Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching

In order to improve the detection accuracy of SSD (Single Shot MultiBox Detector), an A-SSD (Anchor-object SSD) object detection algorithm based on Anchor-object matching is proposed. In the feature extraction part of the algorithm, parallel convolution and hole convolution are used to form the rece...

Full description

Saved in:
Bibliographic Details
Main Authors: LI Cheng-yan, ZHAO Shuai, CHE Zi-xuan
Format: Article
Language:zho
Published: Harbin University of Science and Technology Publications 2022-06-01
Series:Journal of Harbin University of Science and Technology
Subjects:
Online Access:https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2098
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849711609258180608
author LI Cheng-yan
ZHAO Shuai
CHE Zi-xuan
author_facet LI Cheng-yan
ZHAO Shuai
CHE Zi-xuan
author_sort LI Cheng-yan
collection DOAJ
description In order to improve the detection accuracy of SSD (Single Shot MultiBox Detector), an A-SSD (Anchor-object SSD) object detection algorithm based on Anchor-object matching is proposed. In the feature extraction part of the algorithm, parallel convolution and hole convolution are used to form the receptive field module, which increases the receptive field of the feature map and obtains multi-scale feature information, combining the shallow features containing texture, edge and other detailed information with rich fusion of deep features of semantic information. In the detector part of the algorithm, the Anchor-object matching method combined with the SSD multi-layer feature map is used to construct the corresponding Anchor package for each detection target. Through the selection-suppression optimization strategy, the Anchor with higher confidence is selected to update the model. Anchor evaluation score, in continuous iterative learning, the model parameters, Anchor position coordinates and classification confidence are continuously optimized. The mAP of the A-SSD algorithm on the PASCAL VOC data set reached 80.7, and the missed detection rate of the A-SSD algorithm on the workshop pedestrian data set was 3.5%, and the accuracy rate was 91.5%.
format Article
id doaj-art-063069dcd24b436ca0766896833d8a98
institution DOAJ
issn 1007-2683
language zho
publishDate 2022-06-01
publisher Harbin University of Science and Technology Publications
record_format Article
series Journal of Harbin University of Science and Technology
spelling doaj-art-063069dcd24b436ca0766896833d8a982025-08-20T03:14:35ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832022-06-012703738110.15938/j.jhust.2022.03.010Object Detection Algorithm Based on Feature Enhancement and Anchor-object MatchingLI Cheng-yan0ZHAO Shuai1CHE Zi-xuan2School of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, ChinaIn order to improve the detection accuracy of SSD (Single Shot MultiBox Detector), an A-SSD (Anchor-object SSD) object detection algorithm based on Anchor-object matching is proposed. In the feature extraction part of the algorithm, parallel convolution and hole convolution are used to form the receptive field module, which increases the receptive field of the feature map and obtains multi-scale feature information, combining the shallow features containing texture, edge and other detailed information with rich fusion of deep features of semantic information. In the detector part of the algorithm, the Anchor-object matching method combined with the SSD multi-layer feature map is used to construct the corresponding Anchor package for each detection target. Through the selection-suppression optimization strategy, the Anchor with higher confidence is selected to update the model. Anchor evaluation score, in continuous iterative learning, the model parameters, Anchor position coordinates and classification confidence are continuously optimized. The mAP of the A-SSD algorithm on the PASCAL VOC data set reached 80.7, and the missed detection rate of the A-SSD algorithm on the workshop pedestrian data set was 3.5%, and the accuracy rate was 91.5%.https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2098object detectionssd algorithmfeature extractiondetectoranchor-object matching
spellingShingle LI Cheng-yan
ZHAO Shuai
CHE Zi-xuan
Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching
Journal of Harbin University of Science and Technology
object detection
ssd algorithm
feature extraction
detector
anchor-object matching
title Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching
title_full Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching
title_fullStr Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching
title_full_unstemmed Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching
title_short Object Detection Algorithm Based on Feature Enhancement and Anchor-object Matching
title_sort object detection algorithm based on feature enhancement and anchor object matching
topic object detection
ssd algorithm
feature extraction
detector
anchor-object matching
url https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2098
work_keys_str_mv AT lichengyan objectdetectionalgorithmbasedonfeatureenhancementandanchorobjectmatching
AT zhaoshuai objectdetectionalgorithmbasedonfeatureenhancementandanchorobjectmatching
AT chezixuan objectdetectionalgorithmbasedonfeatureenhancementandanchorobjectmatching