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

Multi-scale small target equipment defect identification monitoring method

A defect recognition and small target technology, applied in neural learning methods, character and pattern recognition, image data processing, etc., can solve problems such as size, length and width uncertainty, target detection and recognition difficulties, etc.

Pending Publication Date: 2021-04-06
STATE GRID QINGHAI ELECTRIC POWER CO HAINAN POWER SUPPLY CO +1
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] In the task of equipment defect target detection and recognition, since the target to be detected may appear at any position in the image, and its size, length and width are also uncertain, it brings difficulties to target detection and recognition

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
  • Multi-scale small target equipment defect identification monitoring method
  • Multi-scale small target equipment defect identification monitoring method
  • Multi-scale small target equipment defect identification monitoring method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] The technical solutions of the present invention will be further explained below in conjunction with the accompanying drawings.

[0022] In the target detection and recognition algorithm based on the regional convolutional neural network, the method of target detection is performed by classifying the candidate region (Region Proposal) + convolutional neural network (CNN), that is, to find out the possible position of the target in the picture in advance, That is, the candidate area, and then use the convolutional neural network to extract features.

[0023] The present invention proposes a ResNet50 variant network structure design method that can enhance multi-scale small target convolution feature extraction. By increasing the width of the network, not only can each layer in the network learn sparse or non-sparse features, but also increase the adaptability of the network to multi-scale small targets. At the same time, the continuous use of two 3X3 convolution operati...

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 relates to the technical field of machine vision, in particular to an image recognition monitoring method for defects of small target equipment. The multi-scale small target equipment defect identification monitoring method is characterized by comprising the following steps: (1) constructing a single target detector for a plurality of categories; (2) predicting category scores and position offsets of a group of default bounding boxes fixed on the feature map by using a small convolution filter; and (3) generating predictions of different scales from the feature maps of different scales, and clearly separating the predictions through an aspect ratio. According to the method provided by the invention, an object region is predicted on feature maps of different convolution layers, discretized multi-scale and multi-proportion default boxes coordinates are output, and meanwhile, frame coordinate compensation of a series of candidate frames and confidence of each category are predicted by utilizing a small convolution kernel.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to an image recognition and monitoring method for small target equipment defects. Background technique [0002] In the task of equipment defect target detection and recognition, since the target to be detected may appear in any position in the image, and its size, length and width are also uncertain, it brings difficulties to target detection and recognition. Due to the uncertain size of the image, it takes a lot of computing resources to classify all possible locations and sizes on the image. Therefore, it is necessary to first generate some candidate regions (Region Proposals) to find out the regions that may contain objects. [0003] Convolutional neural network is a kind of neural network, and it is one of the most commonly used networks for deep learning. It has been widely used in the fields of machine vision, word processing and numerical analysis. Deep learning is th...

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): G06T7/00G06K9/32G06K9/62G06N3/04G06N3/08
CPCG06T7/0004G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/20024G06T2207/30181G06V10/25G06N3/045G06F18/241
Inventor 封琰谭毓卿袁源张海林吴童生王兴顺李沛然樊海峰梁珑田洪滨展毅晟芦国云郭妍谢占兰卢涛冯小霞张青梅沈娟马雅静刘有文严隆兴余国栋杨品梅邓蓉
Owner STATE GRID QINGHAI ELECTRIC POWER CO HAINAN POWER SUPPLY CO
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