accuracy of the weld joint boundary is improved. The image binarization processing is carried out, so that the boundary between the welding line and the background is clear, the visualization of the welding line boundary in the image is enhanced, and the abnormal temperature distribution is accurately identified. Through the inside inhomogeneity of pixel intensity analysis discernment, the system can mark and track the abnormal point more effectively, promotes the positioning accuracy to abnormal temperature welding seam. The timestamp index analysis helps capture key time points, and the accurate record of abnormal occurrence time is increased by comparing the quality change of the welding line in the differentiated time period. Based on the quality classification of the shape and the heat distribution of the welding seam and the adjustment of welding parameters, the production efficiency is optimized, and the integral standard of the welding quality is improved.
Drawings
FIG. 1 is a schematic workflow diagram of the present invention;
FIG. 2 is a S1 refinement flowchart of the present invention;
FIG. 3 is a S2 refinement flowchart of the present invention;
FIG. 4 is a S3 refinement flowchart of the present invention;
FIG. 5 is a S4 refinement flowchart of the present invention;
FIG. 6 is a S5 refinement flowchart of the present invention;
FIG. 7 is a S6 refinement flowchart of the present invention;
fig. 8 is a system flow diagram of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, the invention provides a technical scheme, a visual inspection method for an automobile welding production line, comprising the following steps:
s1, based on video data of an automobile welding production line, carrying out frame segmentation on an original image sequence, extracting key visual information frame by frame, adjusting contrast and brightness by utilizing an image processing technology, optimizing image quality, and generating a serialized image frame;
s2, recognizing the weld joint boundary of the automobile welding by adopting a serialized image frame, and performing image binarization processing to distinguish the weld joint from the background to obtain a weld joint boundary image;
S3, analyzing the non-uniformity in the welding of the automobile by utilizing the pixel intensity through the weld boundary image, identifying abnormal temperature distribution, collecting abnormal characteristics, marking, tracking and positioning abnormal points, and generating a calibrated abnormal weld;
S4, calibrating abnormal welding seams, analyzing key time points in an automobile welding production line by using a time stamp index, and recording the occurrence time of the abnormality by comparing the quality changes of the welding seams in the differentiated time period to obtain a time positioning result;
S5, analyzing the quality of the automobile welding based on a time positioning result, classifying the welding quality according to the shape and the thermal distribution of the welding seam, classifying the automobile welding according to quality indexes, and generating a welding quality grade;
And S6, utilizing the welding quality grade to adjust and optimize welding parameters, updating the current and the speed of a welding machine on a production line, matching the differentiated welding requirements, and optimizing the production efficiency and the quality to obtain a production parameter adjustment scheme.
The sequential image frames comprise images with adjusted contrast, images with adjusted frame brightness and key visual information of each frame, the weld boundary images comprise images with distinguished weld boundaries and images with contrasted backgrounds and weld seams, the calibrated abnormal weld seams comprise weld seam images with abnormal temperature distribution marks, in-weld non-uniformity analysis images and abnormal point tracking positioning maps, the time positioning results comprise timestamp indexes of key time points and time period records of weld seam quality changes, the welding quality grades comprise weld seam shape classification result images, heat distribution analysis result images and welding quality grading standards, the production parameter adjustment scheme comprises updated current setting parameters, adjusted welding speed parameters and a differential welding requirement matching scheme.
Referring to fig. 2, based on the video data of the welding line of the automobile, the original image sequence is subjected to frame segmentation, key visual information is extracted frame by frame, contrast and brightness are adjusted by using an image processing technology, and the steps for generating the serialized image frame are specifically as follows:
S101, carrying out frame segmentation processing on video data of an automobile welding production line, extracting a target number of frames per second at fixed time intervals, verifying continuity and integrity of the data, and generating an execution flow of a frame sequence image as follows;
the frame segmentation processing is carried out on the video data of the automobile welding production line, and the frame segmentation processing relates to the segmentation of the video according to fixed time intervals, so that the predetermined number of frames can be extracted every second, and continuous actions and key moments on the production line are captured. The key of frame segmentation is to ensure the continuity and integrity of data, avoid losing important information in the extraction process, and effectively extract static images from dynamic video, thereby providing a basis for subsequent analysis. It is also necessary to verify the time stamp and the order of each frame image to generate a frame sequence image.
S102, extracting coordinates of a welding point in each frame and a contact point position of an automobile body by adopting a frame sequence image, distinguishing a key visual area through color filtering, and marking visual information to obtain an execution flow of the key visual frame sequence as follows;
And extracting key elements by adopting frame sequence images for identification and coordinate extraction, wherein the positions of contact points of welding points and automobile bodies in each frame are required to be identified, and key visual areas are distinguished by a color filtering technology. Color filtering helps identify welds and important visual indicia for further analysis and processing. The coordinates are extracted to ensure that the movement track and the operation point of the welding point on the production line can be accurately tracked. The extracted visual information will be used for marking, providing data support for automated monitoring and quality control systems. Is the key for realizing an automatic visual monitoring system, has important significance for improving the efficiency and welding quality of a production line, obtains a key visual frame sequence, and adopts the following formula:
Wherein, Is the coordinates of the welding point,AndIs the X and Y coordinates of the point of contact of the weld with the body of the vehicle,AndThe correction value is processed and the correction value,AndIs the adjustment parameter of the X and Y coordinates,AndIs a weight factor for coordinate adjustment.
S103, adjusting the contrast and brightness parameters of each frame of image through a key visual frame sequence, respectively increasing the brightness and contrast of the target quantity, optimizing the image quality and visual effect, and generating the execution flow of the serialized image frames as follows;
The image quality of the extracted sequence of key visual frames is optimized by adjusting the contrast and brightness parameters of the image through the sequence of key visual frames. By increasing the brightness and contrast of the target number, important features in the image can be more prominent, and subsequent analysis and recognition are facilitated. Adjusting parameters not only improves the visual effect of the image, but also helps the automated system to more accurately perform tasks such as spot weld detection, alignment, etc. The optimized image is more friendly for operators to monitor the state of the production line, and problems and anomalies can be more clearly observed. Generating a serialized image frame by enhancing the quality of each frame of image, and adopting the following formula:
Wherein, Is the brightness of the image after the adjustment,For the brightness of the original image,AndIn order to adjust the parameters of contrast and brightness,For the sensitivity parameter of the contrast adjustment,Is the average brightness of the image.
Referring to fig. 3, a sequential image frame is adopted to identify a weld boundary of an automobile welding, and an image binarization process is performed to distinguish the weld from a background, so that a weld boundary image is obtained specifically as follows:
S201, adopting a serialized image frame, adjusting the image to a uniform gray scale range according to the contrast difference between a welding line and a non-welding line area of an automobile welding production line, verifying the visualization of welding line characteristics in the image, and obtaining an execution flow of a welding line salient image as follows;
the method comprises the steps of adopting a serialized image frame, highlighting a welding seam area on an automobile welding production line by adjusting the gray scale range of the image, mainly utilizing the difference of the welding seam and a non-welding seam area in contrast, adjusting the image to a uniform gray scale range so as to more clearly identify the position of the welding seam, adjusting the visual effect which is conducive to enhancing the characteristics of the welding seam, making the welding seam more obvious in the image, facilitating subsequent analysis and processing, improving the image quality by the uniform gray scale processing of the image, providing clearer visual input for an automatic detection system, improving the recognition precision and the processing speed, and generating the welding seam highlighting image, wherein the formula is as follows:
Wherein, Representing the gray value of the image after adjustment,AndIs the maximum and minimum of the target gray scale range,Is the gray value of the original image,Is the minimum gray value of the original image,AndThe maximum and minimum gray values in the original image, respectively.
S202, based on a weld joint salient image, setting a gray threshold value to binarize the image, and separating a weld joint region from a non-weld joint region by optimizing the contrast degree of a weld joint boundary and a background to obtain an execution flow of a background distinguishing image as follows;
Based on the weld salient image, binarization processing is carried out, a proper gray threshold is set to separate a weld area from a non-weld area, the contrast between the weld boundary and the background is optimized by adjusting the threshold, the weld area is ensured to be clearly separated from the surrounding non-weld area visually, the binarization is a common technology in image processing, the image is simplified into black and white two colors, the white represents the weld area and the black represents the background, the complexity of the image is simplified, the subsequent image analysis and feature extraction work such as automatic detection of the weld quality are facilitated, and a background distinguishing image is obtained.
S203, recognizing and extracting the position of a weld joint boundary through a background distinguishing image, optimizing the continuity and definition of the boundary line, evaluating and verifying the visual recognition effect of the boundary line, and generating an execution flow of the weld joint boundary image as follows;
The method is characterized in that the positions of the weld joint boundaries are identified and extracted through background distinguishing images, the boundaries are required to be identified, continuity and definition of the boundary lines are required to be optimized, the fact that the representation of the weld joint boundaries in the images is accurate is ensured, the process of optimizing the boundary lines comprises the steps of adjusting image processing parameters to improve the visual identification effect of the boundaries, the boundaries are more obvious, a machine vision system is convenient to accurately read, the key of ensuring the welding quality control precision is that the automatic equipment can accurately execute subsequent welding tasks only by clearly identifying and accurately analyzing the weld joint boundaries, and the weld joint boundary images are generated by adopting the following formula:
Wherein, Representing an image of the weld boundary,Represents the firstImage of the individual weld seam boundaryThe image of the object is a single image,Indicating the definition of the boundary line,Is a weight coefficient for adjusting the sharpness impact,A discontinuity in the boundary line is indicated,Representing the total number of images.
Referring to fig. 4, through a weld boundary image, the method includes the steps of analyzing the non-uniformity inside the automobile welding by using pixel intensity, identifying abnormal temperature distribution, collecting abnormal features, marking, tracking and locating abnormal points, and generating a calibrated abnormal weld, wherein the method specifically includes the steps of:
s301, evaluating the non-uniformity inside the welding device by analyzing the intensity of pixels by adopting a weld boundary image, recording the area of the intensity change of the pixels, and marking the potential non-uniform area to obtain an execution flow of a non-uniform weld analysis chart as follows;
By adopting the welding seam boundary image to analyze the non-uniformity inside the welding assembly, through the detailed evaluation of the pixel intensity in the image, the non-uniformity region appearing in the welding process can be identified, the region is represented as the obvious change of the pixel intensity, and is important for detecting potential welding defects, because the non-uniformity indicates the insufficient intensity of the material or the welding technical problem, quantitative analysis and quality control can be further carried out by recording the change region and marking, the consistency of the welding process and the structural integrity of the product are ensured, and a non-uniform welding seam analysis chart is generated.
S302, identifying an abnormal temperature region in an automobile welding production line by adopting an edge detection technology based on an uneven welding line analysis chart, recording a region deviated from a standard welding line temperature distribution, and processing the abnormal temperature distribution to obtain an execution flow of the abnormal temperature welding line chart as follows;
based on the uneven weld analysis graph, an abnormal temperature region is identified from the uneven weld analysis graph by using an edge detection technology, the region with uneven temperature distribution in the welding process can be accurately positioned, the welding quality is ensured, the welding defect is avoided, the abnormal temperature region is related to potential welding problems, such as overheating or not reaching the required welding temperature, the welding parameters can be processed and adjusted more accurately by recording the deviation region and comparing with the standard weld temperature distribution, the problem is corrected, and the abnormal temperature weld graph is generated.
The formula of the edge detection technique is as follows:
Wherein, For the magnitude of the gradient,AndThe gradients of the image in the horizontal and vertical directions respectively,AndFor the weighting coefficients in the horizontal and vertical directions,As a weighting factor for the diagonal gradient,For the gradient in the diagonal direction of the image,Is the coefficient of the gaussian blur,Is the standard deviation of the two-dimensional image,Is the base of the natural index.
The execution flow is as follows:
Calculating the gradient of each pixel point of the image in the horizontal and vertical directions AndApplying a weight coefficient to the gradient valuesAndAutomatic adjustment of weights by analyzing image content to enhance important edge features, introducing diagonal gradientsAnd apply weightsTo identify edge directions that are ignored in standard edge detection, applying Gaussian blur coefficientsAnd standard deviationThe influence of the gradient value is regulated through a Gaussian function so as to smooth noise and optimize an edge detection result, the optimal value of the parameter is determined through early analysis of an image, or the optimal value is obtained by learning from a batch of samples through a machine learning technology, the whole edge detection process is completed, a more accurate edge intensity diagram is obtained, and the identification capability of an abnormal temperature region is enhanced.
S303, marking the position of abnormal temperature distribution through an abnormal temperature weld map, tracking and positioning abnormal points, and associating the mark with the real-time position of an abnormal area to generate an execution flow of calibrating the abnormal weld;
Through the abnormal temperature weld map, mark the position of abnormal temperature distribution, track and position the abnormal point to with marking and the real-time position of unusual region correlate, allow operating personnel quick response and adjustment welding parameter in order to solve potential quality problem, through accurate calibration unusual region, can ensure that the welding quality control on the production line is more accurate, reduce the rejection rate effectively and improve the product uniformity, produce the unusual weld of demarcation, adopt the formula:
Wherein, A normalized difference measure representing an abnormal temperature distribution in the image,To at the same timePoint observed firstThe value of the temperature is set to be the same,As a reference to the normal temperature value of the sample,Is the number of measurement points that are to be measured,Is the standard deviation of the temperature value and,The number of measurement points is reduced by 1.
Referring to fig. 5, a calibration abnormal weld is adopted, key time points in an automobile welding production line are analyzed by using a timestamp index, and the abnormal occurrence time is recorded by comparing the quality changes of the weld in a differentiated time period, so that a time positioning result is obtained specifically by the steps of:
S401, adopting calibration of abnormal welding seams, distributing a time stamp to each abnormal welding seam, synchronizing the time stamp with a log of a welding production line, and identifying and extracting abnormal time to obtain an execution flow of an abnormal time stamp index map as follows;
By calibrating abnormal welding seams, distributing a time stamp to each identified abnormal welding seam and synchronizing the time stamp with an operation log of a welding production line, allowing specific time points of each abnormal occurrence to be accurately tracked, performing deeper analysis and root cause exploration, accurately identifying and extracting time of each abnormal welding by synchronizing the time stamp with the production line log, helping an operation team to quickly locate problems, helping preventive maintenance and operation optimization, and generating an abnormal time stamp index map, the formula is as follows:
Wherein, Representing an abnormal time stamp index map,Represents the firstThe time stamp of the individual abnormal weld joint,Is a weight coefficient for adjusting the influence of time synchronization,A time synchronization is indicated and a time synchronization is indicated,Representing a time record of the production log,The time-offset is indicated as such,Representing the total number of abnormal welds.
S402, analyzing the quality of a welding seam near a key time point based on an abnormal time stamp index chart, comparing the quality change trend of the welding seam image in a differentiated time period, and analyzing the time interval of the welding seam quality change to obtain an execution flow of a time differentiated welding seam quality chart as follows;
based on the abnormal timestamp index map, the quality of the welding seam near the key time point is analyzed, the quality change trend can be identified by comparing the welding seam images of different time periods, the continuous optimization of the welding quality is ensured, the problems and the fluctuation occurring in the production process can be better understood by analyzing the time interval of the welding seam quality change, the team is helped to predict the potential risk point, measures are taken to improve or adjust, and the time-differentiated welding seam quality map is generated.
S403, recording the time of each quality abnormality through time differentiation weld quality diagrams, comparing the time with the history of an automobile welding production line, marking the occurrence time of the abnormality, and generating an execution flow of a time positioning result as follows;
The specific time of each quality abnormality is recorded through time differentiation weld quality diagrams and is compared with historical data of an automobile welding production line, the root of the problem can be tracked and positioned more accurately through marking the occurrence time of the abnormality, the periodic problem and the random fault appearing on the production line are important to understanding, the production flow can be optimized through associating the abnormality time with the actual operation condition, the problem is ensured to be identified and solved in time, a time positioning result is generated, and the following formula is adopted:
Wherein, In order to locate the result of the time-alignment,For the point in time of the current quality anomaly,As a point in time of the historical data,For the quality anomaly to record a value,Is a specific gravity coefficient of the historical data.
Referring to fig. 6, based on the time positioning result, the quality of the automobile welding is analyzed, the welding quality is classified according to the shape and the thermal distribution of the welding seam, the automobile welding is classified according to the quality index, and the step of generating the welding quality grade specifically comprises:
S501, analyzing the welding quality of a target time point by adopting a time positioning result, and evaluating the consistency with the standard quality by monitoring the shape and the heat distribution of a welding line to obtain an execution flow of a welding line shape and heat distribution analysis chart as follows;
And analyzing the welding quality at a specific time point by adopting a time positioning result, wherein the welding quality comprises the detailed monitoring of the shape and the heat distribution of the welding seam so as to evaluate the consistency with the standard quality, and quality problems affecting the structural integrity and the performance of the product, such as irregular shape of the welding seam or uneven heat distribution, can be identified. The welding process is crucial to ensuring that each welding operation reaches high standard quality, can help a production team to timely adjust welding parameters or processes so as to prevent quality deviation from affecting the performance of products, and generates a welding line shape and thermal distribution analysis chart, wherein the formula is as follows:
Wherein, For the quality analysis value of the weld joint,AndThe shape of the weld and the coefficient of thermal distribution are respectively,AndIs the shape and thermal distribution coefficient of the standard mass,Is the standard mass deviation.
S502, classifying the welding seam by applying a preset quality evaluation index based on the welding seam shape and the thermal distribution analysis chart, wherein the welding seam comprises the integrity, the uniformity and the regularity of thermal distribution of the welding seam, and the execution flow of the classified welding quality chart is obtained as follows;
Based on the weld shape and the thermal distribution analysis chart, the welding is classified by applying a preset quality evaluation index, including evaluating the integrity, uniformity and regularity of thermal distribution of the weld, so that the quality control of the welding process is improved, the welding process can be ensured to reach a preset quality standard by quantitatively analyzing various aspects of the weld, the quality condition of each weld is understood by an operation team, the basis is provided for the subsequent process improvement, a classified welding quality chart is generated, and the formula is adopted:
Wherein, In order to classify the welding quality value,In order to be a weld integrity value,As a value for the uniformity of the weld,As the value of the regularity of the thermal distribution,To evaluate the weight coefficient.
S503, classifying the automobile welding device according to quality indexes by classifying the welding quality graphs, sorting the priority of the quality, verifying that each level accords with the associated quality standard, and generating an execution flow of the welding quality level as follows;
By classifying the welding quality diagrams, classifying the automobile welding according to quality indexes and sorting the priority of the quality, whether each level accords with the associated quality standard can be effectively verified, and the quality consistency of the whole production line is important. Through accurate quality grading, the welding process can be better managed, the occurrence of quality problems can be prevented, each welding part can reach or exceed the expected performance standard, the welding quality grade is generated, and the formula is adopted as follows:
Wherein, For the quality level of the weld,For the welding quality classification value,The coefficients are ordered in order of priority and,In order to be a standard threshold value of compliance,Is a historical data consistency coefficient.
Referring to fig. 7, the steps of using the welding quality level to adjust and optimize the welding parameters, updating the current and speed of the welding machine on the production line, and matching the differentiated welding requirements to obtain the production parameter adjustment scheme are specifically as follows:
S601, adopting welding quality grades, analyzing welding quality data under differential grades, identifying welding parameters including current and speed to be adjusted, and formulating adjustment requirements for each quality grade to obtain an execution flow of a welding parameter requirement analysis table as follows;
the weld quality data at the differential level is analyzed using the weld quality level, including identifying the weld parameters, such as current and speed, that need to be adjusted. By analyzing the welding data corresponding to each quality grade in detail, key factors influencing welding quality can be identified, specific adjustment requirements are formulated for each grade, welding quality and production efficiency are improved, and the welding parameter requirement analysis table is generated because the operation team is allowed to purposefully adjust the settings of the welding machine so as to adapt to the specific requirements of different quality grades.
S602, setting a current and speed adjustment scheme based on a welding parameter demand analysis table, and verifying that the welding parameters of each level are matched with quality standards according to the requirements of welding quality levels in an adjustment range, so as to obtain an execution flow of a welding parameter adjustment scheme draft as follows;
Based on a welding parameter demand analysis table, an adjustment scheme of current and speed is set, the range of the adjustment scheme is determined according to specific requirements of welding quality grades, the key of ensuring the matching of welding parameters and quality standards is that the welding quality and the overall efficiency of a production line can be remarkably improved through accurate parameter setting. Reasonable current and speed are set to influence welding quality, production cost and equipment service life are related, and a scheme of welding parameter adjustment is generated.
S603, updating the current and the speed of a welding machine on a production line through a welding parameter adjustment scheme, and matching the welding requirements which change in real time through adjusting the setting of a differentiated welding station to generate an execution flow of the production parameter adjustment scheme as follows;
The current and the speed of a welding machine on a production line are updated through a welding parameter adjustment scheme, and the setting of a differential welding station is adjusted to match the welding requirement which changes in real time. It is important to ensure that the welding process adapts to the changes in production requirements, and effectively adjusting parameters can reduce quality problems and improve production efficiency. By implementing the adjustment, the production line can more flexibly cope with different production challenges, and a production parameter adjustment scheme is generated.
Referring to fig. 8, a visual inspection system for an automobile welding line is provided, where the visual inspection system is configured to execute the visual inspection method for an automobile welding line, and the system includes:
The frame segmentation module segments an original image sequence based on video data of an automobile welding production line, extracts visual information frame by frame, adjusts contrast and brightness of each frame and generates a serialized image frame;
the boundary recognition module recognizes a welding line and a background based on the serialized image frame, performs image binarization processing and generates a welding line boundary image;
the abnormal temperature analysis module analyzes pixel intensity by utilizing the weld boundary image, identifies abnormal temperature distribution, collects and marks abnormal characteristics, tracks and locates abnormal points and generates a calibrated abnormal weld;
the timestamp index module analyzes key time points in an automobile welding production line by calibrating abnormal welding seams, and records the occurrence time of the abnormality by comparing the quality changes of the welding seams to obtain a time positioning result;
The welding quality classification module analyzes the shape and the heat distribution of the welding seam according to the time positioning result, classifies the welding quality, adjusts and optimizes welding parameters according to differentiated welding requirements, and generates a production parameter adjustment scheme.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.
Claims (8)
1. The visual inspection method for the automobile welding production line is characterized by comprising the following steps of:
Based on the video data of the automobile welding production line, carrying out frame segmentation on an original image sequence, extracting key visual information frame by frame, and adjusting contrast and brightness by utilizing an image processing technology to generate a serialized image frame;
Identifying the weld joint boundary of the automobile welding by adopting the serialized image frame, and performing image binarization processing to distinguish the weld joint from the background so as to obtain a weld joint boundary image;
Analyzing the non-uniformity inside the automobile welding by utilizing the pixel intensity through the weld boundary image, identifying abnormal temperature distribution, collecting abnormal characteristics, marking, tracking and positioning abnormal points, and generating a calibrated abnormal weld;
The calibration abnormal weld joint is adopted, key time points in an automobile welding production line are analyzed by using a time stamp index, and the time of occurrence of the abnormality is recorded by comparing the quality changes of the weld joint in the differentiated time period, so that a time positioning result is obtained;
Based on the time positioning result, analyzing the quality of the automobile welding, classifying the welding quality according to the shape and the thermal distribution of the welding seam, classifying the automobile welding according to the quality index, and generating a welding quality grade;
Adjusting and optimizing welding parameters by utilizing the welding quality grade, updating the current and the speed of a welding machine on a production line, and matching the differentiated welding requirements to obtain a production parameter adjustment scheme;
Analyzing the non-uniformity inside the automobile welding by utilizing the pixel intensity through the weld boundary image, identifying abnormal temperature distribution, collecting abnormal characteristics, marking, tracking and positioning abnormal points, and generating a calibrated abnormal weld, wherein the method specifically comprises the following steps of:
adopting the weld boundary image, evaluating the non-uniformity in the welding device by analyzing the intensity of pixels, recording the area of the intensity change of the pixels, and marking the potential non-uniform area to obtain a non-uniform weld analysis chart;
Based on the uneven weld analysis chart, an abnormal temperature area in an automobile welding production line is identified by adopting an edge detection technology, an area deviated from the standard weld temperature distribution is recorded, and abnormal temperature distribution is processed to obtain an abnormal temperature weld chart;
marking the position of abnormal temperature distribution through the abnormal temperature weld map, tracking and positioning abnormal points, and correlating the mark with the real-time position of the abnormal region to generate a calibrated abnormal weld;
The formula of the edge detection technology is as follows:
Wherein, For the magnitude of the gradient,AndThe gradients of the image in the horizontal and vertical directions respectively,AndFor the weighting coefficients in the horizontal and vertical directions,As a weighting factor for the diagonal gradient,For the gradient in the diagonal direction of the image,Is the coefficient of the gaussian blur,Is the standard deviation of the two-dimensional image,Is the base of the natural index.
2. The visual inspection method of an automobile welding production line according to claim 1, wherein the serialized image frames comprise contrast-adjusted images, frame brightness-adjusted images and key visual information of each frame, the weld boundary images comprise images of weld boundary distinction and images of background contrast with the weld, the calibrated abnormal weld comprises weld images of abnormal temperature distribution marks, intra-weld non-uniformity analysis images and abnormal point tracking localization maps, the time localization results comprise timestamp indexes of key time points and time period records of weld quality changes, the weld quality grades comprise a weld shape classification result map, a thermal distribution analysis result map and a weld quality grading standard, and the production parameter adjustment schemes comprise updated current setting parameters, adjusted welding speed parameters and a differential welding demand matching scheme.
3. The visual inspection method of an automobile welding line according to claim 1, wherein the steps of performing frame segmentation on an original image sequence based on the video data of the automobile welding line, extracting key visual information frame by frame, adjusting contrast and brightness by using an image processing technology, and generating a serialized image frame are specifically as follows:
Carrying out frame segmentation processing on video data of an automobile welding production line, extracting frames of a target number per second at fixed time intervals, verifying continuity and integrity of the data, and generating frame sequence images;
Coordinate extraction is carried out on the positions of the welding points in each frame and the contact points of the automobile body by adopting the frame sequence image, key visual areas are distinguished through color filtering, visual information marking is carried out, and a key visual frame sequence is obtained;
And adjusting the contrast and brightness parameters of each frame of image through the key visual frame sequence, respectively increasing the brightness and contrast of the target quantity, optimizing the image quality and visual effect, and generating the serialized image frames.
4. The visual inspection method of an automobile welding production line according to claim 1, wherein the step of identifying a weld boundary of an automobile welding by using the serialized image frame and performing image binarization processing to distinguish the weld from a background, and obtaining a weld boundary image comprises the steps of:
Adopting the serialized image frame, adjusting the image to a uniform gray scale range according to the contrast difference between the welding line and the non-welding line area of the automobile welding production line, and verifying the visualization of the welding line characteristics in the image to obtain a welding line salient image;
Based on the weld salient image, setting a gray threshold value to binarize the image, and separating a weld region from a non-weld region by optimizing the contrast degree of a weld boundary and a background to obtain a background distinguishing image;
And recognizing and extracting the position of the weld boundary through the background distinguishing image, optimizing the continuity and definition of the boundary, evaluating and verifying the visual recognition effect of the boundary, and generating a weld boundary image.
5. The visual inspection method of an automobile welding production line according to claim 1, wherein the step of using the calibration abnormal weld to analyze key time points in the automobile welding production line by using a timestamp index, and recording abnormal occurrence time by comparing quality changes of the weld in a differential time period, and obtaining a time positioning result comprises the following steps:
adopting the calibrated abnormal weld joints to allocate a time stamp to each abnormal weld joint, synchronizing the time stamp with the log of the welding production line, and identifying and extracting the abnormal time to obtain an abnormal time stamp index map;
based on the abnormal timestamp index map, analyzing the quality of the welding seam near the key time point, comparing the welding seam images in the differentiated time period to identify the quality change trend, and analyzing the time interval of the welding seam quality change to obtain a time differentiated welding seam quality map;
And recording the time of each quality abnormality through the time-differentiated welding seam quality map, comparing the time with the history of an automobile welding production line, marking the occurrence time of the abnormality, and generating a time positioning result.
6. The visual inspection method of an automobile welding production line according to claim 1, wherein the step of analyzing the quality of the automobile welding based on the time positioning result, classifying the welding quality according to the shape and heat distribution of the welding seam, classifying the automobile welding according to the quality index, and generating the welding quality grade is specifically as follows:
Analyzing the welding quality of a target time point by adopting the time positioning result, and evaluating the consistency with the standard quality by monitoring the shape and the thermal distribution of the welding seam to obtain a welding seam shape and thermal distribution analysis chart;
Classifying the welding seam by applying a preset quality evaluation index based on the welding seam shape and the thermal distribution analysis chart, wherein the welding seam comprises the integrity, the uniformity and the regularity of thermal distribution of the welding seam, and a classified welding quality chart is obtained;
And grading the automobile welding according to the quality index through the classified welding quality map, and sorting the priority of the quality, verifying that each grade accords with the associated quality standard, and generating a welding quality grade.
7. The visual inspection method of an automobile welding production line according to claim 1, wherein the step of using the welding quality grade to adjust and optimize welding parameters, updating the current and speed of a welding machine on the production line, and matching the differentiated welding requirements to obtain a production parameter adjustment scheme comprises the following steps:
Adopting the welding quality grades, analyzing welding quality data under the differential grades, identifying welding parameters which need to be adjusted, including current and speed, and formulating adjustment requirements for each quality grade to obtain a welding parameter requirement analysis table;
Based on the welding parameter demand analysis table, setting a current and speed adjustment scheme, and verifying that the welding parameters of each level are matched with quality standards according to the requirements of welding quality levels in an adjustment range to obtain a welding parameter adjustment scheme;
and updating the current and the speed of a welding machine on a production line through the draft of the welding parameter adjustment scheme, and generating the production parameter adjustment scheme by adjusting the setting of a differentiated welding station and matching the welding requirement of real-time change.
8. A visual inspection system for an automobile welding production line is characterized in that, the visual inspection method of an automobile welding line according to any one of claims 1 to 7, the system comprising:
The frame segmentation module segments an original image sequence based on video data of an automobile welding production line, extracts visual information frame by frame, adjusts contrast and brightness of each frame and generates a serialized image frame;
The boundary recognition module recognizes a welding line and a background based on the serialized image frame, executes image binarization processing and generates a welding line boundary image;
the abnormal temperature analysis module analyzes pixel intensity, identifies abnormal temperature distribution, collects and marks abnormal characteristics, tracks and locates abnormal points by utilizing the weld boundary image, and generates a calibrated abnormal weld;
The timestamp index module analyzes key time points in an automobile welding production line through the calibration abnormal welding line, and records the occurrence time of the abnormality by comparing the quality change of the welding line to obtain a time positioning result;
And the welding quality classification module analyzes the shape and the heat distribution of the welding seam according to the time positioning result, classifies the welding quality, adjusts and optimizes welding parameters according to differentiated welding requirements, and generates a production parameter adjustment scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411537129.6A CN119035863B (en) | 2024-10-31 | 2024-10-31 | A visual inspection method and system for automobile welding production line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411537129.6A CN119035863B (en) | 2024-10-31 | 2024-10-31 | A visual inspection method and system for automobile welding production line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119035863A CN119035863A (en) | 2024-11-29 |
| CN119035863B true CN119035863B (en) | 2025-02-18 |
Family
ID=93579988
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411537129.6A Active CN119035863B (en) | 2024-10-31 | 2024-10-31 | A visual inspection method and system for automobile welding production line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119035863B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN119574571B (en) * | 2025-02-06 | 2025-05-06 | 长春光华学院 | Automobile body weld joint detection method and system based on machine vision |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115100127A (en) * | 2022-06-13 | 2022-09-23 | 杭州安脉盛智能技术有限公司 | A detection method for circular weld defects based on visual recognition |
| CN118744296A (en) * | 2024-06-14 | 2024-10-08 | 苏州诺协智能科技有限公司 | System and method for weld seam recognition and trajectory detection based on vision technology |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102021117714A1 (en) * | 2021-07-08 | 2023-01-12 | Endress+Hauser SE+Co. KG | Automatic seam detection for a welding process |
| CN118435046A (en) * | 2021-12-24 | 2024-08-02 | 松下知识产权经营株式会社 | Appearance inspection device, welding system, shape data correction method, and appearance inspection method for welding part |
-
2024
- 2024-10-31 CN CN202411537129.6A patent/CN119035863B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115100127A (en) * | 2022-06-13 | 2022-09-23 | 杭州安脉盛智能技术有限公司 | A detection method for circular weld defects based on visual recognition |
| CN118744296A (en) * | 2024-06-14 | 2024-10-08 | 苏州诺协智能科技有限公司 | System and method for weld seam recognition and trajectory detection based on vision technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119035863A (en) | 2024-11-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109969736B (en) | An intelligent detection method for deviation fault of large carrying belt | |
| CN117664990A (en) | Intelligent PCBA appearance defect detection method | |
| CN119035863B (en) | A visual inspection method and system for automobile welding production line | |
| CN119035861B (en) | Intelligent welding control method and system for automobile processing | |
| CN118226236B (en) | Intelligent recognition method and device for PCB defects | |
| CN112053318A (en) | Two-dimensional PCB defect real-time automatic detection and classification device based on deep learning | |
| CN117649404A (en) | Medicine packaging box quality detection method and system based on image data analysis | |
| CN102455171B (en) | A method for detecting the geometric shape of the back side of laser tailor welding seam | |
| KR101256369B1 (en) | Flat display pannel test equipment and test method using multi ccd camera | |
| JP5168215B2 (en) | Appearance inspection device | |
| CN119198754B (en) | A visual inspection system for appearance defects of pressure sensors | |
| CN107797517B (en) | Method and system for realizing inspection of steel strip punching by machine vision | |
| CN116429768A (en) | Sealing nail welding quality detection method, system, equipment and storage medium | |
| CN110567963B (en) | Alloy analysis visual positioning method and device and alloy analysis system | |
| CN118890790B (en) | Method and system for rapidly detecting processing defects of circuit board | |
| KR20240030907A (en) | Automatic Detection System for Surface Defects of Test Subjects Based on Machine Vision | |
| CN117195439B (en) | Intelligent optimization method for large-shaft-diameter full-section mechanical sealing performance | |
| CN117314829A (en) | Industrial part quality inspection method and system based on computer vision | |
| CN119658151B (en) | Coding quality control method and system of laser vision coding machine | |
| CN119600032B (en) | A method and system for industrial product quality inspection based on machine vision | |
| CN118840709B (en) | A dynamic real-time monitoring, analysis and processing system for coal flow based on image recognition | |
| CN119304387B (en) | An automated monitoring method and system for metal sheet laser cutting operations | |
| CN118781097B (en) | PCB detection method based on artificial intelligence, terminal equipment and storage medium | |
| CN118537031B (en) | A product production supervision system and method based on visual inspection | |
| CN119044173A (en) | Lithium battery pole piece detection system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |