CN115672778A - Intelligent visual recognition system - Google Patents
Intelligent visual recognition system Download PDFInfo
- Publication number
- CN115672778A CN115672778A CN202211316732.2A CN202211316732A CN115672778A CN 115672778 A CN115672778 A CN 115672778A CN 202211316732 A CN202211316732 A CN 202211316732A CN 115672778 A CN115672778 A CN 115672778A
- Authority
- CN
- China
- Prior art keywords
- image
- unit
- module
- sending
- sorting
- 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.)
- Withdrawn
Links
Images
Landscapes
- Image Processing (AREA)
Abstract
The invention is suitable for the technical field of visual recognition, and provides an intelligent visual recognition system, which comprises: the image acquisition module is used for acquiring an image of a product and sending the acquired image to the image processing module; the image processing module is used for receiving the image sent by the image acquisition module, carrying out sharpening processing on the image and sending the processed image to the identification module; the recognition module is used for receiving the image sent by the image processing module, comparing and classifying the image with the image in the image storage module, and sending an instruction to the sorting module; and the sorting module is used for receiving the instruction sent by the identification module and sorting the products. The identification module can classify the images, generate different sorting instructions and send the instructions to the sorting module to sort the products, so that the product sorting module not only can distinguish whether the products are qualified, but also can subdivide different defective products, and is favorable for subsequent sorting and quality analysis.
Description
Technical Field
The invention belongs to the technical field of visual recognition, and particularly relates to an intelligent visual recognition system.
Background
The image information becomes an important source for acquiring information and an important means for utilizing the information by human beings due to a series of advantages of large information amount, high transmission speed, long acting distance and the like, the efficiency and the accuracy of manual processing identification are difficult to meet the requirements as the society develops into an information era, and in the current big data era, the intelligent visual identification system can be applied to various fields of life, work and the like, such as product detection.
The intelligent visual recognition system is vital in manufacturing enterprises, can help quickly complete the defect detection work of product quality, improves the reliability of product quality detection, and effectively avoids subjectivity and individual difference in the process of manual naked eye detection.
However, most of the existing intelligent visual recognition systems can only distinguish qualified products from defective products, and different types of defective products cannot be subdivided, which brings inconvenience to subsequent sorting and analysis.
Disclosure of Invention
The embodiment of the invention aims to provide an intelligent visual recognition system, which aims to solve the technical problems in the prior art determined in the background technology.
The embodiment of the invention is realized in such a way that an intelligent visual recognition system comprises:
the image acquisition module is used for acquiring an image of a product and sending the acquired image to the image processing module;
the image processing module is used for receiving the image sent by the image acquisition module, carrying out sharpening processing on the image and sending the processed image to the recognition module;
the identification module is used for receiving the image sent by the image processing module, comparing and classifying the image with the image in the image storage module and sending an instruction to the sorting module;
and the sorting module is used for receiving the instruction sent by the identification module and sorting the products.
Preferably, the image acquisition module comprises:
the shooting unit is used for acquiring an image of a product;
the light supplementing unit is used for supplementing a light source in the shooting range of the shooting unit;
and the first sending unit is used for sending the acquired image to the image processing module.
Preferably, the image processing module includes:
the first receiving unit is used for receiving the image sent by the image acquisition module;
the noise reduction unit is used for reducing noise in the image;
the clear restoration unit is used for carrying out clear processing on the denoised image;
and the second sending unit is used for sending the processed image to the identification module.
Preferably, the image processing module further comprises a detection unit, and the detection unit is used for detecting whether the image is clear; when the image is clear, the detection unit is used for sending the image to the second sending unit; and when the image is not clear, the detection unit is used for sending the image to the clear restoration unit.
Preferably, the identification module comprises:
the second receiving unit is used for receiving the image sent by the image processing module;
the analysis unit is used for sequentially analyzing and comparing the images with the defective product images in the image storage module;
the classification unit is used for classifying the images according to the analysis and comparison results of the analysis unit so as to generate different sorting instructions;
and the third sending unit is used for sending the sorting instruction to the sorting module.
Preferably, the identification module further comprises a statistic unit, and the statistic unit is used for counting the number of different inferior-quality products according to the analysis and comparison result of the analysis unit to generate data information.
Preferably, the identification module further comprises an output unit for outputting data showing the number of inferior goods according to the data information of the statistical unit.
Preferably, the sorting module comprises:
the third receiving unit is used for receiving the instruction sent by the identification module;
and the execution unit is used for operating the machine according to the instruction of the third receiving unit so as to sort the products.
According to the intelligent visual identification system provided by the embodiment of the invention, the image acquisition module is used for acquiring the image of the product and then sending the acquired image to the image processing module, when the image is acquired, multi-angle acquisition can be selected, images of different angles of the product are acquired at the same time, so that the quality of the product can be conveniently and accurately detected and judged in the follow-up process, and whether the product is qualified or defective is judged according to the acquired image; the image processing module processes the received image, and carries out sharpening processing on the fuzzy part in the image, so that errors in subsequent comparison are avoided, misjudgment and misjudgment are avoided, the recognition and sorting accuracy is improved, and the comparison and judgment efficiency is also improved; the recognition module receives the processed images, the image storage module stores images of qualified products and images of different defects of the products, and after the recognition module receives the processed clear images, the recognition module sequentially traverses and compares the images with the images in the image storage module until the images are compared to be accordant, so that the images can be classified, different sorting instructions are generated, and the instructions are sent to the sorting module; when the letter sorting module received the letter sorting instruction, can sort the operation to the product, according to the difference of letter sorting instruction, with the product classification to different positions, through the letter sorting instruction of difference, not only can be qualified to distinguish the product, can also subdivide different substandard products, classify the substandard product according to the defect, be favorable to subsequent letter sorting and quality analysis.
Drawings
Fig. 1 is a block diagram of an intelligent visual recognition system according to an embodiment of the present invention;
fig. 2 is a block diagram of an image acquisition module according to an embodiment of the present invention;
FIG. 3 is a block diagram of an image processing module according to an embodiment of the present invention;
FIG. 4 is a block diagram of another image processing module according to an embodiment of the present invention;
fig. 5 is a block diagram of a structure of an identification module according to an embodiment of the present invention;
FIG. 6 is a block diagram of another identification module according to an embodiment of the present invention;
FIG. 7 is a block diagram of a further identification module according to an embodiment of the present invention;
fig. 8 is a block diagram of a sorting module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It will be understood that, as used herein, the terms "first," "second," and the like may be used herein to describe various elements, but these elements are not limited by these terms unless otherwise specified. These terms are only used to distinguish one element from another. For example, a first receiving unit may be referred to as a second receiving unit, and similarly, a second receiving unit may be referred to as a first receiving unit, without departing from the scope of the present application.
As shown in fig. 1, a block diagram of an intelligent visual recognition system according to an embodiment of the present invention is provided, where the system includes:
the image acquisition module 100 is configured to acquire an image of a product and send the acquired image to the image processing module 200.
In this system, image acquisition module 100 is arranged in gathering the image of product, sends the image of gathering to image processing module 200 again, when gathering the image, can select the multi-angle to gather, gathers the image of the different angles of product simultaneously, and the quality to the product of convenient follow-up is accurately detected and is judged, judges whether qualified or the defect of product according to the image of gathering.
The image processing module 200 is configured to receive the image sent by the image acquisition module 100, perform sharpening processing on the image, and send the processed image to the recognition module 300;
in the system, the image processing module 200 processes the received image, and performs the sharpening process on the blurred portion of the image, so as to avoid the occurrence of errors during the subsequent comparison, which causes erroneous judgment and erroneous judgment, improve the accuracy of identification and sorting, and improve the efficiency of comparison and judgment.
The recognition module 300 is used for receiving the image sent by the image processing module 200, comparing and classifying the image with the image in the image storage module 500, and sending an instruction to the sorting module 400;
in the system, the recognition module 300 receives the processed image, the image storage module 500 stores the image of the qualified product and the image of the different defects of the product, and after the recognition module 300 receives the processed clear image, the image is sequentially compared with the image in the image storage module 500 in a traversing manner until the matched image is obtained, the image can be classified, different sorting instructions are generated, and the instructions are sent to the sorting module 400.
And the sorting module 400 is used for receiving the instruction sent by the identification module 300 and sorting the product.
In this system, when sorting module 400 received the letter sorting instruction, can sort the operation to the product, according to the difference of letter sorting instruction, with the product classification to different positions, through the letter sorting instruction of difference, not only can be qualified to distinguish the product, can also subdivide different substandard products, classify the substandard product according to the defect, be favorable to subsequent letter sorting and quality analysis.
As shown in fig. 2, as a preferred embodiment of the present invention, the image capturing module 100 includes:
and the shooting unit 101 is used for acquiring an image of the product.
In this system, shoot unit 101 through shoot the product and gather the image of product, shoot unit 101 specifically can be the camera, can adopt a plurality of cameras when the image of a plurality of angles of product is gathered to needs, the camera is put and is installed in different angular position, also can adopt a camera, this camera can carry out the adjustment of position, change the position in proper order again after shooting an angle and carry out the shooting of different angles, be favorable to obtaining comprehensive, complete product image.
And a light supplement unit 102 for supplementing the light source within the shooting range of the shooting unit 101.
In this system, light filling unit 102 is used for assisting shooting unit 101, an environment with sufficient light source is provided for shooting unit 101, it is clear to have guaranteed that the picture of shooing is clear, it is bright, light filling unit 102 specifically can adopt the LED light source, in addition, the photosensor unit can also be add, and link to each other photosensor unit and light filling unit 102, when the photosensor unit detects the ambient light that the product was located is not enough, light filling unit 102 is opened in control, when the photosensor unit detects light enough, light filling unit 102 is closed in control, the effect of shooing the image has both been guaranteed, and energy saving has been guaranteed again.
A first sending unit 103, configured to send the acquired image to the image processing module 200.
In the present system, the first transmission unit 103 transmits the captured image to the image processing module 200 for subsequent sharpening.
As shown in fig. 3, as a preferred embodiment of the present invention, the image processing module 200 includes:
the first receiving unit 201 is configured to receive an image sent by the image capturing module 100.
In the present system, the image transmitted by the image capturing module 100 is received by the first receiving unit 201.
And the noise reduction unit 202 is used for reducing noise in the image.
In the system, the noise reduction unit 202 is used to reduce noise in the image, and the noise reduction method may adopt any one of an average filter, an adaptive wiener filter, a median filter, a morphological noise filter, and wavelet denoising, or other noise reduction techniques, so as to adapt to the problem of poor quality of the image captured by the low-cost image capturing device.
And the clear restoration unit 203 is used for performing clear processing on the noise-reduced image.
In the system, a clear reduction unit 203 is used for carrying out clear processing on the image subjected to noise reduction, and a cleaning reduction unit inputs a target image into a clear processing model so that the clear processing model can process a target fuzzy picture and output a target clear picture; the method comprises the steps that a sharpening processing model is obtained through a training neural network model, a sample picture needed by the training neural network model is a fuzzy picture synthesized through a plurality of sharpening pictures, a generated sharpening picture corresponding to the fuzzy picture is a sharpening picture before the fuzzy picture is synthesized, namely, a superposed picture can be used as the sample picture, the sharpening picture is a training target of the sharpening processing model through inputting the fuzzy picture and outputting the sharpening picture, and the sharpening picture corresponding to the superposed picture can be output after the trained sharpening processing model is obtained and inputting the superposed picture into the sharpening processing model; in addition, the sharpening unit 203 may also adopt other sharpening processing techniques.
A second sending unit 204, configured to send the processed image to the recognition module 300.
In the system, the clear image is transmitted to the recognition module 300 by the second transmitting unit 204, so that the clear image can be compared and classified.
As shown in fig. 4, as a preferred embodiment of the present invention, the image processing module 200 further includes a detecting unit 205, where the detecting unit 205 is configured to detect whether an image is clear; when the image is clear, the detection unit 205 is configured to send the image to the second sending unit 204; when the image is not clear, the detection unit 205 is configured to send the image to the clear restoration unit 203.
In the system, the image processing module 200 may further include a detection unit 205, where the detection unit 205 is connected to the noise reduction unit 202 and is configured to detect whether an image subjected to noise reduction is clear, and when the image is clear, the image may be filtered by the clear restoration unit 203 and directly sent to the identification module 300 through the second sending unit 204, and when the image is blurred, the image is sent to the clear identification unit;
in the system, when the detection unit 205 receives an image, the detection unit performs area division on the image, divides a central area and an edge area, and acquires a central outline and an edge outline corresponding to the two areas, wherein the central outline and the edge outline can reflect the definition degree of the image, and can detect whether the image is fuzzy or clear according to the acquired central outline and the edge outline, the outline is an effect graph after the image is subjected to outline presentation according to a certain image depth threshold, areas occupied by the central area and the edge area on the original image respectively can be set by a user, whether the original image is clear is judged according to whether the central outline is presented in the central area, when the central outline is not presented, the original image is considered to be fuzzy, when the central outline is presented, the edge outline and the central outline presented by the edge area and the central area of the original image are acquired respectively, then the two are compared, and whether the original image is a clear or fuzzy result is judged according to a comparison result; specifically, the value to be compared here refers to the extracted value of the color level lightness and darkness after the image is subjected to the threshold processing, and the extracted value of the color level lightness and darkness can be obtained simultaneously when the image is subjected to the threshold processing. For example, when the central contour is not empty, that is, the central contour can be presented in the central area, the corresponding original picture is generally considered to be clear, and if the central contour is empty, that is, no central contour can be presented in the central area, it is necessary to perform detection as to whether the original picture is clear or not according to the edge contour.
As shown in fig. 5, as a preferred embodiment of the present invention, the identification module 300 includes:
a second receiving unit 301, configured to receive the image sent by the image processing module 200.
In the present system, the image after the sharpening process is received by the second receiving unit 301.
And an analyzing unit 302 for sequentially analyzing and comparing the images with defective images in the image storage module 500.
In the system, the product images are sequentially compared with defective images in the image storage module 500 by using the analysis unit 302, when images with a plurality of angles are collected, the images with the plurality of angles are spliced into one image, defective images with qualified products and different defect types are stored in the image storage module 500, and the product images are sequentially compared with the images in the image storage module 500 until the corresponding images are found, so that the product images can be classified;
when the product image contains different defect problems, the product image can be stored in the image storage module 500 and added with new classifications.
A classifying unit 303, configured to classify the images according to the analysis and comparison result of the analyzing unit 302 to generate different sorting instructions.
In the system, the classifying unit 303 generates sorting instructions according to the classification results obtained by the comparison of the analyzing unit 302, and different product images can be generated into different sorting instructions according to the qualification or defect problems, so as to drive the products to be sorted to different directions.
A third sending unit 304 for sending sorting instructions to the sorting module 400.
In the present system, the sorting instructions are transmitted to the sorting module 400 using the third transmitting unit 304, so that the sorting module 400 can operate on the product.
As shown in fig. 6, as a preferred embodiment of the present invention, the identification module 300 further includes a statistic unit 305, and the statistic unit 305 is configured to count the number of different inferior-quality products according to the analysis comparison result of the analysis unit 302 to generate data information.
In the system, in order to facilitate the staff to know the condition of the product, the identification module 300 may further include a statistical unit 305, the statistical unit 305 is used to mark the unqualified position of the product, and then statistics is performed according to the number of times of the mark, the statistics is performed once when the mark appears, the number of the statistical unit 305 is adapted to the number of categories of the qualified product and the defective product in the classification unit 303, each statistical unit 305 corresponds to one category, and statistics is performed on different categories respectively;
the statistical unit 305 may further include a calculation unit for calculating a pass rate of the product based on the received pass and fail numbers.
As shown in fig. 7, as a preferred embodiment of the present invention, the recognition module 300 further includes an output unit 306, and the output unit 306 is configured to output data showing the number of defective products based on the data information of the statistic unit 305.
In the system, the output unit 306 corresponds to the counting unit 305, the counting unit 305 counts the number of the received images of different categories and transmits the counted images to the output unit 306, the number of the output unit 306 is suitable for the number of the counting unit 305, and the output unit 306 displays the number of each category;
when the statistical unit 305 further includes a calculation unit, the yield is transmitted to the output unit 306, and the output unit 306 is used for displaying the received data, thereby facilitating the staff to visually obtain the product quality.
As shown in fig. 8, as a preferred embodiment of the present invention, the sorting module 400 includes:
a third receiving unit 401, configured to receive the instruction sent by the identification module 300.
In the present system, the sorting instruction transmitted from the identification module 300 is received by the third receiving unit 401.
An execution unit 402 for operating the machine according to the instructions of the third receiving unit 401 to sort the products.
In the system, the execution unit 402 is used to execute information in sorting instructions, where different sorting instructions correspond to different execution operation processes, and thus the machine can be operated to sort products, where the machine should be suitable for the structure of the products and the sorting requirements of the products.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that various changes and modifications can be made by those skilled in the art without departing from the spirit of the invention, and these changes and modifications are all within the scope of the invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
The above description is intended to be illustrative of the preferred embodiment of the present invention and should not be taken as limiting the invention, but rather, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
Claims (8)
1. An intelligent visual recognition system, the system comprising:
the image acquisition module is used for acquiring an image of a product and sending the acquired image to the image processing module;
the image processing module is used for receiving the image sent by the image acquisition module, carrying out sharpening processing on the image and sending the processed image to the identification module;
the recognition module is used for receiving the image sent by the image processing module, comparing and classifying the image with the image in the image storage module, and sending an instruction to the sorting module;
and the sorting module is used for receiving the instruction sent by the identification module and sorting the products.
2. The system of claim 1, wherein the image acquisition module comprises:
the shooting unit is used for acquiring an image of a product;
the light supplementing unit is used for supplementing a light source in the shooting range of the shooting unit;
and the first sending unit is used for sending the acquired image to the image processing module.
3. The system of claim 1, wherein the image processing module comprises:
the first receiving unit is used for receiving the image sent by the image acquisition module;
the noise reduction unit is used for reducing noise in the image;
the clear restoration unit is used for carrying out clear processing on the image subjected to noise reduction;
and the second sending unit is used for sending the processed image to the identification module.
4. The system according to claim 3, wherein the image processing module further comprises a detection unit for detecting whether the image is sharp; when the image is clear, the detection unit is used for sending the image to the second sending unit; and when the image is not clear, the detection unit is used for sending the image to the clear restoration unit.
5. The system of claim 1, wherein the identification module comprises:
the second receiving unit is used for receiving the image sent by the image processing module;
the analysis unit is used for sequentially analyzing and comparing the images with the defective product images in the image storage module;
the classification unit is used for classifying the images according to the analysis and comparison results of the analysis unit so as to generate different sorting instructions;
and the third sending unit is used for sending the sorting instruction to the sorting module.
6. The system of claim 5, wherein the identification module further comprises a statistic unit for counting the number of different inferior goods according to the analysis comparison result of the analysis unit to generate data information.
7. The system of claim 6, wherein the recognition module further comprises an output unit for outputting data showing the number of inferior goods based on the data information of the statistical unit.
8. The system of claim 1, wherein the sorting module comprises:
the third receiving unit is used for receiving the instruction sent by the identification module;
and the execution unit is used for operating the machine according to the instruction of the third receiving unit so as to sort the products.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211316732.2A CN115672778A (en) | 2022-10-26 | 2022-10-26 | Intelligent visual recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211316732.2A CN115672778A (en) | 2022-10-26 | 2022-10-26 | Intelligent visual recognition system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115672778A true CN115672778A (en) | 2023-02-03 |
Family
ID=85098571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211316732.2A Withdrawn CN115672778A (en) | 2022-10-26 | 2022-10-26 | Intelligent visual recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115672778A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117541531A (en) * | 2023-09-28 | 2024-02-09 | 苏州梅曼智能科技有限公司 | Wafer vision detection supervision feedback system based on artificial intelligence |
-
2022
- 2022-10-26 CN CN202211316732.2A patent/CN115672778A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117541531A (en) * | 2023-09-28 | 2024-02-09 | 苏州梅曼智能科技有限公司 | Wafer vision detection supervision feedback system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109454006B (en) | Detection and classification method based on device for online detection and classification of chemical fiber spindle tripping defects | |
| CN109550712B (en) | Chemical fiber filament tail fiber appearance defect detection system and method | |
| CN106841209B (en) | One kind being based on big data self study chemical fiber wire ingot intelligence appearance detection system and method | |
| CN104169945B (en) | To the two-stage classification of the object in image | |
| CN109255787A (en) | Silk ingot scratch detection system and method based on deep learning and image processing techniques | |
| CN105701476A (en) | Machine vision-based automatic identification system and method for production line products | |
| CN111126325A (en) | Intelligent personnel security identification statistical method based on video | |
| CN110533654A (en) | The method for detecting abnormality and device of components | |
| US20180268634A1 (en) | Folded bill identification method and device | |
| CN113642474A (en) | Hazardous area personnel monitoring method based on YOLOV5 | |
| CN108776143A (en) | A kind of online vision inspection apparatus and method of the small stain of egg eggshell surface | |
| CN113111903A (en) | Intelligent production line monitoring system and monitoring method | |
| CN114820626B (en) | Intelligent detection method for automobile front face part configuration | |
| CN114445365A (en) | Banknote printing quality inspection method based on deep learning algorithm | |
| Chao et al. | Design of a dual-camera system for poultry carcasses inspection | |
| CN115672778A (en) | Intelligent visual recognition system | |
| CN114549493A (en) | Magnetic core defect detection system and method based on deep learning | |
| CN116573307A (en) | Automatic garbage classification and recycling intelligent equipment and intelligent garbage classification system thereof | |
| CN113834814B (en) | Glove surface defect detection device | |
| CN114332058A (en) | Serum quality identification method, device, equipment and medium based on neural network | |
| EP3896650A1 (en) | Quality control system for series production | |
| Ignacio et al. | A YOLOv5-based deep learning model for in-situ detection and maturity grading of mango | |
| CN107545565B (en) | Solar screen plate detection method | |
| CN113228049B (en) | Milk analyzer for classifying milk | |
| Aleixos et al. | Assessment of citrus fruit quality using a real-time machine vision system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20230203 |
|
| WW01 | Invention patent application withdrawn after publication |