CN211426295U - Automobile paint surface appearance defect detection system - Google Patents
Automobile paint surface appearance defect detection system Download PDFInfo
- Publication number
- CN211426295U CN211426295U CN201920617734.2U CN201920617734U CN211426295U CN 211426295 U CN211426295 U CN 211426295U CN 201920617734 U CN201920617734 U CN 201920617734U CN 211426295 U CN211426295 U CN 211426295U
- Authority
- CN
- China
- Prior art keywords
- automobile
- light source
- detection
- guide rail
- image
- 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
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The utility model discloses a car finish surface appearance defect detecting system. The utility model comprises a movable lighting source subsystem, an imaging system arranged in multiple visual angles and a detected automobile; during detection, after the detected automobile moves to a detection area, the illumination light source subsystem is moved, the imaging system continuously takes pictures in real time to carry out full-range imaging on the surface of the automobile, and appearance defects of the surface of the automobile paint are extracted through an automobile paint image processing method after imaging. The utility model has high detection precision and efficiency, and can quantitatively identify various surface defects such as paint spattering, scratching and the like; the utility model discloses can reform transform the current source of illumination in automobile production workshop, move through light and shade field of light source system, improve the surface appearance defect detection effect of high bright lacquer painting, can be accurate and quick detect out the position and the yardstick of defect, avoid the subjectivity and the inefficiency of people's eye estimation observation.
Description
Technical Field
The utility model belongs to machine vision detects the field, relates to car finish surface appearance defect detecting system.
Background
The automobile is widely used in daily life of people and becomes a preferred vehicle for people to go out. In the production process of automobiles, the quality of paint spraying reflects the quality of the appearance of the automobiles, but impurity points are inevitably present in the paint spraying process, which can lead to appearance defects such as concave and convex points on the painted surface after paint spraying, in addition, in the assembling process of painted surface parts, the paint surface is inevitably rubbed, which can lead to appearance defects such as partial scratches, paint falling and the like in the assembled vehicles, the appearance defects inevitably generate disputes between sale and production in the automobile sale, and in order to avoid the disputes, the detection of the finished automobile paint surface is very necessary before the automobiles leave the factory.
The current detection means of the automobile finish is mainly a visual method, the visual method is greatly influenced by the proficiency of a detected person and has stronger subjectivity, and in addition, the finish is a high-reflection surface and is greatly influenced by an illumination angle, so that more missed detections cannot be avoided when people visually observe, the long-term detection can cause eye fatigue of people, and the missed detections of appearance defects can also be caused. Because the visual method has the advantages of low detection speed, high omission factor and poor reliability, the assembly line detection of the whole production flow cannot be realized. Therefore, the development of the system and the method for detecting the surface appearance defects of the automobile paint surface can greatly improve the automobile appearance quality and the detection efficiency.
Disclosure of Invention
The utility model aims at the current not enough of visual detection technique, for solving the detection of car finish appearance defect, provide a car finish surface appearance defect detecting system.
The utility model provides a technical scheme that its technical problem adopted as follows:
the utility model comprises a stripe scanning lighting system (S1), an imaging system (S2) with multi-view arrangement, and a detected automobile (S3); during detection, a detected automobile (S3) moves to a fixed position through a guide groove and stops, a stripe scanning lighting system (S1) is arranged on a moving guide rail system (S4) and moves at a constant speed along the advancing direction of the automobile, and an imaging system (S2) arranged in a multi-view angle mode takes pictures in real time at the maximum frame rate to acquire a series of image sequences of each sub-aperture; extracting surface appearance defects of the automobile paint surface after image processing of the image sequence;
saidThe stripe-scan illumination system (S1) includes: a plurality of metal frames which are formed by polygons and can do one-dimensional uniform translation, strip-shaped white light illumination light sources can be arranged on the frames, and each frame is parallelly arranged on a moving guide rail system (S4) according to the same interval T; the travel of the moving guide rail system (S4) is more than 2 times of the interval T of the metal frame; during detection, the light source moves in a one-dimensional constant speed mode through the moving guide rail system (S4), the direction of the light source is perpendicular to the polygonal surface formed by the light source frame, and the image reflected by the strip-shaped light source scans the surface of the automobile so as to image the appearance of the whole surface; the interval T of the metal frame is the width T of the strip-shaped light sourcew3-6 times of the total light quantity of the light source, and continuous illumination must be ensured when the light sources are installed; the subsystem can be built from the beginning, and can also be modified by using the existing lighting frames fixed on the ground in most detection workshops, adjusting the frame interval and increasing the bottom moving guide rail or the conveyor belt.
The imaging system (S2) is arranged to be multi-view tilted so as to cover the full field of view; placing the camera at a specified tilt angle; the number of required cameras is 16-30 or more, and the images and video streams are synchronously shot in real time, so that the sub-apertures can cover the whole automobile measured surface without dead zones.
A method for detecting appearance defects of a paint surface of an automobile comprises the following specific implementation method:
after the detected automobile (S3) passes through the guide groove, the detected automobile stops at a specified position, the relative position of the detected automobile and the camera group is kept unchanged every time, and the illumination light source subsystem (S1) is moved to collect an image video stream. Calibrating the vehicle type detected for the first time, and adjusting the exposure and gain of the cameras according to the color and the light reflection characteristic of the vehicle to ensure that the image of each camera is clear and is not overexposed; then, moving the lighting source subsystem (S1) for 1-2 strip frames at intervals, and acquiring about 40 images by matching with the frame rate of the camera; fusing the collected images, searching a characteristic region of the images for matching, and dividing a detection Region (ROI) and a non-detection region in the images; after calibration, acquiring images by using parameter setting capable of calling calibration of the same vehicle type color, determining the detection area of each sub-aperture image through image matching and the previously calibrated ROI, and detecting and classifying defects by using an image enhancement and feature extraction method.
The utility model discloses beneficial effect as follows:
the utility model provides a fluorescent lamp stripe removes and forms bright dark field, gathers the method that the image fusion was handled in real time, has solved the outward appearance detection of the different colour lacquer painting defects in the complicated curved surface of car, compares in traditional diffuse reflection formation of image detection mode, has the detection precision height, does not have advantages such as too bright dark blind area, detection efficiency height. The method comprises the steps of utilizing a high-speed camera to collect bright and dark field illumination images of the surface of a moving automobile in real time, and utilizing different image fusion and processing methods to detect appearance defects of various automobile paint surfaces. The arrangement of the cameras enables the view fields of all the view angles to cover the whole automobile body, and templates of all the view fields are designed aiming at different automobile types, so that the method is suitable for the online appearance defect detection process of various automobiles.
Drawings
FIG. 1 is a system for detecting surface appearance defects of automotive painted surfaces;
FIG. 2 is a schematic diagram of the surface appearance defect detection imaging of the automobile paint surface;
Detailed Description
The present invention will be further explained with reference to the drawings and examples.
As shown in fig. 1 and 2, an automobile paint surface appearance defect detection system comprises a movable illumination light source subsystem (S1), an imaging system (S2) arranged in multiple viewing angles; during detection, a detected automobile (S3) moves to a fixed position through a guide groove and stops, an illumination light source subsystem (S1) moves at a constant speed on a light source moving guide rail system (S4) along the advancing direction of the automobile, an imaging system (S2) arranged in a multi-view angle mode takes pictures in real time at the maximum frame rate, and a series of image sequences of each sub-aperture are obtained. And after the image sequence is subjected to image processing, extracting the surface appearance defects of the automobile paint surface.
The illumination light source subsystem (S1) includes: a plurality of metal frames which are formed by polygons and can do one-dimensional uniform translation, strip-shaped white light illumination light sources are arranged on the frames, and each frame is parallelly arranged on a moving guide rail system (S4) according to the same interval T; the travel of the moving guide rail system (S4) is more than 2 times of the interval T of the metal frame(ii) a During detection, the light source moves in a one-dimensional constant speed mode through a moving guide rail system (S4), the direction of the light source is perpendicular to a polygonal surface formed by a light source frame, and an image reflected by the strip-shaped light source is scanned on the surface of an automobile so as to image the appearance of the whole surface; the interval T of the metal frame is the width T of the strip-shaped light sourcew3-6 times of the total light quantity of the light source, and continuous illumination must be ensured when the light sources are installed;
the imaging system (S2) of the multi-view arrangement is to be multi-view obliquely arranged so as to cover the full field of view; the automobile paint surface is a high-reflection surface, the strip-shaped light source and the camera are difficult to design according to a common path, and the camera is often required to be placed at a certain inclination angle; the appearance surface shape of the automobile is complex, the size and the scale of the automobile are different, 16-30 or more cameras are needed to shoot image video streams from all positions in real time at different angles, so that the sub-aperture can cover the whole automobile measured surface, and no blind area is left.
The implementation method of the system for detecting the appearance defects of the surface of the automobile paint surface comprises the following specific steps:
step 1, after passing through a guide groove, a detected automobile (S3) stops at a specified position, the relative position of the detected automobile and a camera group is kept unchanged every time, and an illumination light source subsystem (S1) is moved to collect an image video stream. Calibrating the vehicle type detected for the first time, and adjusting the exposure and gain of the cameras according to the color and the light reflection characteristic of the vehicle to ensure that the image of each camera is clear and is not overexposed;
step 2, moving the illumination light source subsystem (S1), wherein the moving distance is 1-2 strip frame intervals, and acquiring about 40 images by matching with the frame rate of a camera;
step 3, fusing the collected images, searching a characteristic region of the image for matching, and dividing a detection Region (ROI) and a non-detection region in the image; after calibration, acquiring images by using parameter setting capable of calling calibration of the same vehicle type color, determining the detection area of each sub-aperture image through the existing image matching and the previously calibrated ROI, and detecting and classifying defects by using an image enhancement and feature extraction method.
Claims (3)
1. An automobile paint surface appearance defect detection system is characterized by comprising a stripe scanning illumination system (S1), an imaging system (S2) arranged in a multi-view mode, and an automobile to be detected (S3); during detection, a detected automobile (S3) moves to a fixed position through a guide groove and stops, a stripe scanning lighting system (S1) is arranged on a moving guide rail system (S4) and moves at a constant speed along the advancing direction of the automobile, and an imaging system (S2) arranged in a multi-view angle mode takes pictures in real time at the maximum frame rate to acquire a series of image sequences of each sub-aperture.
2. The system for detecting surface appearance defects of automotive finishes as claimed in claim 1, characterized in that said stripe scanning illumination system (S1) comprises: a plurality of metal frames which are formed by polygons and can do one-dimensional uniform translation, strip-shaped white light illumination light sources can be arranged on the frames, and each frame is parallelly arranged on a moving guide rail system (S4) according to the same interval T; the travel of the moving guide rail system (S4) is more than 2 times of the interval T of the metal frame; during detection, the light source moves in a one-dimensional constant speed mode through the moving guide rail system (S4), the direction of the light source is perpendicular to the polygonal surface formed by the light source frame, and the image reflected by the strip-shaped light source scans the surface of the automobile so as to image the appearance of the whole surface; the interval T of the metal frame is the width T of the strip-shaped light sourcew3-6 times of the total light quantity of the light source, and continuous illumination must be ensured when the light sources are installed; the stripe scanning lighting system can be built from the beginning, and can also be modified by using the existing lighting frames fixed on the ground in most detection workshops, adjusting the frame interval and increasing the bottom moving guide rail or the conveyor belt.
3. The system for detecting surface appearance defects of automotive finishes as claimed in claim 2, characterized in that said imaging system (S2) is arranged to be multi-view tilted so as to cover the full field of view; placing the camera at a specified tilt angle; the number of required cameras is 16-30, and the image video streams are synchronously shot in real time, so that the sub-apertures can cover the whole automobile measured surface without dead zones.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201920617734.2U CN211426295U (en) | 2019-04-30 | 2019-04-30 | Automobile paint surface appearance defect detection system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201920617734.2U CN211426295U (en) | 2019-04-30 | 2019-04-30 | Automobile paint surface appearance defect detection system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN211426295U true CN211426295U (en) | 2020-09-04 |
Family
ID=72288325
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201920617734.2U Active CN211426295U (en) | 2019-04-30 | 2019-04-30 | Automobile paint surface appearance defect detection system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN211426295U (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110146507A (en) * | 2019-04-30 | 2019-08-20 | 杭州晶耐科光电技术有限公司 | Automobile finish surface appearance defects detection system and method |
CN115290668A (en) * | 2022-09-28 | 2022-11-04 | 苏州振畅智能科技有限公司 | System and method for detecting defects of finish paint of coated car body |
-
2019
- 2019-04-30 CN CN201920617734.2U patent/CN211426295U/en active Active
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110146507A (en) * | 2019-04-30 | 2019-08-20 | 杭州晶耐科光电技术有限公司 | Automobile finish surface appearance defects detection system and method |
CN110146507B (en) * | 2019-04-30 | 2024-01-26 | 杭州晶耐科光电技术有限公司 | System and method for detecting appearance defects of automobile paint surface |
CN115290668A (en) * | 2022-09-28 | 2022-11-04 | 苏州振畅智能科技有限公司 | System and method for detecting defects of finish paint of coated car body |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110146507B (en) | System and method for detecting appearance defects of automobile paint surface | |
US11024020B2 (en) | Method and system for automatic quality inspection of materials and virtual material surfaces | |
RU2763417C2 (en) | System and related method for detecting small defects on/in glass sheet on process line | |
CN112150441A (en) | Smooth paint surface defect detection method based on machine vision | |
RU2762130C2 (en) | System and related method for measuring optical characteristics of glass sheet on process line | |
US5694479A (en) | Process for measuring the optical quality of a glass product | |
CN211426295U (en) | Automobile paint surface appearance defect detection system | |
EP2602763B1 (en) | Method for monitoring the quality of the primer layer applied on a motor-vehicle body before painting | |
CN108550160B (en) | Non-uniform light bar characteristic region extraction method based on light intensity template | |
CN109089013A (en) | A kind of multiple light courcess detection image acquisition methods and Machine Vision Inspecting System | |
CN110108712A (en) | Multifunctional visual sense defect detecting system | |
CN109781739A (en) | Automobile finish surface appearance defects automatic detection system and method | |
CN110987970A (en) | Object surface defect detection system and detection method | |
EP0927348B1 (en) | Automatic, optical quality control process and device for flat, even products | |
CN116500038A (en) | Image acquisition method for detecting defect of outer diameter cylindrical surface of micro workpiece | |
CN109596058A (en) | The size detection recognition methods of plastic workpiece | |
CN111103309A (en) | Method for detecting flaws of transparent material object | |
CN113610083A (en) | Character recognition and character engraving depth detection system and detection method for vehicle VIN code | |
CN112213320A (en) | Biscuit detection device and biscuit detection method based on machine vision | |
CN115541598B (en) | Automobile appearance detection method, device and system | |
CN217766111U (en) | Automobile door trim appearance detection device | |
CN114383522B (en) | Method for measuring surface gap and surface difference of workpiece with reflective difference | |
CN111344553A (en) | Defect detection method and system for curved surface object | |
CN209878610U (en) | Automobile paint surface appearance defect detection system | |
CN111583241B (en) | Mobile detection method and device for regularity of ultra-large area honeycomb products |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |