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

A Pedestrian Recognition Method Based on the Combination of Deep Learning and Attribute Learning

A pedestrian recognition and deep learning technology, applied in the field of pattern recognition, can solve problems such as lack of semantic expression ability, influence recognition ability, non-conformity, etc., and achieve the effect of good semantic expression ability, good recognition rate, and cost reduction.

Active Publication Date: 2018-04-17
南京昭视智能科技有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage is that these low-level feature data do not have good semantic expression ability. In addition, when using the low-level feature data, it is usually necessary to assume that the light and viewing angle are constant, which does not meet the actual environmental conditions and greatly affects the recognition ability.

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
  • A Pedestrian Recognition Method Based on the Combination of Deep Learning and Attribute Learning
  • A Pedestrian Recognition Method Based on the Combination of Deep Learning and Attribute Learning
  • A Pedestrian Recognition Method Based on the Combination of Deep Learning and Attribute Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The invention is based on a pedestrian recognition method combining deep learning and attribute learning, and is divided into four parts: deep learning, attribute learning, attribute category mapping relationship learning and testing. It combines deep learning with attribute learning to extract the preferred features of images and represent them with better semantics. Among them, deep learning is divided into two stages: building a deep learning model and model training. In the stage of building a deep learning model, construct a multi-layer convolutional neural network model, initialize the model and set the relevant parameters of the model; The method tunes the parameters of the convolutional neural network. In the attribute learning part, an attribute classifier is set for each attribute. The preferred features trained from the convolutional neural network model are input to each classifier, and the classifiers learn the attributes through the attribute labels of t...

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 discloses a pedestrian recognition method based on the combination of deep learning and attribute learning. The method constructs a convolutional neural network with five hidden layers, uses a deconvolution method to train the network, and combines attribute learning. The concept is to input the preferred features obtained from the convolutional neural network into each attribute classifier to obtain the posterior probability of the attribute of the sample, and then combine the attribute category mapping relationship to obtain the posterior probability of the category, so as to judge the category of the sample. This method has good detection and recognition performance, can extract the essential features of the image, and because the attribute has better semantic expression performance than the low-level features, and is insensitive to light and viewing angle, the algorithm has a better recognition effect .

Description

technical field [0001] The invention relates to the technical field of pattern recognition, in particular to a pedestrian recognition method based on the combination of deep learning and attribute learning. Background technique [0002] Pedestrian recognition has broad application prospects in video control, robotics, intelligent transportation, multimedia retrieval and other fields, and it is also a popular research object in the field of computer vision in recent years. However, because pedestrian recognition involves the calculation and analysis of a large amount of data, plus the interference of environmental factors such as light and viewing angle, traditional recognition algorithms cannot extract the preferred features of the image and express them through better semantics, resulting in a limited recognition rate. [0003] A traditional recognition algorithm for pedestrian recognition is artificial neural network. It abstracts the human brain neuron network from the p...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06K9/66
CPCG06V40/10G06V30/194G06F18/2411
Inventor 成科扬张纯许方洁王卫东汪树胜羊立
Owner 南京昭视智能科技有限公司
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