Journal
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
Volume 32, Issue 10, Pages 6615-6626Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCSVT.2022.3176055
Keywords
Feature extraction; Remote sensing; Task analysis; Imaging; Transformers; Interference; Computer vision; Building change detection; deep learning; multi-scale; attention mechanism
Categories
Funding
- National Natural Science Foundation of China [61806206, 62172417]
- Natural Science Foundation of Jiangsu Province [BK20201346, BK20180639]
- Six Talent Peaks Project in Jiangsu Province [2015-DZXX-010, 2018-XYDXX-044]
Ask authors/readers for more resources
The neural network-based remote sensing image change detection method proposed in this study addresses the challenges of imaging interference and class imbalance problems under high-resolution conditions. It uses the siamese strategy and multi-head self-attention mechanism to reduce imaging differences and exploit inter-temporal information. It also incorporates a learnable multi-part feature learning module to obtain more comprehensive features. The mixed loss function strategy ensures effective convergence and excludes negative sample interference.
The neural network-based remote sensing image change detection method faces a large amount of imaging interference and severe class imbalance problems under high-resolution conditions, which bring new challenges to the accuracy of the detection network. In this work, to address the imaging interference caused by different imaging angles and times, the siamese strategy and multi-head self-attention mechanism are used to reduce the imaging differences between the dual-temporal images and fully exploit the inter-temporal information. Secondly, a learnable multi-part feature learning module is used to adaptively exploit features from different scales to obtain more comprehensive features. Finally, a mixed loss function strategy is used to ensure that the network converges effectively and excludes the adverse interference of a large number of negative samples to the network. Extensive experiments show that our method outperforms numerous methods on LEVIR-CD, WHU, and DSIFN datasets.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available