CN113220924A - Product model visual identification method and visual identification system - Google Patents
Product model visual identification method and visual identification system Download PDFInfo
- Publication number
- CN113220924A CN113220924A CN202110397807.3A CN202110397807A CN113220924A CN 113220924 A CN113220924 A CN 113220924A CN 202110397807 A CN202110397807 A CN 202110397807A CN 113220924 A CN113220924 A CN 113220924A
- Authority
- CN
- China
- Prior art keywords
- product
- model
- feature
- image processing
- processing system
- 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.)
- Pending
Links
- 230000000007 visual effect Effects 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012545 processing Methods 0.000 claims abstract description 84
- 238000003708 edge detection Methods 0.000 claims description 12
- 238000005259 measurement Methods 0.000 claims description 8
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000007569 slipcasting Methods 0.000 claims description 5
- 229910000831 Steel Inorganic materials 0.000 claims description 4
- 239000010959 steel Substances 0.000 claims description 4
- 239000000178 monomer Substances 0.000 claims description 2
- 239000000047 product Substances 0.000 description 186
- 238000004519 manufacturing process Methods 0.000 description 18
- 230000008569 process Effects 0.000 description 12
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 8
- 230000001502 supplementing effect Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 239000010865 sewage Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005498 polishing Methods 0.000 description 2
- 238000005507 spraying Methods 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5854—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using shape and object relationship
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/5862—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/04—Manufacturing
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/005—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles during manufacturing process
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of visual identification, in particular to a visual identification method and a visual identification system for a product model, which comprises the following steps: s1, acquiring side-view and oblique-view images of the product and inputting the images of the two views into an image processing system; s2, the image processing system detects and identifies according to the side view to obtain the type characteristic of the product, and the image processing system detects and identifies according to the oblique view to obtain the model characteristic A of the product; s3, combining the type characteristics A with the model characteristics A to obtain a product characteristic combination; and S4, judging whether the product database has a model product corresponding to the product characteristic combination, and if so, outputting the corresponding product model to the control system. The invention can save manpower, improve the recognition accuracy and ensure the accuracy of the product model input; by simultaneously comparing the type characteristic and the model characteristic A, the comparison between the product and the model is more accurate.
Description
Technical Field
The invention relates to the technical field of visual identification, in particular to a product model visual identification method and a visual identification system.
Background
For a production line capable of processing products of various types or models, each product needs to be identified in the production and processing process so as to judge the specific model, and then the corresponding model is input to complete the corresponding processing step. At present, more, manually comparing a product manual with the appearance of a product, judging the model of the product, and manually inputting the model into a control system; the traditional mode has repeated work and low efficiency, and simultaneously has the problems of wrong judgment and wrong input of workers, so that the accuracy is low, and the processing efficiency and the yield are influenced. For example, in the production process of sanitary ware, when products of different models are subjected to processes of polishing, glaze spraying, carrying and the like, automation equipment needs to call corresponding formulas according to the models of the different products; in the prior stage of the bathroom industry, the models are manually input, workers are used for contrasting product manuals and product appearances, and the product models are manually input into the control system, so that the work is repeated, the efficiency is low, and mistakes are easily made.
Disclosure of Invention
The invention aims to provide a product model visual identification method and a visual identification system, and aims to solve the technical problems of low production efficiency, high error rate and low yield caused by manual product model identification in the prior art.
In order to achieve the above object, the present invention provides a product model visual identification method, which comprises the following steps:
s1, acquiring side-view and oblique-view images of the product and inputting the images of the two views into an image processing system;
s2, the image processing system detects and identifies according to the side view to obtain the type characteristic of the product, and the image processing system detects and identifies according to the oblique view to obtain the model characteristic A of the product;
s3, combining the type characteristics A with the model characteristics A to obtain a product characteristic combination;
and S4, judging whether the product database has a model product corresponding to the product characteristic combination, and if so, outputting the corresponding product model to the control system.
Preferably, in step S1, a top view image of the product is also acquired and input into the image processing system;
in step S2, the image processing system detects and identifies according to the top view to obtain the model feature B of the product;
in step S3, the genre feature, the model feature a, and the model feature B are combined to obtain a product feature combination.
Preferably, the model feature B comprises a first component feature and a second component feature.
Preferably, in step S2, the category features include a type feature, a height feature and a length feature.
Preferably, in step S2, the image processing system performs product color recognition based on the obtained image to obtain product color features; in step S3, the product feature combination further includes product color features.
Preferably, in step S2, the image processing system performs recognition to obtain code texture features according to the obtained image; in step S3, the product feature combination further includes a code texture feature.
Preferably, the code texture features comprise at least one of characters, numbers, letters, symbols and images; the forming mode of the code texture is at least one of slip casting, label pasting and steel seal printing.
Preferably, various products are placed on the visual recognition station, product models are imported or product characteristic images are collected, a deep learning network is established, the corresponding relation between the product models and the characteristic combinations is established, a product database is established, and the product database is stored in the image processing system.
Preferably, the image processing system comprises at least one image processing operator of an image binarization processing module, a fringe enhancement module, a Sobel edge detection module, a Canny edge detection module, a contour searching module, a single body measurement module and a relation measurement module.
On the other hand, the invention also provides a product model visual identification system, which uses any one visual identification method as described above, wherein the visual identification system comprises an industrial camera, an image acquisition card, an industrial control computer and the image processing system; the image processing system is installed on the industrial control computer, the industrial camera is in data connection with the image acquisition cards, and the number of the image acquisition cards is in data connection with the industrial control computer.
The invention discloses a product model visual identification method and a visual identification system, which at least have the following beneficial effects: by adopting the visual identification method, the work of manually repeating the singleness is saved, the identification accuracy is improved, and the accuracy of inputting the model of the product is ensured; by simultaneously comparing the type characteristic and the model characteristic A, the comparison between the product and the model is more accurate, and the product identification accuracy can be ensured for a product production line with multiple and complicated models.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
FIG. 1 is a flow chart illustrating the steps of the visual identification method according to the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative positional relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
In addition, if there is a description of "first", "second", etc. in an embodiment of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In addition, technical solutions between various embodiments may be combined with each other, but must be realized by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination should not be considered to exist, and is not within the protection scope of the present invention.
As shown in fig. 1, a visual identification method for product models includes the following steps:
s1, acquiring side-view and oblique-view images of the product and inputting the images of the two views into an image processing system;
s2, the image processing system detects and identifies according to the side view to obtain the type characteristic of the product, and the image processing system detects and identifies according to the oblique view to obtain the model characteristic A of the product;
s3, combining the type characteristics A with the model characteristics A to obtain a product characteristic combination;
s4, judging whether the product database has a model product corresponding to the product characteristic combination, if so, outputting the corresponding product model to the control system; if not, an alarm signal is output to the control system.
In the production and processing process of the product, the specific model of the product needs to be identified first, so that the production system can process the product correspondingly. The products are circulated on a production line to be transported to a visual recognition station, and the industrial cameras take pictures of the products, wherein one industrial camera takes a side view (left view or right view) of the products, and the other industrial camera takes an oblique view of the products; inputting the shot picture into an image processing system; the image processing system detects and identifies the oblique view to obtain the model characteristic A of the product; combining the type characteristic and the model characteristic A to form a characteristic combination of the product; searching in a product database according to the characteristic combination, judging whether the characteristic combination of the corresponding product in the product database corresponds to the characteristic combination of the product, if so, outputting the model of the corresponding product to a control system to inform the control system of the specific model of the product, and facilitating the control system to control a processing station to process the product; if not, an alarm signal is output to the control system to inform the staff that the product does not have a corresponding model, the processing cannot be carried out, and the staff is required to confirm a specific reason (whether the model of the product is not stored in the product database or the product characteristic identification is wrong).
According to the technical scheme, an image processing system identifies and acquires a type characteristic and a product model characteristic A through collected side views and oblique views, a product characteristic combination is formed through combination, and then the product characteristic combination is searched in a product database to obtain the specific model of the product; by adopting the visual identification method, the work of manually repeating the singleness is saved, the identification accuracy is improved, and the accuracy of inputting the model of the product is ensured; by simultaneously comparing the type characteristic and the model characteristic A, the comparison between the product and the model is more accurate, and the product identification accuracy can be ensured for a product production line with multiple and complicated models.
Further, in step S1, a top view image of the product is also acquired and input into the image processing system;
in step S2, the image processing system detects and identifies according to the top view to obtain the model feature B of the product;
in step S3, the genre feature, the model feature a, and the model feature B are combined to obtain a product feature combination.
In step S2, the industrial camera simultaneously takes a top view of the product and inputs the taken picture into the image processing system; the image processing system detects and identifies the top view to obtain a model feature B; and combining the model characteristic B with the category characteristic and the model characteristic A to form a product characteristic combination. By adding the top view of the shot product, the image recognition system can detect and recognize the model number characteristic B, the product characteristic combination is more perfect, the product model recognition precision is higher, and the requirement of a more-model and more-complex product processing production line can be met.
Further, the model feature B includes a first component feature and a second component feature.
Through the shot top view, the image processing system can detect and identify the first component characteristic and the second component characteristic and combine the first component characteristic and the second component characteristic into a model characteristic B; for example, when the visual identification method is used to identify a sanitary ware product, the first component feature can be a seat ring feature and the second component feature can be a tank feature; and identifying different seat rings and water tanks to obtain different model characteristics B so as to match different product models. The image processing system can detect and identify the first component characteristic and the second component characteristic by shooting the top view, fewer pictures need to be shot, more characteristics can be detected, and the corresponding identification accuracy is higher.
Further, in step S2, the category features include a type feature, a height feature and a length feature.
The image processing system performs feature detection on the product according to the side view to acquire a type feature, a height feature and a length feature of the product; and combining the acquired type feature, height feature and length feature to obtain a category feature. For example, in the production process of sanitary ware, the type characteristics can be the product appearance characteristics of sanitary ware products such as a conjoined closestool, a split closestool, a wall-hung closestool, a counter basin, a water tank, a urinal, a bathtub and the like; the height characteristic and the length characteristic are the external dimension characteristics of the product. The type features contain more subdivision features, the detection precision is higher, and the model matching is more accurate.
Further, in step S2, the image processing system performs product color recognition based on the obtained image to obtain product color features; in step S3, the product feature combination further includes product color features.
The existing products are various and high in individuation, the types, sizes, textures and outlines of partial products are always different in color, so that the color characteristics of the products are obtained by adding the colors for detecting and identifying the products, the visual identification method is suitable for detecting and identifying the products with the consistent other parameters and the difference only in color, and the application range is wider.
Further, in step S2, the image processing system performs recognition to obtain code texture features according to the obtained image; in step S3, the product feature combination further includes a code texture feature.
In the production and processing processes of some products, code textures are added on the products, which is equivalent to giving the products an identity number; the image processing system identifies according to the obtained image to obtain the code texture feature, and the code texture feature also belongs to one part of the product feature combination, so that the subdivision feature of the product feature combination is more perfect, a specific product can be identified and matched, and the method is suitable for a production line for producing personalized products.
Further, the code texture features comprise at least one of characters, numbers, letters, symbols and images; the forming mode of the code texture is at least one of slip casting, label pasting and steel seal printing.
In particular, the code texture may be words, numbers, letters, symbols, images, or combinations thereof, such as one or more words alone, or a combination of words and numbers, or a combination of numbers, letters and symbols, and so forth. The formation mode of the code texture is also various, and can be slip casting, most of the production of the sanitary ware is slip casting, so the code texture can be synchronously produced in the forming process; the code texture can also be a label, namely, a unique label is adhered to a product, and the mode can not influence the product and is suitable for a processing process with higher requirements on the surface of the product; the code texture can also be steel seal printing which is mainly used for printing raised marks such as characters, numbers and the like on a product by using a seal. The code texture features are various, the code texture features can be formulated according to the type of the product, the definition of the collected image and the processing precision of an image processing system, and the production mode of the code texture features can be selected according to the type of the product, the requirement on the appearance of the product and the like.
Further, various products are placed on the visual recognition station, product models are led in or product characteristic images are collected, a deep learning network is established, the corresponding relation between the product models and the characteristic combinations is established, a product database is established, and the product database is stored in the image processing system.
The characteristic image of the product can be acquired by photographing and identifying modes, or a model of the product is directly manufactured, characteristic information is labeled, characteristic classification is carried out, and a deep learning network is established; establishing the relationship between the model of the product and the feature combination, such as: the characteristic combination of the product model 1 is type 1 characteristic + length 1 characteristic + height 1 characteristic, and the characteristic combination of the product model 2 is type 2 characteristic + back 1 characteristic + water tank 1 characteristic; a product database is established through information acquisition, characteristic classification and model and characteristic matching, and various models of products and various characteristics of each model of product are stored in the product database.
Further, the image processing system comprises at least one image processing operator in an image binarization processing module, a fringe enhancement module, a Sobel edge detection module, a Canny edge detection module, a contour searching module, a single body measurement module and a relation measurement module.
The image processing operator is an operator used for processing the image, and comprises a global feature description operator and a local feature description operator. The image processing system at least comprises one of an image binarization processing module, a stripe enhancement module, a Sobel edge detection module, a Canny edge detection module, a contour searching module and a monomer measurement module. The image binarization processing module sets the gray value of a pixel point on an image to be 0 or 255, namely, the whole image presents an obvious black-and-white effect; the image binarization processing module can greatly reduce the data volume in the image, thereby highlighting the outline of the target. The streak enhancement module can process the shot picture to strengthen the streak part in the image. The edge is a place where the pixel value is in transition (a place with the maximum change rate and a place with the maximum derivative), is one of the significant features of the image, and has important functions in the aspects of image feature extraction, object detection, pattern recognition and the like; the Sobel edge detection module has higher efficiency than canny edge detection in practical application, and is often used when the requirement on efficiency is higher and the requirement on fine texture precision is lower; soble edge detection is usually directional and can detect only vertical edges or both. The Canny edge detection module is not easily disturbed by noise, can detect true weak edges, detects strong edges and weak edges respectively using two different thresholds, and includes weak edges in the output image when the weak edges and the strong edges are connected. The profile searching module can search the profile of the product from the shot image and identify the shape of the product. The single body measuring module can measure the shot image, measure the parameter information of the area, the width, the height, the perimeter, the inclination angle and the like of the outline or the shape, and the measured information is the characteristic information of the product. The relation measurement module can measure the shot images, measure parameter information such as the center distance, the shortest distance, the straight line included angle and the like among the outlines or the shapes, and the measured information is the characteristic information of the product. The image processing system may include one or more of the above modules, and may be selected according to the identification features and the identification process of the product in practical applications.
On the other hand, a product model visual identification system uses any one visual identification method as described above, and the visual identification system comprises an industrial camera, an image acquisition card, an industrial control computer and the image processing system; the image processing system is installed on the industrial control computer, the industrial camera is in data connection with the image acquisition card, and the number of the image acquisition cards is in data connection with the industrial control computer
The industrial camera is used for acquiring images of products; the image acquisition card is used for acquiring image information; the industrial control computer is a hardware basis for data processing of the image processing system and is an interactive platform for debugging, modifying and monitoring the image processing system; the image processing system is used for processing the image information to obtain product characteristics. The industrial camera and the image acquisition card are responsible for image acquisition, and the industrial control computer and the image processing system are responsible for image processing and result output. The visual identification system uses the visual identification method as described above, so the beneficial effects of the visual identification method are included, and the details are not repeated herein.
The visual recognition system specifically comprises a fixing support, a light supplementing lamp, a background plate and the like, wherein the fixing support is used for fixing the industrial cameras, each industrial camera is correspondingly installed and fixed on one fixing support, and the arrangement position of each fixing support needs to ensure that the industrial cameras can shoot photos of products at corresponding angles; the light supplementing lamp is used for supplementing light to the product, so that the appearance, the outline and the like of the product can be shot clearly by the industrial camera, and particularly, the light supplementing lamp can be installed at the level or the top of the product to ensure the light supplementing effect; the background plate is used for setting off the product, makes the product show more clearly in the photo of shooing, contains more irrelevant object on the photo of avoiding shooing, and the background plate is specifically placed in the opposite side of shooing the product, and the background plate generally chooses the solid background board for use, and the colour of background board can select according to the colour of product, and the main principle is that the difference in colour of product and background board is great, guarantees to show appearance, the profile of product clearly.
For the convenience of understanding, the application of the visual identification method to the identification of the type of the sanitary ware will be described in detail below.
Firstly, a vision recognition system is installed on a production line, namely three fixing supports are installed on a vision recognition station, then three industrial cameras for shooting a side view, an oblique view and a top view are respectively installed on the three fixing supports, then horizontal and top light supplement lamps are installed, and a background plate is installed on the other side of a shot product (because the color of a sanitary ware product is generally white or beige, a pure black background plate can be selected); and adjusting parameters and positions of the industrial camera according to the shot images, and adjusting the position, orientation and brightness of the light supplement lamp.
Secondly, writing an identification flow in an image processing software platform (including an image processing system) according to the characteristics of the product, such as the size, the texture, the appearance contour and the characteristics of the product, namely formulating a specific identification process; placing various products on a visual identification station, introducing a product model or collecting a product characteristic image, labeling characteristic information, classifying characteristics, establishing a deep learning network, establishing a corresponding relation between a product model and a characteristic combination, and establishing a product database; the relationship between the model number and the feature combination of the product is as follows: the product model 1 combination feature is type 1 feature + length 1 feature + height 1 feature, and the product model 2 combination feature is type 2 feature + back 1 feature + water tank 1 feature.
Then, the visual recognition system and the visual recognition method are formally applied to the production process. The sanitary ware product is conveyed to a visual identification station through a conveying line, and the three industrial cameras photograph the sanitary ware product to obtain a side view, an oblique view and a top view of the product; the three pictures are input into an image processing system. The image processing system detects and identifies the side view, classifies products into corresponding product types according to type characteristics, height characteristics and length characteristics (the three are unified into type characteristics) acquired by the side view, namely, classifies the products into specific sanitary ware products such as a conjoined closestool, a split closestool, a wall-hung closestool, a counter basin, a water tank, a urinal, a bathtub and the like, and the height characteristics and the length characteristics represent the overall dimensions of the products. The image processing system detects and identifies the oblique view, and obtains the back characteristics (namely model characteristics A) of the product according to the oblique view, wherein the back characteristics comprise back shape radian, sewage discharge pipe shape, sewage discharge outlet position, whether the sewage discharge pipe is exposed or not and the like. And the image processing system detects and identifies the top view, and obtains the seat ring characteristics (namely the first component characteristics) and the water tank characteristics (namely the second component characteristics) of the product according to the top view, wherein the seat ring characteristics comprise the outer radian of the seat ring, the shape and the size of the inner ring and the like, and the water tank characteristics comprise the shape and the size of the water tank, the set position and the like. Further, the image processing system obtains product color features and/or code texture features from the side, oblique, and top views.
Combining the obtained characteristics to form a product characteristic combination, searching out a corresponding product model in a product database according to the product characteristic combination, and inputting the corresponding product model into a control system; and if the corresponding product model cannot be searched, outputting an alarm signal to the control system. If the control system receives a specific product model, the control system indicates that the product subjected to visual identification needs to be processed on a production line, and therefore, on stations such as a polishing station, a glaze spraying station and a carrying station, automatic equipment carries out process processing corresponding to the model according to the product model.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all modifications and equivalents of the present invention, which are made by the contents of the present specification and the accompanying drawings, or directly/indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A visual identification method for product models is characterized by comprising the following steps:
s1, acquiring side-view and oblique-view images of the product and inputting the images of the two views into an image processing system;
s2, the image processing system detects and identifies according to the side view to obtain the type characteristic of the product, and the image processing system detects and identifies according to the oblique view to obtain the model characteristic A of the product;
s3, combining the type characteristics A with the model characteristics A to obtain a product characteristic combination;
and S4, judging whether the product database has a model product corresponding to the product characteristic combination, and if so, outputting the corresponding product model to the control system.
2. The visual identification method of product type as claimed in claim 1,
in step S1, simultaneously acquiring an overhead view image of the product and inputting the image into an image processing system;
in step S2, the image processing system detects and identifies according to the top view to obtain the model feature B of the product;
in step S3, the genre feature, the model feature a, and the model feature B are combined to obtain a product feature combination.
3. The visual identification method of product model according to claim 2, wherein the model feature B includes a first component feature and a second component feature.
4. The method for visually recognizing the product type as claimed in claim 1, wherein in step S2, the category feature includes a type feature, a height feature and a length feature.
5. The visual identification method of product type as claimed in claim 1, wherein in step S2, the image processing system performs product color identification based on the obtained image to obtain product color characteristics; in step S3, the product feature combination further includes product color features.
6. The visual identification method of product models according to claim 1, wherein in step S2, the image processing system performs identification according to the obtained images to obtain texture features of code numbers; in step S3, the product feature combination further includes a code texture feature.
7. The visual identification method for the product model according to claim 6, wherein the code texture features comprise at least one of characters, numbers, letters, symbols and images; the forming mode of the code texture is at least one of slip casting, label pasting and steel seal printing.
8. The visual identification method of the product model according to claim 1, characterized in that, various products are placed on a visual identification station, product models are imported or product feature images are collected, a deep learning network is established, a corresponding relation between the product model and the feature combination is established, a product database is established, and the product database is stored in an image processing system.
9. The visual identification method of the product model according to claim 1, wherein the image processing system comprises at least one image processing operator of an image binarization processing module, a streak enhancement module, a Sobel edge detection module, a Canny edge detection module, a contour finding module, a monomer measurement module and a relation measurement module.
10. A visual recognition system for product model, characterized in that the visual recognition method of any one of claims 1 to 9 is used, the visual recognition system comprises an industrial camera, an image acquisition card, an industrial control computer and the image processing system; the image processing system is installed on the industrial control computer, the industrial camera is in data connection with the image acquisition card, and the image acquisition card is in data connection with the industrial control computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110397807.3A CN113220924A (en) | 2021-04-13 | 2021-04-13 | Product model visual identification method and visual identification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110397807.3A CN113220924A (en) | 2021-04-13 | 2021-04-13 | Product model visual identification method and visual identification system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113220924A true CN113220924A (en) | 2021-08-06 |
Family
ID=77087261
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110397807.3A Pending CN113220924A (en) | 2021-04-13 | 2021-04-13 | Product model visual identification method and visual identification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113220924A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743296A (en) * | 2021-09-03 | 2021-12-03 | 广东伟兴电子科技有限公司 | Industrial product identification and traceability realization method based on 5G network |
| CN117593297A (en) * | 2024-01-18 | 2024-02-23 | 福州市展凌智能科技有限公司 | Jig identification and counting system and method based on image processing |
| CN118443681A (en) * | 2024-07-05 | 2024-08-06 | 马夸特开关(威海)有限公司 | Potting AOI multi-point identification detection control system and image data processing method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08155766A (en) * | 1994-12-07 | 1996-06-18 | Toyota Motor Corp | Equipment for inspecting incorrect or defective article |
| CN109241948A (en) * | 2018-10-18 | 2019-01-18 | 杜海朋 | A kind of NC cutting tool visual identity method and device |
| CN209531485U (en) * | 2018-04-09 | 2019-10-25 | 常州轻工职业技术学院 | Automatic sorting device based on image recognition |
| CN111046722A (en) * | 2019-10-16 | 2020-04-21 | 深圳市美思美科智能科技股份有限公司 | Intelligent visual auxiliary material identification system |
-
2021
- 2021-04-13 CN CN202110397807.3A patent/CN113220924A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08155766A (en) * | 1994-12-07 | 1996-06-18 | Toyota Motor Corp | Equipment for inspecting incorrect or defective article |
| CN209531485U (en) * | 2018-04-09 | 2019-10-25 | 常州轻工职业技术学院 | Automatic sorting device based on image recognition |
| CN109241948A (en) * | 2018-10-18 | 2019-01-18 | 杜海朋 | A kind of NC cutting tool visual identity method and device |
| CN111046722A (en) * | 2019-10-16 | 2020-04-21 | 深圳市美思美科智能科技股份有限公司 | Intelligent visual auxiliary material identification system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743296A (en) * | 2021-09-03 | 2021-12-03 | 广东伟兴电子科技有限公司 | Industrial product identification and traceability realization method based on 5G network |
| CN113743296B (en) * | 2021-09-03 | 2024-05-31 | 揭阳职业技术学院 | Industrial product identification and traceability implementation method based on 5G network |
| CN117593297A (en) * | 2024-01-18 | 2024-02-23 | 福州市展凌智能科技有限公司 | Jig identification and counting system and method based on image processing |
| CN117593297B (en) * | 2024-01-18 | 2024-04-19 | 福州市展凌智能科技有限公司 | Jig identification and counting system and method based on image processing |
| CN118443681A (en) * | 2024-07-05 | 2024-08-06 | 马夸特开关(威海)有限公司 | Potting AOI multi-point identification detection control system and image data processing method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113220924A (en) | Product model visual identification method and visual identification system | |
| CN109829914B (en) | Method and device for detecting product defects | |
| CN107941808B (en) | 3D printing forming quality detection system and method based on machine vision | |
| CN109785316B (en) | Method for detecting apparent defects of chip | |
| CN107194395B (en) | Object dynamic positioning method based on color identification and contour extraction | |
| CN105069466B (en) | Pedestrian's dress ornament color identification method based on Digital Image Processing | |
| CN106340044B (en) | Join automatic calibration method and caliberating device outside video camera | |
| CN106251353A (en) | Weak texture workpiece and the recognition detection method and system of three-dimensional pose thereof | |
| CN107945184A (en) | A kind of mount components detection method positioned based on color images and gradient projection | |
| CN107945155B (en) | Toothpaste tube shoulder defect detection method based on Gabor filter | |
| CN108918526A (en) | A kind of chips defect detection method of flexibility IC package base plate line | |
| CN110567976B (en) | Mobile phone cover plate silk-screen defect detection device and detection method based on machine vision | |
| CN109993154B (en) | Intelligent identification method for single-pointer sulfur hexafluoride instrument of transformer substation | |
| CN108710876A (en) | A kind of battery surface mark defect inspection method and system based on machine vision | |
| CN108971025A (en) | A kind of visible detection method and vision detection system | |
| CN115359047B (en) | Abnormal defect detection method for intelligent welding of PCB | |
| CN108709500B (en) | Circuit board element positioning and matching method | |
| CN112345534A (en) | Vision-based bubble plate particle defect detection method and system | |
| CN111681220B (en) | Method, device, system and storage medium for constructing defect detection model | |
| CN114926410A (en) | Method for detecting appearance defects of brake disc | |
| CN110533660B (en) | Method for detecting silk-screen defect of electronic product shell | |
| CN108182707B (en) | Chessboard grid calibration template under incomplete collection condition and automatic identification method thereof | |
| CN109060799A (en) | A kind of assembling line finished product detection determination method | |
| CN111932617B (en) | Method and system for realizing real-time detection and positioning of regular objects | |
| CN105303565A (en) | Product appearance LOGO detection method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210806 |
|
| RJ01 | Rejection of invention patent application after publication |