CN115346118A - Identification and positioning method for vehicles in scrap steel discharge yard - Google Patents
Identification and positioning method for vehicles in scrap steel discharge yard Download PDFInfo
- Publication number
- CN115346118A CN115346118A CN202210962433.XA CN202210962433A CN115346118A CN 115346118 A CN115346118 A CN 115346118A CN 202210962433 A CN202210962433 A CN 202210962433A CN 115346118 A CN115346118 A CN 115346118A
- Authority
- CN
- China
- Prior art keywords
- network
- neural network
- yard
- positioning
- vehicles
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- 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
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (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 Analysis (AREA)
Abstract
The invention discloses a method for identifying and positioning vehicles in a scrap steel discharge yard, which comprises the following steps: arranging a camera on the scrap steel stock yard, wherein the camera moves and collects images of the stock yard; labeling the trucks in the stock ground image to obtain a data set; inputting the data set into a detection neural network for training; expanding the input image into a plurality of feature maps with different scales through the network structure; the neural network utilizes a plurality of feature maps with different scales, and calculates the target category, the specific position and the specific size according to the content of each area in the feature maps with different scales through an ROI Align layer, a classification network and a regression network; the neural network also classifies each target area of the image at a pixel level by predicting the network structure, and particularly marks whether each pixel in the target area is a background or a target, so as to determine the vehicle positioning state, thereby realizing the complex target detection and positioning in a complex scene, being completely suitable for the positioning requirements of various complex scenes and complex vehicles in function, and being capable of providing a high-efficiency and stable vehicle positioning guarantee for the intelligent quality testing system of scrap steel.
Description
Technical Field
The invention relates to the technical field of vehicle identification and positioning, in particular to a method for identifying and positioning vehicles in a scrap steel discharge yard.
Background
The scrap steel is a green resource, has high environmental protection value for the steel production industry, and the scrap iron is the only resource for replacing iron ore for steelmaking. Many steel scrap types, the actual detection sight is complicated, the manual system links up the degree of difficulty greatly, most steel enterprises judge that the steel scrap grade is mainly surveyd by quality control personnel and the callipers measures and judges jointly now, traditional steel scrap is graded and mainly relies on naked eye identification, be difficult to standardize, for realizing that steel scrap grade discernment overall process is unmanned, at first need fix a position the steel scrap vehicle through the camera is automatic, because all be the steel scrap around the steel scrap discharge point, the steel scrap vehicle that loads in the steel scrap vehicle also, according to traditional approach, it is very difficult accurate location steel scrap vehicle.
Disclosure of Invention
The invention aims to provide a method for identifying and positioning vehicles in a scrap steel discharge yard, which aims to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme: a method for identifying and positioning vehicles in a scrap steel discharge yard comprises the following steps:
the method comprises the following steps: arranging a camera on the scrap steel stock yard, wherein the camera moves and collects images of the stock yard;
step two: labeling the trucks in the stock ground image to obtain a data set;
step three: inputting the data set into a detection neural network for training to obtain a training model;
step four: the input image is expanded into a plurality of feature maps with different scales through the network structure, so that the multi-scale detection of vehicles with different sizes in different shooting environments of various factories and discharge openings by the system is realized;
step five: the neural network utilizes a plurality of feature maps with different scales, and calculates the target category, specific position and specific size according to the content of each region in the feature maps with different scales through an ROI Align layer, a classification network and a regression network;
step six: the neural network also carries out pixel-level classification on each target area of the image through a prediction network structure, and particularly marks whether each pixel in the target area is a background or a target, so that the vehicle positioning state is determined.
Further, the neural Network adopts Mask RCNN Network, and the input part adopts Res Net (Residual Net) + FPN (Feature Pyramid Network) structure.
Further, the regression network comprises a plurality of convolution layers, a pooling layer and a full connection layer.
And further, labeling and classifying the material loading/unloading state of the truck in the unloading field image, inputting the labeled and classified stock field image into a classification neural network to obtain a classification model, classifying the stock field image through the classification model, and determining the loading state of the truck.
A computer readable medium having stored thereon instructions, which when executed by one or more processors, cause an apparatus to perform the method for locating a vehicle in a scrap discharge yard.
Compared with the prior art, the invention has the beneficial effects that: the steel scrap vehicle positioning system based on the neural network can automatically train parameters through a large number of vehicle example images loaded with steel scrap, so that the parameter quantity of the system is increased by thousands of times, the complex target detection and positioning under a complex scene are realized, the system can be functionally and completely adapted to the positioning requirements of various complex scenes and complex vehicles, and the system can provide efficient and stable vehicle positioning guarantee for an intelligent steel scrap quality testing system.
Drawings
FIG. 1 is a flow chart of a method for identifying and positioning vehicles in a scrap discharge yard according to the present invention.
FIG. 2 is a schematic diagram of a detection result of the method for identifying and positioning the vehicles in the scrap steel discharge yard.
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.
Example 1
In the embodiment of the invention, the method for identifying and positioning the vehicles in the scrap steel discharge yard comprises the following steps:
step one S1: arranging a camera on the scrap steel stock yard, wherein the camera moves and collects images of the stock yard;
step two S2: labeling the trucks in the stock ground image to obtain a data set;
step three, S3: inputting the data set into a detection neural network for training to obtain a training model;
step four S4: the input image is expanded into a plurality of feature maps with different scales through the network structure, so that the system can realize multi-scale detection of vehicles with different sizes in different shooting environments of various factories and discharge openings.
And step five S5, the neural network utilizes a plurality of feature maps with different scales, and calculates the target category, the specific position and the specific size according to the content of each area in the feature maps with different scales through the ROI Align layer, the classification network and the regression network.
Step six S6: the neural network also carries out pixel-level classification on each target area of the image through a prediction network structure, and particularly marks whether each pixel in the target area is a background or a target, so as to determine the vehicle positioning state.
The regression network comprises a plurality of convolution layers, a pooling layer and a full-connection layer.
And step two S2, labeling and classifying the material loading/unloading state of the truck in the unloading field image, inputting the labeled and classified stock field image into a classification neural network to obtain a classification model, classifying the stock field image through the classification model, and determining the loading state of the truck.
Example 2
Under the same method as that in embodiment 1, the neural Network in this embodiment adopts Mask RCNN Network, and the input portion adopts Res Net (Residual Net) + FPN (Feature Pyramid Network) structure.
Example 3
On the premise of the same methods as embodiments 1 and 2, the present embodiment is a computer-readable medium having stored thereon instructions that, when executed by one or more processors, cause an apparatus to perform the methods described in embodiments 1 and 2.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (5)
1. A method for identifying and positioning vehicles in a scrap steel discharge yard is characterized by comprising the following steps: the method comprises the following steps:
the method comprises the following steps: arranging a camera on the scrap steel stock yard, wherein the camera moves and collects images of the stock yard;
step two: labeling the trucks in the stock ground image to obtain a data set;
step three: inputting the data set into a detection neural network for training to obtain a training model;
step four: the input image is expanded into a plurality of feature maps with different scales through the network structure, so that the system can realize multi-scale detection of vehicles with different sizes in different shooting environments of various factories and discharge openings.
Step five: the neural network utilizes a plurality of feature maps with different scales, and calculates the target category, the specific position and the specific size according to the content of each region in the feature maps with different scales through the ROI Align layer, the classification network and the regression network.
Step six: the neural network also carries out pixel-level classification on each target area of the image through a prediction network structure, and particularly marks whether each pixel in the target area is a background or a target, so as to determine the vehicle positioning state.
2. The method as claimed in claim 1, wherein the neural Network is Mask RCNN Network, and the input part is Res Net (Residual Net) + FPN (Feature Pyramid Network) structure.
3. The method as claimed in claim 1, wherein the regression network comprises a plurality of convolution layers, pooling layers, and full connection layers.
4. The method for identifying and positioning vehicles in the scrap steel discharge yard according to claim 1, wherein in the second step, the material loading/unloading state of the trucks in the images of the discharge yard is labeled and classified, the labeled and classified images of the discharge yard are input into a classification neural network to obtain a classification model, and the images of the discharge yard are classified through the classification model to determine the loading state of the trucks.
5. A computer-readable medium having instructions stored thereon, which when executed by one or more processors, cause an apparatus to perform the method recited in claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210962433.XA CN115346118A (en) | 2022-08-11 | 2022-08-11 | Identification and positioning method for vehicles in scrap steel discharge yard |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210962433.XA CN115346118A (en) | 2022-08-11 | 2022-08-11 | Identification and positioning method for vehicles in scrap steel discharge yard |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115346118A true CN115346118A (en) | 2022-11-15 |
Family
ID=83951544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210962433.XA Pending CN115346118A (en) | 2022-08-11 | 2022-08-11 | Identification and positioning method for vehicles in scrap steel discharge yard |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115346118A (en) |
-
2022
- 2022-08-11 CN CN202210962433.XA patent/CN115346118A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103824081B (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
| CN106919910B (en) | Traffic sign identification method based on HOG-CTH combined features | |
| CN111080607B (en) | Rolling bearing oil slinging fault detection method based on image recognition | |
| CN109993138A (en) | A kind of car plate detection and recognition methods and device | |
| CN106934392A (en) | Vehicle-logo recognition and attribute forecast method based on multi-task learning convolutional neural networks | |
| CN110991447B (en) | Train number accurate positioning and identifying method based on deep learning | |
| CN109657664A (en) | A kind of recognition methods, device and the electronic equipment of license plate type | |
| CN111753612A (en) | Method and device for detecting sprinkled object and storage medium | |
| CN113221839B (en) | Automatic truck image identification method and system | |
| CN113723300A (en) | Artificial intelligence-based fire monitoring method and device and storage medium | |
| CN115131283A (en) | Defect detection and model training method, device, equipment and medium for target object | |
| CN111523415A (en) | Image-based two-passenger one-dangerous vehicle detection method and device | |
| CN115424217A (en) | AI vision-based intelligent vehicle identification method and device and electronic equipment | |
| CN116993970A (en) | Oil and gas pipeline excavator occupation pressure detection method and system based on yolov5 | |
| CN108734113A (en) | Vehicle automatic marking method, storage medium, electronic equipment, system | |
| CN103544716A (en) | Method and device for classifying colors of pixels of image | |
| CN111652846A (en) | Semiconductor defect identification method based on characteristic pyramid convolution neural network | |
| CN109858310A (en) | Vehicles and Traffic Signs detection method | |
| CN112784675B (en) | Target detection method and device, storage medium and terminal | |
| CN115346118A (en) | Identification and positioning method for vehicles in scrap steel discharge yard | |
| CN113177528A (en) | License plate recognition method and system based on multi-task learning strategy training network model | |
| CN112329858A (en) | Image recognition method for breakage fault of anti-loosening iron wire of railway motor car | |
| CN106824826A (en) | A kind of corn monoploid sorting system | |
| CN112017065A (en) | Vehicle loss assessment and claim settlement method and device and computer readable storage medium | |
| CN116129100A (en) | Truck part positioning and fault identifying method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |