CN114120109A - Belt longitudinal tearing detection method based on neural network - Google Patents
Belt longitudinal tearing detection method based on neural network Download PDFInfo
- Publication number
- CN114120109A CN114120109A CN202111375025.6A CN202111375025A CN114120109A CN 114120109 A CN114120109 A CN 114120109A CN 202111375025 A CN202111375025 A CN 202111375025A CN 114120109 A CN114120109 A CN 114120109A
- Authority
- CN
- China
- Prior art keywords
- convolutional
- layer
- neural network
- pictures
- subset
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The disclosure relates to a belt longitudinal tearing detection method based on a neural network, which comprises the following steps: s1, constructing a convolutional neural network algorithm model, wherein the convolutional neural network algorithm model comprises a first convolutional network and a second convolutional network; s2, training the convolutional neural network algorithm model through a preset training set, wherein the preset training set comprises a first picture set containing longitudinal tearing characteristics and a second picture set containing frame labeling of the longitudinal tearing characteristics; s3, collecting image information of a coal conveying belt area; s4, inputting the image information into the convolutional neural network algorithm model; and S5, judging whether the longitudinal tearing characteristics exist in the image information, and outputting the picture frame containing the frame marks. Compared with a general algorithm based on edge detection or pixel identification, the belt longitudinal tearing detection method based on the neural network has the beneficial effect of high accuracy.
Description
Technical Field
The invention relates to a production monitoring method, in particular to a belt longitudinal tearing detection method based on a neural network.
Background
The coal conveying belt machine is an important device for fuel transportation in a thermal power plant, and the coal conveying belt is generally divided into a common belt and a steel rope core belt. The thermal power plant usually uses a common belt, the bandwidth of the belt is generally about 1-1.4m, and the price is more and more expensive along with the increase of the bandwidth. Because the belt conveyor has a high running speed, the belt is sometimes torn due to equipment or other reasons in the transportation process, once the belt is torn, the tearing range is generally dozens of meters or hundreds of meters in length, so that economic loss which is possibly brought about is few tens of thousands or more than hundreds of thousands, and meanwhile, the normal operation of a unit can be directly influenced. Although some units install a belt tearing prevention device on a coal conveying system, the belt tearing cannot be effectively prevented in actual operation due to various problems of system design, signal transmission and the like. Therefore, the tearing detection, prevention and timely treatment of the coal conveying belt are fully performed, and the method is an important work for ensuring the normal production of the unit. And for timely treatment afterwards, the tearing detection of the coal conveying belt is a key link for prevention and timely treatment. Generally, the tearing detection of the belt can be realized by manual inspection or video monitoring. With the development of science and technology, the belt tearing detection based on machine vision is applied at present. For example, chinese patent application CN105293003A discloses a belt longitudinal tear detection method based on machine vision, in which an image of a collected coal conveying belt related region is segmented, and then a local mode is assigned to the segmented region, where the mode of pixels is the local mode whose gray value is closest to the average value of the local mode; then grouping pixels with geometric characteristics of line structures together; and further, calculating 2-4 order Zernike moments of pixels in the area of the line structure, judging whether the line is a line by using a template matching method, if the line is matched with the line, considering that the belt is longitudinally torn, and if not, judging that the belt is normal. However, the inventor finds that the method for calculating based on the pixel gray-scale value is low in accuracy and prone to false recognition.
Disclosure of Invention
In view of the above problems in the prior art, an object of the present invention is to provide a belt rip detection method based on a neural network, which has higher detection accuracy.
In order to achieve the above object, an aspect of the present invention provides a belt rip detection method based on a neural network, including:
s1, constructing a convolutional neural network algorithm model, wherein the convolutional neural network algorithm model comprises a first convolutional network and a second convolutional network;
s2, training the convolutional neural network algorithm model through a preset training set, wherein the preset training set comprises a first picture set containing longitudinal tearing characteristics and a second picture set containing frame labeling of the longitudinal tearing characteristics;
s3, collecting image information of a coal conveying belt area;
s4, inputting the image information into the convolutional neural network algorithm model;
and S5, judging whether the longitudinal tearing characteristics exist in the image information, and outputting the picture frame containing the frame marks.
Preferably, the first convolution network includes a first convolution layer and a pooling layer, and the first convolution layer is configured to perform convolution processing on the first picture set to obtain a first picture subset including a longitudinal tearing feature; the pooling layer is configured to pool the first subset of pictures to obtain a first feature vector having a dimension M.
Preferably, the second convolution network includes a second convolution layer and a feature extraction layer, and the second convolution layer is configured to perform convolution processing on the second picture set to obtain a second picture subset including vertical tearing features; the feature extraction layer is configured to extract, based on the second picture subset, M second feature vectors respectively corresponding to the M regions.
Preferably, the convolutional neural network further includes a combination layer configured to combine the first feature vector and the second feature vector to obtain M combination vectors.
Preferably, the convolutional neural network further comprises a bounding box regression layer, wherein the bounding box regression layer is configured to output a predicted longitudinally-torn region in the second picture subset based on the M combined vectors, and the longitudinally-torn region is output with a visual bounding box label.
Preferably, the first convolution layer and the second convolution layer share the same convolution layer.
Preferably, the combination layers are combined by a support vector machine when vector combination is performed.
Preferably, the frame regression layer includes a first hidden layer and a second hidden layer, and the first hidden layer is configured to predict the probability of each region of the pictures of the first picture subset; the second hidden layer is configured to produce a visual border in the second subset of pictures.
Compared with the prior art, the belt longitudinal tearing detection method based on the neural network provided by the invention can be used for respectively training the determined longitudinal tearing picture and the longitudinal tearing picture with the frame mark to obtain the trained convolutional neural network algorithm model, so that the picture input by the image acquisition device can firstly obtain a picture subset possibly having longitudinal tearing through the first convolutional network, then secondary screening is carried out through the second convolutional network, and then the longitudinal tearing picture with the frame mark is output to a human-computer interface through feature vector combination and frame regression, so that a worker can conveniently check the longitudinal tearing picture. Compared with the common algorithm based on edge detection or pixel identification, the method has the advantage of high accuracy.
Drawings
Fig. 1 is a flowchart of a method for detecting belt rip based on a neural network according to the present invention.
Fig. 2 is a block diagram of a system to which the belt rip detection method based on the neural network of the present invention is applied.
Detailed Description
In order to make the technical solutions of the present invention better understood, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
Various aspects and features of the present invention are described herein with reference to the drawings.
These and other characteristics of the invention will become apparent from the following description of a preferred form of embodiment, given as a non-limiting example, with reference to the accompanying drawings.
It should also be understood that, although the invention has been described with reference to some specific examples, a person of skill in the art shall certainly be able to achieve many other equivalent forms of the invention, having the characteristics as set forth in the claims and hence all coming within the field of protection defined thereby.
The above and other aspects, features and advantages of the present invention will become more apparent in view of the following detailed description when taken in conjunction with the accompanying drawings.
As shown in fig. 1, a belt rip detection method based on a neural network according to an embodiment of the present invention includes:
s1, constructing a convolutional neural network algorithm model, wherein the convolutional neural network algorithm model comprises a first convolutional network and a second convolutional network;
s2, training the convolutional neural network algorithm model through a preset training set, wherein the preset training set comprises a first picture set containing longitudinal tearing characteristics and a second picture set containing frame labeling of the longitudinal tearing characteristics;
s3, collecting image information of a coal conveying belt area;
s4, inputting the image information into the convolutional neural network algorithm model;
and S5, judging whether the longitudinal tearing characteristics exist in the image information, and outputting the picture frame containing the frame marks.
It is worth to be noted that, for the construction of the convolutional neural network algorithm model, the prior art belongs to the prior art, and the difference is that, as shown in fig. 2, the first convolutional network in the present invention includes a first convolutional layer and a pooling layer, where the first convolutional layer is configured to perform convolutional processing on a first picture set, so as to obtain a first picture subset including a longitudinal tear feature; the pooling layer is configured to pool the first subset of pictures to obtain a first feature vector having a dimension M. Correspondingly, in the invention, the second convolution network includes a second convolution layer and a feature extraction layer, and the second convolution layer is configured to perform convolution processing on the second picture set to obtain a second picture subset including longitudinal tearing features; the feature extraction layer is configured to extract, based on the second picture subset, M second feature vectors respectively corresponding to the M regions.
Further, in the present invention, the convolutional neural network further includes a combination layer, the combination layer is configured to combine the first feature vector and the second feature vector to obtain M combination vectors, and the combination layer performs combination by using a support vector machine when performing vector combination. The convolutional neural network further comprises a bounding box regression layer configured to output a predicted longitudinally torn region in the second subset of pictures based on the M combined vectors, the longitudinally torn region being output with a visual bounding box label.
Moreover, in some modifications, in order to further optimize the convolutional neural network algorithm model and improve the operation speed and accuracy, the first convolutional layer and the second convolutional layer share the same convolutional layer.
Various operations or functions are described herein that may be implemented as or defined as software code or instructions. Such content may be directly executable ("object" or "executable" form) source code or differential code ("delta" or "patch" code). Software implementations of embodiments described herein may be provided via an article of manufacture having code or instructions stored therein or via a method of operating a communication interface to transmit data via the communication interface. A machine or computer-readable storage medium may cause a machine to perform the functions or operations described, and includes any mechanism for storing information in a form accessible by a machine (e.g., a computing device, an electronic system, etc.), such as recordable/non-recordable media (e.g., Read Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, etc.). A communication interface includes any mechanism that interfaces to any of a hardwired, wireless, optical, etc. medium to communicate with another device, such as a memory bus interface, a processor bus interface, an internet connection, a disk controller, etc. The communication interface may be configured by providing configuration parameters and/or transmitting signals to prepare the communication interface to provide data signals describing the software content. The communication interface may be accessed via one or more commands or signals sent to the communication interface.
The above embodiments are only exemplary embodiments of the present invention, and are not intended to limit the present invention, and the scope of the present invention is defined by the claims. Various modifications and equivalents may be made by those skilled in the art within the spirit and scope of the present invention, and such modifications and equivalents should also be considered as falling within the scope of the present invention.
Claims (8)
1. The belt longitudinal tearing detection method based on the neural network comprises the following steps:
s1, constructing a convolutional neural network algorithm model, wherein the convolutional neural network algorithm model comprises a first convolutional network and a second convolutional network;
s2, training the convolutional neural network algorithm model through a preset training set, wherein the preset training set comprises a first picture set containing longitudinal tearing characteristics and a second picture set containing frame labeling of the longitudinal tearing characteristics;
s3, collecting image information of a coal conveying belt area;
s4, inputting the image information into the convolutional neural network algorithm model;
and S5, judging whether the longitudinal tearing characteristics exist in the image information, and outputting the picture frame containing the frame marks.
2. The method of claim 1, the first convolutional network comprising a first convolutional layer and a pooling layer, the first convolutional layer configured to perform convolutional processing on a first set of pictures to obtain a first subset of pictures comprising vertical tearing features; the pooling layer is configured to pool the first subset of pictures to obtain a first feature vector having a dimension M.
3. The method of claim 2, the second convolutional network comprising a second convolutional layer and a feature extraction layer, the second convolutional layer configured to perform convolutional processing on a second set of pictures to obtain a second subset of pictures comprising vertical tearing features; the feature extraction layer is configured to extract, based on the second picture subset, M second feature vectors respectively corresponding to the M regions.
4. The method of claim 3, the convolutional neural network further comprising a combining layer configured to combine the first eigenvector and the second eigenvector to obtain M combined vectors.
5. The method of claim 4, the convolutional neural network further comprising a bounding box regression layer configured to output a predicted longitudinally-torn region in the second subset of pictures based on the M combined vectors, the longitudinally-torn region output with a visual bounding box label.
6. The method of claim 4, the first convolutional layer and the second convolutional layer sharing a same convolutional layer.
7. The method of claim 4, the combining layer, when performing vector combining, performs combining by a support vector machine.
8. The method of claim 4, the bounding box regression layer comprising a first hidden layer and a second hidden layer, the first hidden layer configured to predict a probability for each region of a picture of the first subset of pictures; the second hidden layer is configured to produce a visual border in the second subset of pictures.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111375025.6A CN114120109A (en) | 2021-11-19 | 2021-11-19 | Belt longitudinal tearing detection method based on neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111375025.6A CN114120109A (en) | 2021-11-19 | 2021-11-19 | Belt longitudinal tearing detection method based on neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114120109A true CN114120109A (en) | 2022-03-01 |
Family
ID=80396689
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111375025.6A Pending CN114120109A (en) | 2021-11-19 | 2021-11-19 | Belt longitudinal tearing detection method based on neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114120109A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114529824A (en) * | 2022-04-24 | 2022-05-24 | 江苏科比泰智能科技有限公司 | Image recognition-based belt monitoring system and method for adhesive tape conveyor |
-
2021
- 2021-11-19 CN CN202111375025.6A patent/CN114120109A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114529824A (en) * | 2022-04-24 | 2022-05-24 | 江苏科比泰智能科技有限公司 | Image recognition-based belt monitoring system and method for adhesive tape conveyor |
| CN114529824B (en) * | 2022-04-24 | 2022-06-28 | 江苏科比泰智能科技有限公司 | Image recognition-based belt monitoring system and method for adhesive tape conveyor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111079747B (en) | Railway wagon bogie side frame fracture fault image identification method | |
| CN110490842B (en) | Strip steel surface defect detection method based on deep learning | |
| CN110533023B (en) | Method and device for detecting and identifying foreign matters of railway wagon | |
| WO2021151277A1 (en) | Method and apparatus for determining severity of damage on target object, electronic device, and storage medium | |
| CN114120109A (en) | Belt longitudinal tearing detection method based on neural network | |
| CN111709455A (en) | Evaluation method of wine bottle surface defect model, electronic device and storage medium | |
| CN113408361B (en) | Mining conveyor belt massive material detection method and system based on deep learning | |
| CN115810132B (en) | Crack orientation identification method, device, equipment and storage medium | |
| CN111709452A (en) | Evaluation method of wine bottle surface defect model, electronic device and storage medium | |
| CN116434346A (en) | Method and device for detecting customer behaviors in unattended store and storage medium | |
| CN116206155A (en) | Waste steel classification and identification method based on YOLOv5 network | |
| CN112818987B (en) | Method and system for identifying and correcting display content of electronic bus stop board | |
| CN114120175A (en) | Method for identifying foreign matters on coal conveying belt based on computer vision | |
| CN114039279A (en) | Control cabinet monitoring method and system in rail transit station | |
| CN114399671A (en) | Target identification method and device | |
| CN112256906A (en) | Method, device and storage medium for marking annotation on display screen | |
| CN112200107A (en) | Invoice text detection method | |
| CN112329783B (en) | Image processing-based coupler yoke break identification method | |
| CN115713758B (en) | Carriage identification method, system, device and storage medium | |
| CN111401104B (en) | Classification model training method, classification method, device, equipment and storage medium | |
| CN114066895B (en) | Detection method and device for pantograph slide plate | |
| CN115331085B (en) | Bogie spring supporting plate detection method based on deep learning | |
| CN117197131B (en) | Method and apparatus for conveyor belt tear identification and computing device | |
| CN112730427B (en) | Product surface defect detection method and system based on machine vision | |
| CN115330872A (en) | Temperature monitoring method, system and medium of cable processing controller based on Internet of things |
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 |