Object detection algorithm based on improved Faster R-CNN

Min Yu; Dan Qu

doi:10.1117/12.2634517

12 May 2022 Object detection algorithm based on improved Faster R-CNN

Min Yu, Dan Qu

Proceedings Volume 12173, International Conference on Optics and Machine Vision (ICOMV 2022); 1217316 (2022) https://doi.org/10.1117/12.2634517
Event: International Conference on Optics and Machine Vision (ICOMV 2022), 2022, Guangzhou, China

Abstract

Aiming at the problem of the classic two-stage object detection algorithm Faster R-CNN that it is difficult to fully extract and fuse multi-scale features, an improved Faster R-CNN algorithm based on switchable atrous convolution and involution is proposed. This method first replaces the 3×3 convolution in the feature extraction module with a switchable atrous convolution to better capture the original rich information of the feature; then, replaces the 3×3 convolution in the feature fusion module with an involution further enhances the feature fusion effect at different stages. In order to test the detection effect of the algorithm, experiments were carried out on the MS COCO data set and the PASCAL VOC data set. Compared with the improved Faster R-CNN algorithm, the average precision of the proposed algorithm on the two data sets is improved by 5.5% and 2.2%, respectively. Among them, for the detection of larger objects in the COCO data set, the effect is improved more significantly.

Citation Download Citation

Min Yu and Dan Qu "Object detection algorithm based on improved Faster R-CNN", Proc. SPIE 12173, International Conference on Optics and Machine Vision (ICOMV 2022), 1217316 (12 May 2022); https://doi.org/10.1117/12.2634517

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Convolution

Detection and tracking algorithms

Feature extraction

Data modeling

Network architectures

Visualization

Computer vision technology

Show All Keywords

Keywords/Phrases

Search In:

Publication Years