Paper
29 October 2018 A multi-scale feature fusion target detection algorithm
Dong Chong, Jingmei Li, Jiaxiang Wang
Author Affiliations +
Proceedings Volume 10836, 2018 International Conference on Image and Video Processing, and Artificial Intelligence; 108361N (2018) https://doi.org/10.1117/12.2514046
Event: 2018 International Conference on Image, Video Processing and Artificial Intelligence, 2018, Shanghai, China
Abstract
For existing Faster R-CNN and single shot multibox detector (SSD) target detection algorithms, they all have the problem of low object detection accuracy under small target conditions. This paper proposes a general and effective target detection algorithm and the detection accuracy has greatly improved for smaller targets. The algorithm is divided into two parts. In the first part, in the feature extraction process, the feature map extracted by the basic feature extraction network is deconvoluted and merged with the previous layer feature map to generate Multi-scale feature maps with rich semantics and high resolution. Using proposed multi-scale feature maps to generate proposals. The second part uses the generated proposals to be sent to the Faster R-CNN network for classification and detection. Experiments show that using this algorithm for target detection can not only improve the recall of proposals, but also improve the accuracy of target detection, especially for small targets. The algorithm provides a new idea for small target detection.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Dong Chong, Jingmei Li, and Jiaxiang Wang "A multi-scale feature fusion target detection algorithm", Proc. SPIE 10836, 2018 International Conference on Image and Video Processing, and Artificial Intelligence, 108361N (29 October 2018); https://doi.org/10.1117/12.2514046
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Detection and tracking algorithms

Target detection

Feature extraction

Convolution

Image resolution

Algorithm development

Classification systems

RELATED CONTENT

Object detection based on deep learning
Proceedings of SPIE (December 16 2021)
Remote logo detection using angle-distance histograms
Proceedings of SPIE (May 19 2016)
Consecutive pedestrian tracking in large scale space
Proceedings of SPIE (September 28 2016)
Parallel Processing For Computer Vision
Proceedings of SPIE (November 22 1982)

Back to Top