LCD-Net: A Lightweight Remote Sensing Change Detection Network Combining Feature Fusion and Gating Mechanism

TL;DR

A lightweight remote sensing change detection network (LCD-Net) that reduces model size and computational cost while maintaining high detection performance, and is well-suited for real-time applications in resource-limited settings.

Abstract

Remote sensing image change detection (RSCD) is crucial for monitoring dynamic surface changes, with applications ranging from environmental monitoring to disaster assessment. While traditional CNN-based methods have improved detection accuracy, they often suffer from high computational complexity and large parameter counts, limiting their use in resource-constrained environments. To address these challenges, we propose a Lightweight remote sensing Change Detection Network (LCD-Net in short) that reduces model size and computational cost while maintaining high detection performance. LCD-Net employs MobileNetV2 as the encoder to efficiently extract features from bitemporal images. A Temporal Interaction and Fusion Module (TIF) enhances the interaction between bitemporal features, improving temporal context awareness. Additionally, the Feature Fusion Module (FFM) aggregates multiscale features to better capture subtle changes while suppressing background noise. The Gated Mechanism Module (GMM) in the decoder further enhances feature learning by dynamically adjusting channel weights, emphasizing key change regions. Experiments on LEVIR-CD+, SYSU, and S2Looking datasets show that LCD-Net achieves competitive performance with just 2.56M parameters and 4.45G FLOPs, making it well-suited for real-time applications in resource-limited settings. The code is available at https://github.com/WenyuLiu6/LCD-Net.

Figures (10)

• Figure 1: Comparison of model complexity on the SYSU dataset, illustrating FLOPs vs. F1-score, FLOPs vs. Params, and Params vs. F1-score for different methods.
• Figure 2: This set of images is from the SYSU dataset. Among them, FFM represents the Feature Fusion Module. The displayed colors correspond to true positives (white), false positives (red), true negatives (black), and false negatives (blue).
• Figure 3: The proposed LCD-Net architecture. The framework includes an encoder, Feature Fusion Module(FFM), and decoder. The encoder, based on MobileNetV2, extracts bi-temporal features enhanced by the TIF. The decoder upsamples features and uses the Gating Mechanism Module (GMM) to improve key feature extraction, producing the change detection map.
• Figure 4: Illustration of the proposed Temporal Interaction and Fusion Module (TIF).
• Figure 5: Illustration of the proposed Feature Fusion Module (FFM).