Journal
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
Volume 18, Issue 5, Pages 811-815Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LGRS.2020.2988032
Keywords
Feature extraction; Buildings; Semantics; Dams; Decoding; Task analysis; Image segmentation; Attention module; building change detection; deep learning; sample imbalance; semantic segmentation
Categories
Funding
- National Key Research and Development Program of China [2016YFC0802500]
- Major Project of High Resolution Earth Observation Systems [11-Y20A03-9001-16/17]
- National Natural Science Foundation of China [61871295]
Ask authors/readers for more resources
The proposed dual-task constrained deep Siamese convolutional network (DTCDSCN) model achieves both change detection and semantic segmentation simultaneously, improving feature extraction and representation to address the issue of lack of discriminative features in change detection. Experimental results demonstrate state-of-the-art performance on the WHU building data set, showcasing the effectiveness of the proposed method.
In recent years, building change detection methods have made great progress by introducing deep learning, but they still suffer from the problem of the extracted features not being discriminative enough, resulting in incomplete regions and irregular boundaries. To tackle this problem, we propose a dual-task constrained deep Siamese convolutional network (DTCDSCN) model, which contains three subnetworks: a change detection network and two semantic segmentation networks. DTCDSCN can accomplish both change detection and semantic segmentation at the same time, which can help to learn more discriminative object-level features and obtain a complete change detection map. Furthermore, we introduce a dual attention module (DAM) to exploit the interdependencies between channels and spatial positions, which improves the feature representation. We also improve the focal loss function to suppress the sample imbalance problem. The experimental results obtained with the WHU building data set show that the proposed method is effective for building change detection and achieves state-of-the-art performance in terms of four metrics on the WHU building data set: precision, recall, F1-score, and intersection over union.
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