CN116977998B - Workpiece feeding visual identification system and method for coating production line - Google Patents
Workpiece feeding visual identification system and method for coating production line Download PDFInfo
- Publication number
- CN116977998B CN116977998B CN202310933428.0A CN202310933428A CN116977998B CN 116977998 B CN116977998 B CN 116977998B CN 202310933428 A CN202310933428 A CN 202310933428A CN 116977998 B CN116977998 B CN 116977998B
- Authority
- CN
- China
- Prior art keywords
- workpiece
- image
- image information
- information
- result
- 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
- 238000000576 coating method Methods 0.000 title claims abstract description 37
- 239000011248 coating agent Substances 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000000007 visual effect Effects 0.000 title claims abstract description 34
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 25
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 70
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 70
- 238000010586 diagram Methods 0.000 claims description 43
- 238000009826 distribution Methods 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 12
- 230000002194 synthesizing effect Effects 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 4
- 238000010422 painting Methods 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 description 13
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000036544 posture Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 230000003867 tiredness Effects 0.000 description 1
- 208000016255 tiredness Diseases 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/16—Image acquisition using multiple overlapping images; Image stitching
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/273—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion removing elements interfering with the pattern to be recognised
-
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/1444—Selective acquisition, locating or processing of specific regions, e.g. highlighted text, fiducial marks or predetermined fields
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/06—Recognition of objects for industrial automation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Automation & Control Theory (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention provides a visual recognition system and a recognition method for workpiece loading on a coating production line, wherein the system comprises the following components: the binocular image acquisition module is used for acquiring binocular stereoscopic image information of the workpiece; the interference information removing module is used for removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; the multidimensional information synthesis module is used for carrying out multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; and the multi-dimensional result identification module is used for determining the identification result of the current workpiece according to the multi-dimensional synthesis result.
Description
Technical Field
The invention relates to the technical field of coating equipment, in particular to a visual recognition system and a visual recognition method for workpiece loading on a coating production line.
Background
Along with the rapid development of the Chinese manufacturing industry, particularly the engineering machinery industry develops to centralized, intelligent and environment-friendly type. The intelligent system is applied, and the four processes of stamping, welding, coating and final assembly are connected efficiently. However, in the present day of automation development, the equipment has strict usage space and boundary conditions. In particular, the coating process has strict requirements on the trafficability of equipment, and the most economical and reasonable trafficability size is the key to the reasonable design of a coating line.
Meanwhile, the existing coating production line has strong compatibility, and has technological requirements of different workpieces and different postures, errors are unavoidable under the conditions of automatic workpiece feeding and manual assistance, and the existing solution method adopts a mode of scanning two-dimension codes to identify the workpieces, but the method can lead to incomplete scanning of the two-dimension codes when the workpieces with different postures are faced, so that the identification result is inaccurate, and therefore, a workpiece feeding visual identification system and a workpiece feeding visual identification method for the coating production line are needed.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a visual recognition system and a visual recognition method for workpiece loading of a coating production line, which are used for solving the problems.
A visual identification system for workpiece loading on a coating production line, comprising: the binocular image acquisition module is used for acquiring binocular stereoscopic image information of the workpiece; the interference information removing module is used for removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; the multidimensional information synthesis module is used for carrying out multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; and the multi-dimensional result identification module is used for determining the identification result of the current workpiece according to the multi-dimensional synthesis result.
As one embodiment of the present invention, the binocular image acquisition module performs operations comprising: acquiring a first user instruction, and setting a coating production line range shot by a binocular vision camera according to the first user instruction; when a new workpiece appears in the coating production line, binocular stereo image information of the new workpiece is acquired in real time based on a binocular vision camera according to preset acquisition time length, binocular stereo image information of a plurality of different time points is obtained, and acquired marks are carried out on the new workpiece after the acquisition is completed.
As one embodiment of the present invention, the interference information rejection module performs operations including: acquiring a plurality of binocular stereo image information of the same workpiece, and performing background removal processing on each binocular stereo image information according to a digital image processing method to obtain a plurality of background-free first image information; acquiring a second user instruction, and determining a pixel threshold according to the second user instruction; removing partial image information with the pixel value smaller than the pixel threshold value in the first image information based on the pixel threshold value to obtain clean image information to be identified; wherein the removed portion is replaced with a blank in the clean image information to be identified.
As one embodiment of the present invention, the multidimensional information synthesis module performs operations comprising: acquiring all the clean image information to be identified of the current workpiece, and extracting five views of the current workpiece in each clean image information to be identified; obtaining a plurality of plane diagrams of any view angle of a current workpiece as plane diagrams to be spliced, taking any one plane diagram to be spliced as a reference diagram, taking other plane diagrams to be spliced as comparison diagrams, perfecting the reference diagram according to the comparison diagrams, and determining multidimensional image information; and synthesizing the five multi-dimensional image information according to the splicing relation of the five visual angles and the stereoscopic image outline of the binocular stereoscopic image information to generate a multi-dimensional synthesis result.
As an embodiment of the present invention, perfecting a reference map according to a comparison map, determining multidimensional image information includes: step 1, determining an overlapping area of a reference image and a comparison image based on an image overlapping identification method; determining an image superposition positioning point based on the superposition area, and performing image superposition operation on the reference image and the comparison image according to the image superposition positioning point, wherein when the image superposition operation is performed, if the image of the reference image in the superposition area is not blank, the reference image is taken as the main part when the image of the reference image in the superposition area is superposed, and the image of the comparison image in the superposition area is filtered; if the image of the reference image in the overlapping area is blank, the comparison image is taken as the main part during overlapping, the blank of the reference image in the overlapping area is covered by the image of the comparison image in the overlapping area, and a new reference image is obtained after the overlapping is completed; and 2, perfecting the new reference diagram according to the new comparison diagram until the perfecting is finished when no blank exists in the latest reference diagram, and determining multidimensional image information.
As one embodiment of the present invention, the multi-dimensional result recognition module performs operations comprising: obtaining a multi-dimensional synthesis result, performing similarity matching according to the multi-dimensional synthesis result and a plurality of pieces of workpiece image information in a database based on a preset matching rule, determining pieces of workpiece image information corresponding to the multi-dimensional synthesis result, and determining the identification result of the current workpiece according to the corresponding pieces of workpiece image information.
As an embodiment of the present invention, the multi-dimensional result recognition module further performs operations comprising: when the workpiece image information successfully matched with the multi-dimensional synthesis result does not exist in the database, the current workpiece is judged to be the suspicious workpiece, a grabbing instruction is sent to the mechanical arm, and the suspicious workpiece is grabbed to the suspicious region.
As an embodiment of the present invention, before sending the grabbing instruction, the method further includes: acquiring a multi-dimensional synthesis result, and performing text extraction on the multi-dimensional synthesis result based on a picture recognition text technology to obtain multi-segment text information; determining the associated distribution characteristics of the key texts according to the positions of the key text information in the existing workpiece image information in the database; acquiring position information of a plurality of pieces of text information on a multi-dimensional synthesis result, screening the plurality of pieces of text information according to the associated distribution characteristics, and determining a plurality of pieces of key text information; determining effective text information of each position information according to the associated distribution characteristics, and carrying out text recognition and keyword extraction on the multi-section key text information according to the effective text information to obtain effective keywords; judging whether an effective keyword is obtained, if the effective keyword is empty, returning an unrecognizable instruction to a multidimensional result recognition module, wherein the multidimensional result recognition module sends an unknown in-doubt workpiece grabbing instruction to the mechanical arm, and grabs the in-doubt workpiece to an unknown in-doubt area; if the effective keywords are not empty and unique, acquiring workpiece image information corresponding to the effective keywords to determine the identification result of the current workpiece, returning the identification result to a multi-dimensional result identification module, and sending an identification suspicious workpiece grabbing instruction to the mechanical arm by the multi-dimensional result identification module to grab the suspicious workpiece to the suspicious region of the corresponding identification workpiece image information.
As an embodiment of the present invention, a visual recognition system for workpiece loading on a coating line further includes: and the cloud service module is used for displaying the doubt information, the grabbing instruction information and the system operation data.
A visual recognition method for workpiece loading on a coating production line comprises the following steps: acquiring binocular stereo image information of a workpiece, and removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; performing multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; and determining the identification result of the current workpiece according to the multi-dimensional synthesis result.
The beneficial effects of the invention are as follows:
the invention provides a workpiece feeding visual recognition system and a workpiece feeding visual recognition method for a coating production line, wherein three-dimensional image information is collected through a binocular visual system, multi-dimensional recognition is carried out on a workpiece feeding according to the three-dimensional image information, and therefore, the workpiece feeding visual recognition system is beneficial to accurately recognizing workpieces.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic block diagram of a visual recognition system and a recognition method for workpiece loading on a coating line according to an embodiment of the present invention;
FIG. 2 is a flowchart of determining multidimensional image information in a visual recognition system and a recognition method for workpiece loading on a coating production line according to an embodiment of the present invention;
fig. 3 is a flowchart of a method in a visual recognition system and a recognition method for workpiece loading on a coating line according to an embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
Referring to fig. 1, a workpiece loading visual recognition system for a coating line includes: the binocular image acquisition module 1 is used for acquiring binocular stereo image information of a workpiece; the interference information removing module 2 is used for removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; the multidimensional information synthesis module 3 is used for carrying out multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; the multidimensional result identification module 4 is used for determining the identification result of the current workpiece according to the multidimensional synthesis result;
The working principle of the technical scheme is as follows: the system comprises a binocular image acquisition module, an interference information rejection module, a multi-dimensional information synthesis module and a multi-dimensional result recognition module, wherein the binocular image acquisition module is used for acquiring images of a workpiece according to a plurality of binocular cameras which are well arranged in advance, and generating binocular stereo image information of the workpiece according to multiple images; the interference information eliminating module is used for eliminating redundant interference information in the binocular stereo image information according to a preset interference eliminating method to obtain clean image information to be identified, wherein the preset interference eliminating method comprises the processes of denoising, removing image areas lower than a preset pixel threshold value and the like; then the multidimensional information synthesis module performs multidimensional synthesis on the clean image information to be identified of the same workpiece to generate a multidimensional synthesis result; finally, the multi-dimensional result recognition module determines the recognition result of the current workpiece according to the multi-dimensional synthesis result;
the beneficial effects of the technical scheme are as follows: the binocular vision system is used for collecting three-dimensional image information, carrying out multidimensional identification on the workpiece according to the three-dimensional image information, and being beneficial to accurately identifying the workpiece.
In one embodiment, the binocular image acquisition module performs operations comprising: acquiring a first user instruction, and setting a coating production line range shot by a binocular vision camera according to the first user instruction; when a new workpiece appears in the coating production line, acquiring binocular stereo image information of the new workpiece in real time based on a binocular vision camera according to preset acquisition time length to obtain binocular stereo image information of a plurality of different time points, and carrying out acquired marking on the new workpiece after acquisition is completed;
The working principle of the technical scheme is as follows: the system comprises a binocular image acquisition module, wherein when a production line starts to work, the binocular image acquisition module acquires a first user instruction, and the range of a coating production line shot by a binocular vision camera is set according to the first user instruction; when a new workpiece appears in the coating production line, acquiring binocular stereo image information of the new workpiece in real time based on a binocular vision camera according to preset acquisition time length or the preset acquisition area, so as to obtain binocular stereo image information of a plurality of different time points, and after acquisition, marking the new workpiece after acquisition;
The beneficial effects of the technical scheme are as follows: through the technical scheme, the workpiece image information is acquired, compared with two-dimensional code scanning, three-dimensional image acquisition is more comprehensive, and binocular vision acquisition is more reliable compared with the situation that two-dimensional codes are located on an image acquisition surface.
In one embodiment, the interference information culling module performs operations comprising: acquiring a plurality of binocular stereo image information of the same workpiece, and performing background removal processing on each binocular stereo image information according to a digital image processing method to obtain a plurality of background-free first image information; acquiring a second user instruction, and determining a pixel threshold according to the second user instruction; removing partial image information with the pixel value smaller than the pixel threshold value in the first image information based on the pixel threshold value to obtain clean image information to be identified; wherein the removing part is replaced by a blank in the clean image information to be identified;
The working principle of the technical scheme is as follows: the system comprises an interference information eliminating module, wherein after corresponding data are acquired by the binocular image acquisition module, the interference information eliminating module carries out background elimination processing on each binocular stereoscopic image information according to a digital image processing method to obtain a plurality of first image information without background; acquiring a second user instruction, and determining a pixel threshold according to the second user instruction, wherein the second user instruction is related to model information of binocular vision cameras in the binocular image acquisition module, and shooting precision of different models is different, so that the pixel threshold needs to be set in a targeted manner, and the condition that the follow-up recognition precision is too low due to the same setting is avoided; removing partial image information with the pixel value smaller than the pixel threshold value in the first image information based on the pixel threshold value to obtain clean image information to be identified; wherein the removing part is replaced by a blank in the clean image information to be identified;
The beneficial effects of the technical scheme are as follows: through the technical scheme, useless information in the image information is removed, so that the subsequent image recognition effect is improved, and the recognition result is more reliable.
In one embodiment, the multidimensional information synthesis module performs operations comprising: acquiring all the clean image information to be identified of the current workpiece, and extracting five views of the current workpiece in each clean image information to be identified; obtaining a plurality of plane diagrams of any view angle of a current workpiece as plane diagrams to be spliced, taking any one plane diagram to be spliced as a reference diagram, taking other plane diagrams to be spliced as comparison diagrams, perfecting the reference diagram according to the comparison diagrams, and determining multidimensional image information; synthesizing the five multi-dimensional image information according to the splicing relation of the five visual angles and the stereoscopic image outline of the binocular stereoscopic image information to generate a multi-dimensional synthesis result;
The working principle of the technical scheme is as follows: the system comprises a multidimensional information synthesis module, wherein after the clean image information to be identified is obtained, five views of a current workpiece in each clean image information to be identified are extracted; obtaining a plurality of plane diagrams of any view angle of a current workpiece as plane diagrams to be spliced, taking any one plane diagram to be spliced as a reference diagram, taking other plane diagrams to be spliced as comparison diagrams, perfecting the reference diagram according to the comparison diagrams, and determining multidimensional image information; synthesizing the five multi-dimensional image information according to the splicing relation of the five visual angles and the three-dimensional image outline of the binocular three-dimensional image information to generate a multi-dimensional synthesis result, wherein the five views comprise a top view, a front view, a left view, a right view and a rear view;
The beneficial effects of the technical scheme are as follows: through the technical scheme, the accuracy of the image information is improved by synthesizing the image information at the multi-dimensional angle, the follow-up image recognition effect is improved, and the recognition result is more reliable.
Referring to fig. 2, in one embodiment, refining the reference map according to the comparison map, determining multidimensional image information includes: step 1, determining an overlapping area of a reference image and a comparison image based on an image overlapping identification method; determining an image superposition positioning point based on the superposition area, and performing image superposition operation on the reference image and the comparison image according to the image superposition positioning point, wherein when the image superposition operation is performed, if the image of the reference image in the superposition area is not blank, the reference image is taken as the main part when the image of the reference image in the superposition area is superposed, and the image of the comparison image in the superposition area is filtered; if the image of the reference image in the overlapping area is blank, the comparison image is taken as the main part during overlapping, the blank of the reference image in the overlapping area is covered by the image of the comparison image in the overlapping area, and a new reference image is obtained after the overlapping is completed; step 2, perfecting the new reference diagram according to the new comparison diagram until the latest reference diagram is finished perfecting when no blank exists in the latest reference diagram, and determining multidimensional image information;
The working principle of the technical scheme is as follows: determining an overlapping region of the reference image and the comparison image based on an image overlapping identification method; determining an image superposition positioning point based on the superposition area, and performing image superposition operation on the reference image and the comparison image according to the image superposition positioning point, wherein when the image superposition operation is performed, if the image of the reference image in the superposition area is not blank, the reference image is taken as the main part when the image of the reference image in the superposition area is superposed, and the image of the comparison image in the superposition area is filtered; if the images of the reference images in the overlapping area range are blank, the comparison images are taken as the main images in the overlapping process, the blank of the reference images in the overlapping area range is covered by the images of the comparison images in the overlapping area range, a new reference image is obtained after the overlapping is completed, then the new reference image is perfected according to the new comparison images, the perfecting method refers to the step 1, the perfecting is finished until the blank does not exist in the latest reference image, and multidimensional image information is determined;
The beneficial effects of the technical scheme are as follows: through the technical scheme, the accuracy of the image information is improved, and compared with the single image recognition according to the workpiece, the method is beneficial to improving the subsequent image recognition effect, so that the recognition result is more reliable.
In one embodiment, the multi-dimensional result recognition module performs operations comprising: acquiring a multi-dimensional synthesis result, performing similarity matching according to the multi-dimensional synthesis result and a plurality of pieces of workpiece image information in a database based on a preset matching rule, determining workpiece image information corresponding to the multi-dimensional synthesis result, and determining the identification result of the current workpiece according to the corresponding workpiece image information;
The beneficial effects of the technical scheme are as follows: through the technical scheme, compared with the method for carrying out similarity comprehensive calculation on multiple images of the same workpiece, the method can reduce errors generated during the similarity comprehensive calculation, and the recognition result is more reliable.
In one embodiment, the multi-dimensional result recognition module further performs operations comprising: when the workpiece image information successfully matched with the multi-dimensional synthesis result does not exist in the database, the current workpiece is judged to be the suspicious workpiece, a grabbing instruction is sent to the mechanical arm, and the suspicious workpiece is grabbed to the suspicious region.
In one embodiment, before sending the grabbing instruction, the method further comprises: acquiring a multi-dimensional synthesis result, and performing text extraction on the multi-dimensional synthesis result based on a picture recognition text technology to obtain multi-segment text information; determining the associated distribution characteristics of the key texts according to the positions of the key text information in the existing workpiece image information in the database; acquiring position information of a plurality of pieces of text information on a multi-dimensional synthesis result, screening the plurality of pieces of text information according to the associated distribution characteristics, and determining a plurality of pieces of key text information; determining effective text information of each position information according to the associated distribution characteristics, and carrying out text recognition and keyword extraction on the multi-section key text information according to the effective text information to obtain effective keywords; judging whether an effective keyword is obtained, if the effective keyword is empty, returning an unrecognizable instruction to a multidimensional result recognition module, wherein the multidimensional result recognition module sends an unknown in-doubt workpiece grabbing instruction to the mechanical arm, and grabs the in-doubt workpiece to an unknown in-doubt area; if the effective keywords are not empty and unique, acquiring workpiece image information corresponding to the effective keywords to determine the identification result of the current workpiece, returning the identification result to a multi-dimensional result identification module, and sending an identification suspicious workpiece grabbing instruction to the mechanical arm by the multi-dimensional result identification module to grab the suspicious workpiece to a suspicious region corresponding to the identification workpiece image information;
the working principle of the technical scheme is as follows: before sending a grabbing instruction, acquiring a multi-dimensional synthesis result, and carrying out text extraction on the multi-dimensional synthesis result based on a picture recognition text technology to obtain multi-segment text information; determining associated distribution characteristics of key texts according to positions of key text information in the existing workpiece image information in a database, wherein the associated distribution characteristics refer to position distribution characteristics of each different workpiece which can represent the own key text characteristics in the image; the method comprises the steps of obtaining position information of a plurality of pieces of text information on a multi-dimensional synthesis result, screening the plurality of pieces of text information according to associated distribution characteristics, and determining a plurality of pieces of key text information, wherein the method is remarkable in that when two workpieces cannot be distinguished from a key text at a certain distribution position, the key text at the plurality of distribution positions is adopted to distinguish the two workpieces; according to the associated distribution characteristics, determining effective text information of each piece of position information, namely, key text information of the existing workpiece on the position information, carrying out text recognition and keyword extraction on multiple pieces of key text information according to the effective text information to obtain effective keywords, wherein the purpose of the effective keyword extraction is to reduce recognition difficulty, and recognition errors caused by unclear images can be reduced to a great extent only by carrying out recognition extraction on keywords with unique characteristics; judging whether an effective keyword is obtained, if the effective keyword is empty, returning an unrecognizable instruction to a multidimensional result recognition module, wherein the multidimensional result recognition module sends an unknown in-doubt workpiece grabbing instruction to the mechanical arm, and grabs the in-doubt workpiece to an unknown in-doubt area; if the effective keywords are not empty and unique, acquiring workpiece image information corresponding to the effective keywords to determine the identification result of the current workpiece, returning the identification result to a multi-dimensional result identification module, and sending an identification suspicious workpiece grabbing instruction to the mechanical arm by the multi-dimensional result identification module to grab the suspicious workpiece to a suspicious region corresponding to the identification workpiece image information;
the beneficial effects of the technical scheme are as follows: through the technical scheme, the suspicious workpiece information is determined by utilizing the mode of the associated distribution characteristics of the text and the position, so that the suspicious workpiece information is convenient to grasp to the corresponding position correspondingly, the corresponding workpiece is convenient to review by subsequent staff, the inquiry in a stack of mixed workpieces is not needed, and the degree of automation is improved.
In one embodiment, a visual recognition system for workpiece loading on a coating line further comprises: and the cloud service module is used for displaying the doubt information, the grabbing instruction information and the system operation data.
Referring to fig. 3, a method for visually identifying workpiece loading on a coating line includes: s101, acquiring binocular stereo image information of a workpiece, and removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; s102, performing multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; s103, determining the identification result of the current workpiece according to the multi-dimensional synthesis result.
The working principle and the beneficial effects of the workpiece loading visual identification method of the coating production line can be referred to the working principle and the beneficial effects mentioned with respect to the corresponding modules in the workpiece loading visual identification system of the coating production line, and repeated tiredness is not needed here.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (7)
1. A workpiece loading visual identification system for a coating production line, comprising: the binocular image acquisition module is used for acquiring binocular stereoscopic image information of the workpiece; the interference information removing module is used for removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; the multidimensional information synthesis module is used for carrying out multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; the multidimensional result identification module is used for determining the identification result of the current workpiece according to the multidimensional synthesis result;
The multi-dimensional result recognition module performs operations including: acquiring a multi-dimensional synthesis result, performing similarity matching according to the multi-dimensional synthesis result and a plurality of pieces of workpiece image information in a database based on a preset matching rule, determining workpiece image information corresponding to the multi-dimensional synthesis result, and determining the identification result of the current workpiece according to the corresponding workpiece image information;
the multi-dimensional result recognition module further performs operations including: when the workpiece image information successfully matched with the multi-dimensional synthesis result does not exist in the database, judging that the current workpiece is an in-doubt workpiece, sending a grabbing instruction to the mechanical arm, and grabbing the in-doubt workpiece to an in-doubt region;
Before sending the grabbing instruction, the method further comprises the following steps: acquiring a multi-dimensional synthesis result, and performing text extraction on the multi-dimensional synthesis result based on a picture recognition text technology to obtain multi-segment text information; determining the associated distribution characteristics of the key texts according to the positions of the key text information in the existing workpiece image information in the database; acquiring position information of a plurality of pieces of text information on a multi-dimensional synthesis result, screening the plurality of pieces of text information according to the associated distribution characteristics, and determining a plurality of pieces of key text information; determining effective text information of each position information according to the associated distribution characteristics, and carrying out text recognition and keyword extraction on the multi-section key text information according to the effective text information to obtain effective keywords; judging whether an effective keyword is obtained, if the effective keyword is empty, returning an unrecognizable instruction to a multidimensional result recognition module, wherein the multidimensional result recognition module sends an unknown in-doubt workpiece grabbing instruction to the mechanical arm, and grabs the in-doubt workpiece to an unknown in-doubt area; if the effective keywords are not empty and unique, acquiring workpiece image information corresponding to the effective keywords to determine the identification result of the current workpiece, returning the identification result to a multi-dimensional result identification module, and sending an identification suspicious workpiece grabbing instruction to the mechanical arm by the multi-dimensional result identification module to grab the suspicious workpiece to the suspicious region of the corresponding identification workpiece image information.
2. A workpiece visual identification system in a coating line as claimed in claim 1, wherein the binocular image acquisition module performs operations comprising: acquiring a first user instruction, and setting a coating production line range shot by a binocular vision camera according to the first user instruction; when a new workpiece appears in the coating production line, binocular stereo image information of the new workpiece is acquired in real time based on a binocular vision camera according to preset acquisition time length, binocular stereo image information of a plurality of different time points is obtained, and acquired marks are carried out on the new workpiece after the acquisition is completed.
3. The visual recognition system of workpiece on a coating line of claim 1, wherein the disturbance information rejection module performs operations comprising: acquiring a plurality of binocular stereo image information of the same workpiece, and performing background removal processing on each binocular stereo image information according to a digital image processing method to obtain a plurality of background-free first image information; acquiring a second user instruction, and determining a pixel threshold according to the second user instruction; removing partial image information with the pixel value smaller than the pixel threshold value in the first image information based on the pixel threshold value to obtain clean image information to be identified; wherein the removed portion is replaced with a blank in the clean image information to be identified.
4. A visual recognition system for workpiece-on-a-painting line according to claim 1, wherein the multi-dimensional information synthesizing module performs operations comprising: acquiring all the clean image information to be identified of the current workpiece, and extracting five views of the current workpiece in each clean image information to be identified; obtaining a plurality of plane diagrams of any view angle of a current workpiece as plane diagrams to be spliced, taking any one plane diagram to be spliced as a reference diagram, taking other plane diagrams to be spliced as comparison diagrams, perfecting the reference diagram according to the comparison diagrams, and determining multidimensional image information; and synthesizing the five multi-dimensional image information according to the splicing relation of the five visual angles and the stereoscopic image outline of the binocular stereoscopic image information to generate a multi-dimensional synthesis result.
5. The visual recognition system for workpiece loading on a coating line of claim 4, wherein the refinement of the reference map based on the comparison map to determine the multi-dimensional image information comprises: step 1, determining an overlapping area of a reference image and a comparison image based on an image overlapping identification method; determining an image superposition positioning point based on the superposition area, and performing image superposition operation on the reference image and the comparison image according to the image superposition positioning point, wherein when the image superposition operation is performed, if the image of the reference image in the superposition area is not blank, the reference image is taken as the main part when the image of the reference image in the superposition area is superposed, and the image of the comparison image in the superposition area is filtered; if the image of the reference image in the overlapping area is blank, the comparison image is taken as the main part during overlapping, the blank of the reference image in the overlapping area is covered by the image of the comparison image in the overlapping area, and a new reference image is obtained after the overlapping is completed; and 2, perfecting the new reference diagram according to the new comparison diagram until the perfecting is finished when no blank exists in the latest reference diagram, and determining multidimensional image information.
6. A workpiece-on-workpiece visual identification system for a coating line as set forth in claim 1, further comprising: and the cloud service module is used for displaying the doubt information, the grabbing instruction information and the system operation data.
7. A visual recognition method for workpiece loading on a coating production line is characterized by comprising the following steps: acquiring binocular stereo image information of a workpiece, and removing redundant interference information in the binocular stereo image information to obtain clean image information to be identified; performing multidimensional synthesis on the clean image information to be identified to generate a multidimensional synthesis result; determining the identification result of the current workpiece according to the multi-dimensional synthesis result;
The determining the identification result of the current workpiece according to the multi-dimensional synthesis result comprises the following steps: acquiring a multi-dimensional synthesis result, performing similarity matching according to the multi-dimensional synthesis result and a plurality of pieces of workpiece image information in a database based on a preset matching rule, determining workpiece image information corresponding to the multi-dimensional synthesis result, and determining the identification result of the current workpiece according to the corresponding workpiece image information;
The step of determining the identification result of the current workpiece according to the multi-dimensional synthesis result further comprises the following steps: when the workpiece image information successfully matched with the multi-dimensional synthesis result does not exist in the database, judging that the current workpiece is an in-doubt workpiece, sending a grabbing instruction to the mechanical arm, and grabbing the in-doubt workpiece to an in-doubt region;
Before sending the grabbing instruction, the method further comprises the following steps: acquiring a multi-dimensional synthesis result, and performing text extraction on the multi-dimensional synthesis result based on a picture recognition text technology to obtain multi-segment text information; determining the associated distribution characteristics of the key texts according to the positions of the key text information in the existing workpiece image information in the database; acquiring position information of a plurality of pieces of text information on a multi-dimensional synthesis result, screening the plurality of pieces of text information according to the associated distribution characteristics, and determining a plurality of pieces of key text information; determining effective text information of each position information according to the associated distribution characteristics, and carrying out text recognition and keyword extraction on the multi-section key text information according to the effective text information to obtain effective keywords; judging whether an effective keyword is obtained, if the effective keyword is empty, returning an unrecognizable instruction to a multidimensional result recognition module, wherein the multidimensional result recognition module sends an unknown in-doubt workpiece grabbing instruction to the mechanical arm, and grabs the in-doubt workpiece to an unknown in-doubt area; if the effective keywords are not empty and unique, acquiring workpiece image information corresponding to the effective keywords to determine the identification result of the current workpiece, returning the identification result to a multi-dimensional result identification module, and sending an identification suspicious workpiece grabbing instruction to the mechanical arm by the multi-dimensional result identification module to grab the suspicious workpiece to the suspicious region of the corresponding identification workpiece image information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310933428.0A CN116977998B (en) | 2023-07-27 | 2023-07-27 | Workpiece feeding visual identification system and method for coating production line |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310933428.0A CN116977998B (en) | 2023-07-27 | 2023-07-27 | Workpiece feeding visual identification system and method for coating production line |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116977998A CN116977998A (en) | 2023-10-31 |
| CN116977998B true CN116977998B (en) | 2024-07-05 |
Family
ID=88482635
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310933428.0A Active CN116977998B (en) | 2023-07-27 | 2023-07-27 | Workpiece feeding visual identification system and method for coating production line |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116977998B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109089102A (en) * | 2018-09-05 | 2018-12-25 | 华南智能机器人创新研究院 | A kind of robotic article method for identifying and classifying and system based on binocular vision |
| CN112990148A (en) * | 2021-05-07 | 2021-06-18 | 武汉理工大学 | Target identification method and system for intelligent transfer robot |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0749872A (en) * | 1992-09-22 | 1995-02-21 | Nec Software Ltd | Automatic key word extraction system |
| JPH09101951A (en) * | 1995-10-06 | 1997-04-15 | Oki Electric Ind Co Ltd | Document retrieving device |
| JP4060559B2 (en) * | 2001-09-13 | 2008-03-12 | 株式会社東芝 | Image processing apparatus and image processing method |
| JP4772888B2 (en) * | 2009-03-27 | 2011-09-14 | シャープ株式会社 | Image processing apparatus, image forming apparatus, image processing method, program, and recording medium thereof |
| JP6455158B2 (en) * | 2015-01-09 | 2019-01-23 | ブラザー工業株式会社 | Information processing program and information processing apparatus |
| CN110322515A (en) * | 2019-07-02 | 2019-10-11 | 工极智能科技（苏州）有限公司 | Workpiece identification and grabbing point extraction method based on binocular stereo vision |
| CN112328308A (en) * | 2020-02-26 | 2021-02-05 | 北京字节跳动网络技术有限公司 | Method and device for recognizing text |
| CN114626731A (en) * | 2022-03-22 | 2022-06-14 | 平安普惠企业管理有限公司 | Risk identification method and device, electronic equipment and computer readable storage medium |
| CN114842368B (en) * | 2022-05-07 | 2023-10-03 | 中国电信股份有限公司 | Scene-based visual auxiliary information determination method, system, equipment and storage medium |
| CN116187718B (en) * | 2023-04-24 | 2023-08-04 | 深圳市宏大供应链服务有限公司 | Intelligent goods identification and sorting method and system based on computer vision |
-
2023
- 2023-07-27 CN CN202310933428.0A patent/CN116977998B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109089102A (en) * | 2018-09-05 | 2018-12-25 | 华南智能机器人创新研究院 | A kind of robotic article method for identifying and classifying and system based on binocular vision |
| CN112990148A (en) * | 2021-05-07 | 2021-06-18 | 武汉理工大学 | Target identification method and system for intelligent transfer robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116977998A (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109543753B (en) | License plate recognition method based on self-adaptive fuzzy repair mechanism | |
| CN111652085A (en) | Object identification method based on combination of 2D and 3D features | |
| CN108802041B (en) | Method for rapidly changing small sample set of screen detection | |
| WO2019236885A1 (en) | System and method for finding and classifying patterns in an image with a vision system | |
| CN113240798B (en) | Intelligent material integrity detection and configuration method based on digital twinning and AR | |
| CN116843631B (en) | 3D visual material separating method for non-standard part stacking in light steel industry | |
| CN113628159A (en) | Full-automatic training method and device based on deep learning network and storage medium | |
| CN110992410B (en) | Robot vision guiding method and device based on RGB-D data fusion | |
| CN113963129A (en) | Point cloud-based ship small component template matching and online identification method | |
| CN116337887A (en) | Method and system for detecting defects on upper surface of casting cylinder body | |
| CN111401184A (en) | Machine vision processing method and device, storage medium and electronic equipment | |
| CN116977998B (en) | Workpiece feeding visual identification system and method for coating production line | |
| CN114463425A (en) | Workpiece surface featureless point positioning method based on probability Hough linear detection | |
| CN113936291A (en) | Aluminum template quality inspection and recovery method based on machine vision | |
| CN113706501A (en) | Intelligent monitoring method for airplane assembly | |
| CN117001659A (en) | Automatic grabbing method, device and equipment for rubber blocks and storage medium | |
| CN116001000A (en) | Intelligent control method and system for PCB (printed circuit board) dividing machine | |
| CN113420839B (en) | Semi-automatic labeling method and segmentation positioning system for stacking planar target objects | |
| CN114596439A (en) | Image-based intelligent damage assessment and check method and device for car insurance and electronic equipment | |
| CN114723748B (en) | Detection method, device and equipment of motor controller and storage medium | |
| WO2024065189A1 (en) | Method, system, apparatus, electronic device, and storage medium for evaluating work task | |
| CN116542940A (en) | Automatic flexible visual recognition method and system for large-format glass contour | |
| CN116052185B (en) | System and method for identifying and engraving depth detection of vehicle VIN (vehicle identification number) code matched with template | |
| CN114708155A (en) | CNN-based feature space matching positioning system and method | |
| CN115661305A (en) | Method and device for collaborative annotation of 2D and 3D images, electronic equipment and storage medium |
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 |