Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Urban rail transit panoramic monitoring video fault detection method based on depth learning

A technology for urban rail transit and panoramic monitoring, applied in the field of urban rail transit panoramic surveillance video fault detection, to achieve the effect of reducing orders of magnitude, strong generalization ability, and reducing difficulty

Inactive Publication Date: 2017-05-24
HUAZHONG NORMAL UNIV +1
View PDF2 Cites 88 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the application process of these algorithms, the selection and definition of image features determine the quality of the detection results, which has great limitations.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Urban rail transit panoramic monitoring video fault detection method based on depth learning
  • Urban rail transit panoramic monitoring video fault detection method based on depth learning
  • Urban rail transit panoramic monitoring video fault detection method based on depth learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0056] 1. Implementation method of image dataset

[0057] The ultimate purpose of the present invention is to realize the fault recognition of the panoramic monitoring video of urban rail transit, the essence is to realize the recognition of the fault video image, and the image clarity and color cast are two extremely important judgment factors of the video quality detection. The present invention selects pictures with abnormal clarity, pictures with abnormal color cast and normal pictures as sample sets for training and testing, and the images are in RGB color mode. In this process, it is necessary to process the video with abnormal definition, video with abnormal color shift, and normal video, convert the surveillance video into a single frame image (that is, a frame image), and normalize the size of the image to construct an abnormal resolution image. Image Dataset, Color Cast Anomaly Image Dataset, and Normal Image Dataset. According to the distribution ratio, the images ...

Embodiment 2

[0074] The specific steps of the optimization of the fault video image recognition model based on convolutional neural network are as follows:

[0075] Step 1, expand the image dataset

[0076] If the image data set is too small and the extracted features are not sufficient, the abstraction ability of the model for feature learning will be limited, which will affect the test accuracy of the model; the training set with unbalanced positive and negative samples will also lead to inaccurate training results. rate decreased. Therefore, in order to obtain a model with higher accuracy, the present invention expands the 500 images with abnormal definition, 500 images with abnormal color cast, and 500 normal images in the original image data set to 3700 images with abnormal resolution and 4150 images with abnormal color cast. 4,150 normal pictures (the training of the binary classification model needs to ensure that the total number of positive and negative samples exceeds 10,000, of...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides an urban rail transit panoramic monitoring video fault detection method based on depth learning. The method comprises a data set construction process, a model training generation process and an image classification recognition process. The data set construction process processes a definition abnormity video, a colour cast abnormity video and a normal video in an urban rail transit panoramic monitoring video. A training set and a test set are classified. The model training generation process comprises model training and model test. The model training is to train a fault video image recognition model based on a convolution neural network. The convolutional neural network comprises a plurality of convolution layers and a plurality of full connection layers. The model test is to calculate the test accuracy. If expectation is not fulfilled, the fault video image recognition model is optimized. The image classification recognition process comprises the steps that a single-frame image to be recognized is input into the model, and the fault video image recognition model outputs an image classification result to complete the fault image detection of the urban rail transit panoramic monitoring video.

Description

technical field [0001] The invention belongs to the field of deep learning, and in particular relates to a fault detection method for urban rail transit panorama monitoring video based on deep learning. Background technique [0002] In recent years, with the growth of the number of private cars, the problem of urban traffic congestion has become a major problem faced by urban life, and more and more traffic accidents have occurred. These problems have increasingly affected the normal life of ordinary urban people. Therefore, the normal operation of urban rail transit panoramic surveillance video is particularly important, which requires real-time inspection of the operating status of urban rail transit panoramic surveillance video, timely identification of corresponding faults and timely maintenance. The panorama refers to the fusion and splicing of multiple surveillance videos and regional 3D models or a single surveillance video in the panoramic surveillance system, while ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/044G06F18/241G06F18/214
Inventor 赵刚刘闪徐丽华陈亚雯张傲
Owner HUAZHONG NORMAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com