Research on Target Detection and Recognition Algorithm Based on Deep Learning

With the rapid development of robotics, people have higher requirements for detection and recognition of targets. The traditional target recognition algorithm can identify the type of the target, but can not identify the position information of the target, and the recognition rate of the traditional...

Full description

Saved in:
Bibliographic Details
Main Authors: Wang, Hui, Liu, Chaoda, Yu, Lijun, Zhao, Jingyuan
Format: Conference Proceeding
Language:English
Subjects:
Classification algorithms
Convolution
Convolutional neural networks
Deep learning Convolutional neural network
Feature extraction
Proposals
Target detection
Target recognition
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the rapid development of robotics, people have higher requirements for detection and recognition of targets. The traditional target recognition algorithm can identify the type of the target, but can not identify the position information of the target, and the recognition rate of the traditional target recognition algorithm is low when the same target appears in the image. To this end, this paper proposes a target detection and recognition algorithm based on deep learning technology using convolutional neural network and region segmentation. The algorithm uses the region proposal to segment the image in the recognition process, which can recognize the type of target and precisely calibrate the position of the target in the image. The simulation results show that the algorithm has a high recognition rate and is sensitive to small target recognition.
ISSN:2161-2927
DOI:10.23919/ChiCC.2019.8865560