CN114519704A - Method for detecting scratch defects on surfaces of metal parts of communication electrical appliances - Google Patents

Abstract

The invention discloses a method for detecting scratches and defects on the surface of metal parts of communication electrical appliances, comprising the following specific steps: S1: Scanning sample construction: finding a sample of a component to be detected, and scanning it to establish a sample of the specification to be detected The parts include the inspection position and the inspection requirements of the characteristics of the parts to be inspected; S2: Placement of inspection parts: Place the parts to be inspected on the inspection station stably as required to be inspected; S3: Surface cleaning: Use a fan to treat the appearance of the inspected parts Perform non-contact cleaning to ensure that there will be no floating dust on the surface that will affect the detection results during detection; S4: Scanning image processing: Scan the parts to be detected into a map, and process the scanned images to facilitate better Extract image information; S5: scratch defect analysis; S6: result display. The method for detecting surface scratches and defects of metal parts of communication electrical appliances disclosed by the invention has the technical effect of further ensuring the accuracy of the detection results.

Description

A method for detecting scratches and defects on the surface of metal parts of communication electrical appliances

technical field

The invention relates to the technical field of production and detection of metal fittings, in particular to a method for detecting scratches and defects on the surface of metal parts of communication electrical appliances.

Background technique

During the manufacturing process of communication appliances, a large number of metal parts are involved in the assembly and processing. These parts have been turned over and installed, resulting in scratches on the surface of the parts during the manufacturing and assembly process. These defects need to be detected and dealt with in time. Avoid affecting the performance of communication appliances.

Machine vision has a wide range of applications in the fields of national economy, scientific research and national defense construction. Its biggest advantage is non-contact measurement. Compared with other methods, it has great advantages in safety, reliability, detection accuracy, detection speed, and detection cost. Nowadays, when using machine vision to detect metal parts, the detection range is usually controlled within a general area, but there may be some interference from other factors in this area, resulting in inaccurate detection results. In order to improve this problem, we The following schemes are proposed.

SUMMARY OF THE INVENTION

The invention discloses a method for detecting scratches and defects on the surface of metal parts of communication electrical appliances, aiming at solving the technical problem of inaccurate detection results caused by interference from other factors.

In order to achieve the above object, the present invention adopts the following technical solutions:

A method for detecting scratches and defects on the surface of metal parts of communication electrical appliances, comprising the following specific steps:

S1: Scanning sample construction: find a sample of the part to be inspected, and scan it to establish the inspection requirements of the part to be inspected including the inspection position and the characteristics of the part to be inspected;

S2: Placement of inspection parts: place the parts to be inspected on the inspection station stably as required to be inspected;

S3: Surface cleaning: use a fan to clean the surface of the component to be tested non-contact, to ensure that there is no floating ash on the surface that will affect the test results during testing;

S4: Scanning image processing: scan the component to be detected into a map, and process the scanned image to facilitate better extraction of image information;

S5: Scratch defect analysis: Identify and analyze whether there are scratch defects on the surface according to the processed image of the scanned image, and if so, further analyze the scanned scratch defects;

S6: Result display: display the scratch defects on the surface of the part scanned and various aspects of scratch defects data on the display screen, and give a corresponding treatment plan according to the data.

By setting the surface cleaning, non-contact cleaning is used to blow off the dust that may exist on the parts to be inspected to ensure the cleanliness of the surface of the parts, and then the contour features of the samples are constructed and modeled by scanning the samples, and in the scanning image processing Match the parts to be inspected with the modeling, so as to accurately control the peripheral contour of the inspection surface, and only detect the inspection surface during inspection, thus avoiding the influence of the transmission belt or the fixed panel and other factors on the inspection results during the inspection process. , to ensure the accuracy of the detection results.

In a preferred solution, in the S1, the construction of the scanning sample specifically includes the following steps:

S11: Sample scanning: find a sample of the component to be inspected, and perform 3D scanning in multiple orientations to obtain images of various orientations for analysis;

S12: Image feature extraction: extract meaningful features from the image using image segmentation technology according to the image scanned in the sample scan, and model according to the features;

S13: Detection surface selection: manually select the three-dimensional image scanned by the sample, select one or more specific surfaces to scan and analyze, and save the selected specific surface as the basis for subsequent inspection of large parts;

S14: Image array segmentation: The specific surface image formed by sample scanning and detection surface selection is combined with the data extracted in the image feature extraction, and the image of the surface to be detected is divided into equal parts, and the segmentation position is saved for subsequent detection. Basis for bulk shipment of parts;

In the S12, the meaningful features in the image feature extraction include the outer contour of the image and different scanning surface edges, so as to prepare for subsequent detection and positioning;

In the S4, the scanned image processing specifically includes the following steps:

S41: Part surface scanning: scan the part to be inspected to form a scanned image;

S42: image feature comparison: according to the steps of image feature extraction and detection surface selection, the modeling of the sample is compared with the component to be detected;

S43: Secondary image array segmentation: segment the surface image of the component to be detected according to the segmentation positions stored in the image array segmentation, and divide the overall surface treatment into different array small area treatments;

S44: Image enhancement: the image after the secondary image array segmentation processing is enhanced, the high-frequency components of the image are strengthened, the clarity of the image is improved, and the protruding details in the image are emphasized;

In the S42, the specific comparison method in the image feature comparison is to overlap the edge contour of the component to be detected according to the modeling in the image feature extraction. If there is a little difference, the position of the sample modeling is corrected in the system, so that The modeling can be coincident with the part to be inspected, and then the inspection surface is positioned according to the specific inspection surface selected in the inspection surface selection.

By setting up image array segmentation, secondary image array segmentation and image feature comparison, the sample is first modeled, and then the image of the surface to be detected is divided into equal parts, and the segmentation position is saved as the basis for subsequent detection, and then through The secondary image array segmentation compares the segmentation positions of the image array segmentation. The overall surface treatment is divided into different array small areas for analysis and processing, which can greatly reduce the amount of calculation and speed up the detection speed. quality.

In a preferred solution, in the S6, the result display specifically includes the following steps:

S61: Data measurement: scratch recognition and 3D depth scanning measure various data of scratches;

S62: Composition: compose the scanned surface according to the scanning image processing and scratch defect analysis, and only use lines to form the outline of the scanned surface and the shape and position of the scratches in the figure;

S63: Annotation: through the data measurement step, the measured various data are marked in the composition in detail;

S64: Alarm: alarm with a sound to remind the staff that there is a scratch defect on the surface of this part;

S65: Analysis of treatment plans: According to the measured scratch data, different treatment plans are analyzed, including grinding, filling or refurbishment;

In the S61, the various data measured in the data measurement include the length, depth, quantity and location of the scratches.

By setting up data measurement, composition, labeling and processing plan analysis, the outline of the scanned surface and the shape and position of the scratches are formed by lines through the composition, and then various data measured by the data are marked on the composition, so that the staff can be more intuitive. Viewing a number of scratch data, and assisting the processing plan analysis of the given treatment plan, can enable the staff to classify and store the problematic parts according to different treatment plans in a more timely manner, which is conducive to the unification of the problematic parts. processing to enhance work efficiency.

It can be seen from the above that a method for detecting scratches and defects on the surface of metal parts of communication electrical appliances includes the following specific steps:

S1: Scanning sample construction: find a sample of the part to be inspected, and scan it to establish the inspection requirements of the part to be inspected including the inspection position and the characteristics of the part to be inspected;

S2: Placement of inspection parts: place the parts to be inspected on the inspection station stably as required to be inspected;

S3: Surface cleaning: use a fan to clean the surface of the component to be tested non-contact, to ensure that there is no floating ash on the surface that will affect the test results during testing;

S4: Scanning image processing: scan the component to be detected into a map, and process the scanned image to facilitate better extraction of image information;

S5: Scratch defect analysis: Identify and analyze whether there are scratch defects on the surface according to the processed image of the scanned image, and if so, further analyze the scanned scratch defects;

S6: Result display: display the scratch defects on the surface of the part scanned and various aspects of scratch defects data on the display screen, and give a corresponding treatment plan according to the data. The method for detecting scratches on the surface of metal parts of communication electrical appliances provided by the present invention has the technical effect of further ensuring the accuracy of the detection results.

Description of drawings

FIG. 1 is an overall flow chart of a method for detecting scratches and defects on the surface of a metal part of a communication electrical appliance proposed by the present invention.

FIG. 2 is a flow chart of scanning sample construction of a method for detecting scratch defects on the surface of metal parts of communication electrical appliances proposed by the present invention.

FIG. 3 is a flow chart of scanning image processing of a method for detecting scratch defects on the surface of a metal part of a communication electrical appliance proposed by the present invention.

FIG. 4 is a flow chart of scratch defect analysis of a method for detecting scratch defects on the surface of metal parts of communication electrical appliances proposed by the present invention.

FIG. 5 is a flow chart showing the results of a method for detecting scratches and defects on the surface of a metal part of a communication electrical appliance proposed by the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments.

The method for detecting surface scratches and defects of metal parts of communication electrical appliances disclosed by the invention is mainly applied to the scene of detecting the parts of communication electrical appliances.

Referring to Fig. 1, a method for detecting scratches and defects on the surface of metal parts of communication electrical appliances includes the following specific steps:

S1: Scanning sample construction: find a sample of the part to be inspected, and scan it to establish the inspection requirements of the part to be inspected including the inspection position and the characteristics of the part to be inspected;

S2: Placement of inspection parts: place the parts to be inspected on the inspection station stably as required to be inspected;

S3: Surface cleaning: use a fan to clean the surface of the component to be tested non-contact, to ensure that there is no floating ash on the surface that will affect the test results during testing;

S4: Scanning image processing: scan the component to be detected into a map, and process the scanned image to facilitate better extraction of image information;

S5: Scratch defect analysis: Identify and analyze whether there are scratch defects on the surface according to the processed image of the scanned image, and if so, further analyze the scanned scratch defects;

S6: Result display: display the scratch defects on the surface of the part scanned and various aspects of scratch defects data on the display screen, and give a corresponding treatment plan according to the data.

2 and 3, in a preferred embodiment, in S1, the scanning sample construction specifically includes the following steps:

S11: Sample scanning: find a sample of the component to be inspected, and perform 3D scanning in multiple orientations to obtain images of various orientations for analysis;

S12: Image feature extraction: extract meaningful features from the image using image segmentation technology according to the image scanned in the sample scan, and model according to the features;

S13: Detection surface selection: manually select the three-dimensional image scanned by the sample, select one or more specific surfaces to scan and analyze, and save the selected specific surface as the basis for subsequent inspection of large parts;

S14: Image array segmentation: The specific surface image formed by sample scanning and detection surface selection is combined with the data extracted in the image feature extraction, and the image of the surface to be detected is divided into equal parts, and the segmentation position is saved for subsequent detection. Basis for bulk shipment of parts;

In S12, the meaningful features in the image feature extraction include the outer contour of the image and the edges of different scanning surfaces to prepare for subsequent detection and positioning;

In S4, the scanned image processing specifically includes the following steps:

S41: Part surface scanning: scan the part to be inspected to form a scanned image;

S42: image feature comparison: according to the steps of image feature extraction and detection surface selection, the modeling of the sample is compared with the component to be detected;

S43: Secondary image array segmentation: segment the surface image of the component to be detected according to the segmentation positions stored in the image array segmentation, and divide the overall surface treatment into different array small area treatments;

S44: Image enhancement: the image after the secondary image array segmentation processing is enhanced, the high-frequency components of the image are strengthened, the clarity of the image is improved, and the protruding details in the image are emphasized;

In S42, the specific comparison method in the image feature comparison is to overlap the edge contour of the component to be detected according to the modeling in the image feature extraction. If there is a slight difference, correct the position of the sample modeling in the system to make the modeling It can be overlapped with the part to be inspected, and then the inspection surface is positioned according to the specific inspection surface selected in the inspection surface selection.

4, in a preferred embodiment, in S5, the scratch defect analysis specifically includes the following steps:

S51: Scratch identification: identify the image after the secondary image array segmentation and image enhancement, and distinguish whether there is scratch defect on the surface;

S52: Release: If no scratch defect is identified in the scratch identification, it means that there is no scratch defect on the surface of the part, and it will be released and collected immediately;

S53: Array combination analysis: If scratches are identified on the surface of the part, and the scratches span two or more divided array small areas, the scratches in different array small areas are automatically pieced together to form one Complete scratch defects;

S54: Scratch positioning: according to the identified scratches, it is positioned on the scanning surface of the part;

S55: 3D depth scanning; directly perform laser scanning on the scratches located in the scratch positioning to detect the specific depth of the scratches.

5, in a preferred embodiment, in S6, the result display specifically includes the following steps:

S61: Data measurement: scratch recognition and 3D depth scanning measure various data of scratches;

S62: Composition: compose the scanned surface according to the scanning image processing and scratch defect analysis, and only use lines to form the outline of the scanned surface and the shape and position of the scratches in the figure;

S63: Annotation: through the data measurement step, the measured various data are marked in the composition in detail;

S64: Alarm: alarm with a sound to remind the staff that there is a scratch defect on the surface of this part;

S65: Analysis of treatment plans: According to the measured scratch data, different treatment plans are analyzed, including grinding, filling or refurbishment;

In S61, the various data measured in the data measurement include the length, depth, number and location of the scratches.

Working principle: During use, the surface cleaning step uses non-contact cleaning to blow off the dust that may exist on the part to be tested to ensure the cleanliness of the surface of the part, and then scan the sample to construct the contour feature of the sample and form a modeling, and In the scanning image processing, the parts to be inspected and the edge contours matched by the modeling are overlapped. If there is a slight difference, the position of the sample modeling is corrected in the system, so that the modeling can overlap the parts to be inspected, so as to accurately control the inspection. The peripheral contour of the surface is detected, and then the detection surface is positioned according to the specific detection surface selected in the detection surface selection. During the detection, only the detection surface is detected, which can avoid the interference of other factors during the detection process and cause inaccurate detection results. the accuracy of the detection results.

The above description is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited to this. The equivalent replacement or change of the inventive concept thereof shall be included within the protection scope of the present invention.

Claims (8)

1. A method for detecting scratches and defects on the surface of metal parts of communication electrical appliances, characterized in that, comprising the following concrete steps: S1: Scanning sample construction: find a sample of the part to be inspected, and scan it to establish the inspection requirements of the part to be inspected including the inspection position and the characteristics of the part to be inspected; S2: Placement of inspection parts: place the parts to be inspected on the inspection station stably as required to be inspected; S3: Surface cleaning: use a fan to clean the surface of the component to be tested non-contact, to ensure that there is no floating ash on the surface that will affect the test results during testing; S4: Scanning image processing: scan the component to be detected into a map, and process the scanned image to facilitate better extraction of image information; S5: Scratch defect analysis: Identify and analyze whether there are scratch defects on the surface according to the processed image of the scanned image, and if so, further analyze the scanned scratch defects; S6: Result display: display the scratch defects on the surface of the part scanned and various aspects of scratch defects data on the display screen, and give a corresponding treatment plan according to the data. 2. The method for detecting scratches on the surface of metal parts of a communication electrical appliance according to claim 1, wherein in the S1, the construction of the scanned sample specifically comprises the following steps: S11: Sample scanning: find a sample of the component to be inspected, and perform 3D scanning in multiple orientations to obtain images of various orientations for analysis; S12: Image feature extraction: extract meaningful features from the image using image segmentation technology according to the image scanned in the sample scan, and model according to the features; S13: Detection surface selection: manually select the three-dimensional image scanned by the sample, select one or more specific surfaces to scan and analyze, and save the selected specific surface as the basis for subsequent inspection of large parts; S14: Image array segmentation: The specific surface image formed by the sample scanning and detection surface selection is combined with the data extracted in the image feature extraction, and the image of the surface to be detected is divided into equal parts, and the segmentation position is saved for subsequent detection. Basis when parts are in bulk. 3. The method for detecting scratches on the surface of metal parts of a communication electrical appliance according to claim 2, wherein in the S12, meaningful features in the image feature extraction include the outer contour of the image and different scanning surfaces edge, to prepare for subsequent detection and positioning. 4. The method for detecting scratches on the surface of metal parts of a communication electrical appliance according to claim 3, wherein in the S4, the scanning image processing specifically comprises the following steps: S41: Part surface scanning: scan the part to be inspected to form a scanned image; S42: image feature comparison: according to the steps of image feature extraction and detection surface selection, the modeling of the sample is compared with the component to be detected; S43: Secondary image array segmentation: segment the surface image of the component to be detected according to the segmentation positions stored in the image array segmentation, and divide the overall surface treatment into different array small area treatments; S44: Image enhancement: the image after the secondary image array segmentation processing is enhanced to enhance the high-frequency components of the image, improve the clarity of the image, and emphasize the protruding details in the image. 5. The method for detecting scratches on the surface of metal parts of communication electrical appliances according to claim 4, wherein in the S42, the specific comparison method in the image feature comparison is based on the modeling in the image feature extraction Coincidence with the edge contour of the part to be inspected, if there is a slight difference, correct the position of the sample modeling in the system, so that the modeling can coincide with the part to be inspected, and then inspect the surface according to the specific inspection surface selected in the inspection surface selection. position. 6. The method for detecting scratch defects on the surface of metal parts of a communication electrical appliance according to claim 5, wherein in the S5, the scratch defect analysis specifically comprises the following steps: S51: Scratch identification: identify the image after the secondary image array segmentation and image enhancement, and distinguish whether there is scratch defect on the surface; S52: Release: If no scratch defect is identified in the scratch identification, it means that there is no scratch defect on the surface of the part, and it will be released and collected immediately; S53: Array combination analysis: If scratches are identified on the surface of the part, and the scratches span two or more divided array small areas, the scratches in different array small areas are automatically pieced together to form one Complete scratch defects; S54: Scratch positioning: according to the identified scratches, it is positioned on the scanning surface of the part; S55: 3D depth scanning; directly perform laser scanning on the scratches located in the scratch positioning to detect the specific depth of the scratches. 7. The method for detecting scratches on the surface of metal parts of a communication electrical appliance according to claim 6, wherein in the S6, the result display specifically comprises the following steps: S61: Data measurement: scratch recognition and 3D depth scanning measure various data of scratches; S62: Composition: compose the scanned surface according to the scanning image processing and scratch defect analysis, and only use lines to form the outline of the scanned surface and the shape and position of the scratches in the figure; S63: Annotation: through the data measurement step, the measured various data are marked in the composition in detail; S64: Alarm: alarm with a sound to remind the staff that there is a scratch defect on the surface of this part; S65: Analysis of treatment plan: According to the measured scratch data, different treatment plans are analyzed, including grinding, filling or reprocessing. 8 . The method for detecting scratch defects on the surface of metal parts of a communication electrical appliance according to claim 7 , wherein in the S61 , the various data measured in the data measurement include the length, depth, number of scratches. 9 . and location.

Images