M2F2-RCNN: Multi-functional Faster RCNN Based on Multi-scale Feature Fusion for Region Search in Remote Sensing Images

Shoulin Yin1, Liguo Wang2, Qunming Wang3, Mirjana Ivanović4 and Jinghui Yang5

  1. College of Information and Communication Engineering, Harbin Engineering University
    Harbin, 150001 China
    yslin@hit.edu.cn
  2. College of Information and Communications Engineering, Dalian Minzu University
    Dalian, 116000 China
    wangliguo@hrbeu.edu.cn
  3. College of Surveying and Geo-Informatics, Tongji University
    Shanghai, 200092 China
    111wqm@163.com
  4. Faculty of Sciences, University of Novi Sad
    21000 Novi Sad, Serbia
    mira@dmi.uns.ac.rs
  5. School of Information Engineering, China University of Geosciences
    Beijing, 100083, China
    yang06081102@163.com

Abstract

In order to realize fast and accurate search of sensitive regions in remote sensing images, we propose a multi-functional faster RCNN based on multi-scale feature fusion model for region search. The feature extraction network is based on ResNet50 and the dilated residual blocks are utilized for multi-layer and multi-scale feature fusion. We add a path aggregation network with a convolution block attention module (CBAM) attention mechanism in the backbone network to improve the efficiency of feature extraction. Then, the extracted feature map is processed, and RoIAlign is used to improve the pooling operation of regions of interest and it can improve the calculation speed. In the classification stage, an improved non-maximum suppression is used to improve the classification accuracy of the sensitive region. Finally, we conduct cross validation experiments on Google Earth dataset and the DOTA dataset. Meanwhile, the comparison experiments with the state-of-the-art methods also prove the high efficiency of the proposed method in region search ability.

Key words

remote sensing images, region search, multi-functional faster RCNN, multi-scale feature fusion, convolution block attention module

Digital Object Identifier (DOI)

https://doi.org/10.2298/CSIS230315054Y

Publication information

Volume 20, Issue 4 (September 2023)
Year of Publication: 2023
ISSN: 2406-1018 (Online)
Publisher: ComSIS Consortium

Full text

DownloadAvailable in PDF
Portable Document Format

How to cite

Yin, S., Wang, L., Wang, Q., Ivanović, M., Yang, J.: M2F2-RCNN: Multi-functional Faster RCNN Based on Multi-scale Feature Fusion for Region Search in Remote Sensing Images. Computer Science and Information Systems, Vol. 20, No. 4. (2023), https://doi.org/10.2298/CSIS230315054Y