CN116580233A - Intelligent working robot control system for industrial garbage classification - Google Patents
Intelligent working robot control system for industrial garbage classification Download PDFInfo
- Publication number
- CN116580233A CN116580233A CN202310542935.1A CN202310542935A CN116580233A CN 116580233 A CN116580233 A CN 116580233A CN 202310542935 A CN202310542935 A CN 202310542935A CN 116580233 A CN116580233 A CN 116580233A
- Authority
- CN
- China
- Prior art keywords
- image
- industrial
- waste
- garbage
- classified
- 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
- 239000010813 municipal solid waste Substances 0.000 title claims abstract description 55
- 239000002440 industrial waste Substances 0.000 claims abstract description 34
- 238000003384 imaging method Methods 0.000 claims abstract description 15
- 238000003860 storage Methods 0.000 claims abstract description 8
- 239000002699 waste material Substances 0.000 claims description 30
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 8
- 238000000034 method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000000694 effects Effects 0.000 claims description 6
- 239000010794 food waste Substances 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 239000011449 brick Substances 0.000 claims description 3
- 239000002920 hazardous waste Substances 0.000 claims description 3
- 238000002360 preparation method Methods 0.000 claims description 3
- 238000004451 qualitative analysis Methods 0.000 claims description 3
- 239000002689 soil Substances 0.000 claims description 3
- 239000002023 wood Substances 0.000 claims description 3
- 125000004122 cyclic group Chemical group 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 4
- 238000004064 recycling Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000005294 ferromagnetic effect Effects 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000010819 recyclable waste Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
Abstract
The invention provides an intelligent working robot control system for industrial garbage classification, and belongs to the technical field of garbage classification. The system comprises: the device comprises an image acquisition module, an image storage module, an imaging matching module and a classification module; the image acquisition module is used for acquiring images of the industrial garbage to be classified; the image storage module is used for storing the images of the industrial waste to be classified into a database and defining different labels according to different categories; the imaging matching module is used for matching the instantly generated image of the industrial garbage to be classified with the image stored in the head of the database; and the classification module is used for matching the optimal data set and automatically classifying the labels of the corresponding data set. The invention provides an intelligent working robot control system for industrial garbage classification, which can greatly improve the accuracy of garbage identification respectively and greatly save human resources after the system is operated.
Description
Technical Field
The invention belongs to the technical field of garbage classification, and particularly relates to an intelligent working robot control system for industrial garbage classification.
Background
The industrial garbage is waste garbage generated in the production or construction process of factories or construction sites, is different from common household garbage, contains a lot of renewable resources and harmful substances, and can be partially sent to a resource recycling station for resource recycling, and the rest of harmful substances or other undefined industrial garbage can be sorted out for recycling, so that the waste of the waste industrial garbage can be reduced.
At present, most of the discarded industrial wastes are classified by adopting magnets, and useful ferromagnetic wastes are sorted out, but under the condition that some of the wastes are not strong in magnetism or are wrapped by plastic materials, complicated public sequences are needed, and meanwhile, many other wastes are needed to be classified, so that the problems are needed to be solved again.
Disclosure of Invention
The invention aims to provide an intelligent working robot control system for classifying industrial garbage, which is used for recognizing and classifying garbage by using a yolov2 model, cleaning according to the setting of the system, reducing classification steps and procedures and realizing garbage classification more intelligently.
In order to achieve the above object, in the present invention, an intelligent working robot control system for industrial garbage classification, the system comprises: the device comprises an image acquisition module, an image storage module, an imaging matching module and a classification module;
the image acquisition module is used for acquiring images of the industrial garbage to be classified;
the image storage module is used for storing the images of the industrial garbage to be classified into a database, forming a basic set and defining different labels according to different categories;
the imaging matching module is used for matching the instantly generated image of the industrial garbage to be classified with the image stored in the head of the database;
the classification module is used for matching the optimal data set and automatically classifying the labels of the corresponding data set.
Further, each tag corresponds to a number of images of the same category, and the dataset is placed in a recurring set for repeated images.
Further, the defining different labels according to different categories comprises the following specific steps:
matching the imaging reading of the images of the industrial wastes to be classified with the defined labels according to each acquisition;
establishing a temporary data set for placing the image just generated;
matching the temporary data set with the basic set;
and storing the sorted data set in a database. An intelligent work robot control system for industrial waste classification according to claim 1, wherein the tags specifically include food waste, construction waste and, normal waste and hazardous waste;
the food waste comprises waste generated in the process of processing and eating food;
the common garbage comprises recyclable garbage;
the construction waste comprises soil, stones, concrete blocks, broken bricks, waste wood, waste pipelines and electric appliance waste;
the dangerous garbage comprises a dry battery, a fluorescent tube and a thermometer.
Further, the step for matching the image of the industrial waste to be classified, which is generated in real time, with the image stored in the database header comprises:
converting the collected image of the industrial garbage to be classified into a gray image;
calculating a threshold value of a gray level image for extracting difference characteristics of the garbage area and the background;
selecting an image of a background area with an optimal threshold value, and selecting an iteration method to obtain an image of industrial waste with an optimal threshold value segmentation algorithm;
and transmitting the industrial garbage image of the optimal threshold segmentation algorithm into a VGG-16 convolutional neural network for training, and matching and classifying.
Further, the VGG-16 convolutional neural network is specifically trained by: and correcting each pixel in the image by calculating the brightness ratio and shadow relation between each pixel, determining the color of the pixel, performing qualitative analysis and comparison on the processing effect by image information entropy, and making the convolutional neural network more perfect by repeated imaging and multiple training of the convolutional neural network.
Further, the coordinates of the pixel point are (x, y), the initial image of the industrial garbage to be classified is S (x, y), the reflected image of the industrial garbage to be classified is R (x, y), the brightness image of the industrial garbage to be classified is I (x, y), and the relationship satisfies S (x, y) =i (x, y) ×r (x, y);
r(x，y)＝logR(x，y)；
wherein r (x, y) is the output image;
wherein ,；
where K represents the total number of gaussian center-surrounded functions, K is the number that needs to be calculated for the function representing gaussian center-surrounded functions, x represents the center-surrounded function using convolution operations, and where C is a gaussian-surround scale, is a scale that satisfies the equation, and is satisfied.
Further, the processing effect is qualitatively analyzed and compared through the information entropy of the output image, the statistical format of the industrial garbage image information is displayed, the statistical format is used for displaying the preparation aggregation attribute of the image and the two-dimensional entropy of the created image, the probability of the gray level image appearing in the image exists, the appearance frequency is represented, the characteristic double-tuple characteristic is represented, M is the size of the image, and the information entropy of the output image is represented, wherein n represents a constant.
Further, the industrial garbage is classified and matched by comparing the entropy of the output picture information.
The beneficial technical effects of the invention are at least as follows:
1. the invention provides an intelligent working robot control system for classifying industrial garbage, which can greatly improve the accuracy of recognizing garbage respectively after the system is operated;
2. for the situations that the images cannot be accurately identified, such as unclear darkness, shadow coverage and foreign object overlapping, especially at night, the convolutional neural network used by the system can cope with the influence of the environment on the images.
3. The cost of labor is reduced.
Drawings
The invention will be further described with reference to the accompanying drawings, in which embodiments do not constitute any limitation of the invention, and other drawings can be obtained by one of ordinary skill in the art without inventive effort from the following drawings.
FIG. 1 is a block diagram of an intelligent work robot control system for industrial waste classification according to the present invention.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
Embodiment one:
as shown in fig. 1, the present invention provides an intelligent work robot control system for industrial waste classification, the system comprising: the device comprises an image acquisition module, an image storage module, an imaging matching module and a classification module;
the image acquisition module is used for acquiring images of the industrial garbage to be classified;
the image storage module is used for storing the images of the industrial waste to be classified into a database and defining different labels according to different categories;
under each label there is stored a large number of data sets of the same category that would be placed in a recurring set for repeated images.
Different labels are defined according to different categories, and the specific steps are as follows:
matching is carried out according to the imaging reading of the images of the industrial garbage to be classified obtained each time;
establishing a temporary data set for placing the image just generated;
matching the temporary data set with the basic set;
and storing the sorted data set in a database.
Marking the characteristic information of the industrial waste to be classified to form corresponding labels, wherein the labels specifically comprise food waste, construction waste and common waste and dangerous waste;
the food waste comprises waste generated in the process of processing and eating food;
the common garbage comprises recyclable garbage;
the construction waste comprises soil, stones, concrete blocks, broken bricks, waste wood, waste pipelines and electric appliance waste;
the dangerous garbage comprises a dry battery, a fluorescent tube and a thermometer.
And the imaging matching module is used for matching the instantly generated image of the industrial garbage to be classified with the image stored in the head part of the database.
The step of imaging the matching module comprises:
converting the collected image of the industrial garbage to be classified into a gray image;
calculating a threshold value of a gray level image for extracting difference characteristics of the garbage area and the background;
selecting an image of a background area with an optimal threshold value, and selecting an iteration method to obtain an image of industrial waste with an optimal threshold value segmentation algorithm;
and transmitting the images of the industrial waste into a VGG-16 convolutional neural network for training, and matching and classifying.
The VGG-16 convolutional neural network is specifically trained by: and correcting each pixel in the image by calculating the brightness ratio and shadow relation between each pixel, determining the color of the pixel, performing qualitative analysis and comparison on the processing effect by image information entropy, and making the convolutional neural network more perfect by repeated imaging and multiple training of the convolutional neural network.
The pixels are (x, y), the initial image of the industrial waste to be classified is S (x, y), the reflected image of the industrial waste to be classified is R (x, y), the brightness image of the industrial waste to be classified is I (x, y, and the relation satisfies
S(x，y＝I(x，y*R(x，y；
r(x，y＝logR(x，y；
Where r (x, y is the output image and I (x, y is the input image of the initial image of the industrial waste to be classified;
wherein ,
wherein K represents the number of functions surrounded by a Gaussian center, K is a constant, and F is calculated by convolution k-1 (x, y) represents a center-surround function, andwhere C is a gaussian surrounding scale, α is a scale satisfying the equation, and satisfies ≡ ≡f (x, y) dxdy=1.
The processing effect is qualitatively analyzed and compared by the information entropy obtained by the output image information, the statistical format of the industrial garbage image information is displayed, the preparation aggregation attribute of the image is displayed, the two-bit entropy of the image is created, and the probability P of gray level image appearing in the image exists ij ＝f(i,j)/M 2 Wherein f (i, j) represents the frequency of occurrence, (i, j) represents the feature binary group feature, M is the size of the image, and the entropy of the output picture information isWhere n represents a constant.
And the classification module is used for matching the optimal data set and automatically classifying the labels of the corresponding data set. And classifying the images according to the labels by comparing the information entropy of the output pictures, and classifying and matching the industrial wastes. Firstly, loading the trained optimal weight parameters, then taking a test set as a model input, and predicting to obtain a corresponding classification result
Specific embodiment II:
since the classification standards for each site are different, the specified waste is classified into recyclable waste and hazardous waste, the system using the present invention can set corresponding IP addresses according to different areas and store in the area waste classification standards, respectively. Meanwhile, the original pictures of some articles are uploaded to a cloud platform to serve as the basis for article identification, and then become garbage. And then, when the images are generated through the identification technology and garbage classification, the corresponding basic images are subjected to high-speed matching by using the system, and the image with the highest matching degree is found for final segmentation. Each identified image is automatically stored on the cloud platform, and the information of each stage is gradually perfected, so that the technology is ensured to be more mature after being applied, and the technology can be applied to more places needing the help of the technology.
While embodiments of the invention have been shown and described, it will be understood by those skilled in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.
Claims (8)
1. An intelligent work robot control system for industrial waste classification, the system comprising: the device comprises an image acquisition module, an image storage module, an imaging matching module and a classification module;
the image acquisition module is used for acquiring images of the industrial garbage to be classified;
the image storage module is used for storing the images of the industrial garbage to be classified into a database, forming a basic set and defining different labels according to different categories;
the imaging matching module is used for matching the instantly generated image of the industrial garbage to be classified with the image stored in the head of the database;
the classification module is used for matching the optimal data set and automatically classifying the labels of the corresponding data set.
2. An intelligent work robot control system for industrial waste classification as claimed in claim 1, wherein each tag corresponds to a plurality of images of the same class, and wherein the data sets are placed in a cyclic set for repeated images.
3. An intelligent work robot control system for industrial waste classification according to claim 2, wherein the defining of different labels according to different classes comprises the following specific steps:
matching the imaging reading of the images of the industrial wastes to be classified with the defined labels according to each acquisition;
establishing a temporary data set for placing the image just generated;
matching the temporary data set with the basic set;
and storing the sorted data set in a database. An intelligent work robot control system for industrial waste classification according to claim 1, wherein the tags specifically include food waste, construction waste and, normal waste and hazardous waste;
the food waste comprises waste generated in the process of processing and eating food;
the common garbage comprises recyclable garbage;
the construction waste comprises soil, stones, concrete blocks, broken bricks, waste wood, waste pipelines and electric appliance waste;
the dangerous garbage comprises a dry battery, a fluorescent tube and a thermometer.
4. An intelligent work robot control system for industrial waste classification according to claim 1, wherein said step for matching an image of the industrial waste to be classified generated on the fly with an image stored in a database header comprises:
converting the collected image of the industrial garbage to be classified into a gray image;
calculating a threshold value of a gray level image for extracting difference characteristics of the garbage area and the background;
selecting an image of a background area with an optimal threshold value, and selecting an iteration method to obtain an image of industrial waste with an optimal threshold value segmentation algorithm;
and transmitting the industrial garbage image of the optimal threshold segmentation algorithm into a VGG-16 convolutional neural network for training, and matching and classifying.
5. The intelligent work robot control system for industrial waste classification of claim 4, wherein the VGG-16 convolutional neural network is trained specifically as follows: and correcting each pixel in the image by calculating the brightness ratio and shadow relation between each pixel, determining the color of the pixel, performing qualitative analysis and comparison on the processing effect by image information entropy, and making the convolutional neural network more perfect by repeated imaging and multiple training of the convolutional neural network.
6. The intelligent working robot control system for industrial waste classification according to claim 5, wherein the coordinates of the pixel point are (x, y), the initial image of the industrial waste to be classified is S (x, y), the reflected image of the industrial waste to be classified is R (x, y), the brightness image of the industrial waste to be classified is I (x, y), and the relation satisfies S (x, y) =i (x, y) ×r (x, y);
r(x，y)＝logR(x，y)；
wherein r (x, y) is the output image;
wherein ,
wherein K represents the total number of the functions surrounded by the gaussian center, K represents the number of the functions surrounded by the gaussian center to be calculated, F represents the use of convolution operation k-1 (x, y) represents a center-surround function, andwherein C is a Gaussian surround scale and alpha is a scale satisfying the equationDegree, and satisfies ≡ ≡f (x, y) dxdy=1.
7. The intelligent working robot control system for classifying industrial waste according to claim 6, wherein the statistical format of the industrial waste image information is displayed by qualitatively analyzing and comparing the information entropy of the output image, and the probability P of gray-scale image appearing in the image is given by the preparation aggregation attribute of the display image and the two-dimensional entropy of the created image ij ＝f(i,j)/M 2 Where f (i, j) represents the frequency of occurrence, (i, j) represents the feature-tuple feature, M is the size of the image, and the entropy of the output image information isWhere n represents a constant.
8. An intelligent work robot control system for industrial waste classification as claimed in claim 7, wherein the industrial waste is classified and matched by comparing the entropy of the output picture information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310542935.1A CN116580233A (en) | 2023-05-11 | 2023-05-11 | Intelligent working robot control system for industrial garbage classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310542935.1A CN116580233A (en) | 2023-05-11 | 2023-05-11 | Intelligent working robot control system for industrial garbage classification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116580233A true CN116580233A (en) | 2023-08-11 |
Family
ID=87537161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310542935.1A Pending CN116580233A (en) | 2023-05-11 | 2023-05-11 | Intelligent working robot control system for industrial garbage classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116580233A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116884536A (en) * | 2023-08-29 | 2023-10-13 | 济南明泉数字商务有限公司 | Automatic optimization method and system for production formula of industrial waste residue bricks |
-
2023
- 2023-05-11 CN CN202310542935.1A patent/CN116580233A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116884536A (en) * | 2023-08-29 | 2023-10-13 | 济南明泉数字商务有限公司 | Automatic optimization method and system for production formula of industrial waste residue bricks |
| CN116884536B (en) * | 2023-08-29 | 2024-03-12 | 济南明泉数字商务有限公司 | Automatic optimization method and system for production formula of industrial waste residue bricks |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106845408B (en) | Street garbage identification method under complex environment | |
| US20210224609A1 (en) | Method, system and device for multi-label object detection based on an object detection network | |
| CN107480643B (en) | Intelligent garbage classification processing robot | |
| Gyawali et al. | Comparative analysis of multiple deep CNN models for waste classification | |
| CN109308479B (en) | Deep learning-based campus garbage automatic classification and recovery method | |
| CN107368614A (en) | Image search method and device based on deep learning | |
| CN104881770A (en) | Express bill information identification system and express bill information identification method | |
| CN113128335B (en) | Method, system and application for detecting, classifying and finding micro-living ancient fossil image | |
| CN111784171A (en) | Municipal CIM environmental sanitation work distribution method based on artificial intelligence and image processing | |
| CN110097616B (en) | Combined drawing method and device, terminal equipment and readable storage medium | |
| CN113436184B (en) | Power equipment image defect discriminating method and system based on improved twin network | |
| CN111161213B (en) | Industrial product defect image classification method based on knowledge graph | |
| CN116580233A (en) | Intelligent working robot control system for industrial garbage classification | |
| CN112580647A (en) | Stacked object oriented identification method and system | |
| CN114169381A (en) | Image annotation method and device, terminal equipment and storage medium | |
| CN112750113B (en) | Glass bottle defect detection method and device based on deep learning and linear detection | |
| CN113469264A (en) | Construction method of automatic garbage classification model, garbage sorting method and system | |
| CN110598564A (en) | OpenStreetMap-based high-spatial-resolution remote sensing image transfer learning classification method | |
| CN110533629A (en) | A kind of detection method and detection device of Bridge Crack | |
| CN111652214A (en) | Garbage bottle sorting method based on deep learning | |
| CN107563327B (en) | Pedestrian re-identification method and system based on self-walking feedback | |
| CN113160136A (en) | Wood defect identification and segmentation method based on improved Mask R-CNN | |
| CN106709474A (en) | Handwritten telephone number identification, verification and information sending system | |
| CN115375672A (en) | Coal gangue detection method based on improved YOLOv4 algorithm | |
| CN113971746B (en) | Garbage classification method and device based on single hand teaching and intelligent sorting 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 |