DWTFreqNet: Infrared Small Target Detection via Wavelet-Driven Frequency Matching and Saliency-Difference Optimization
Top Cited Papers
- 11 September 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-15
Abstract
In the field of infrared small target detection, targets generally exhibit dim characteristics, and difficult to distinguish from background clutter. Learning-based methods enhance feature representation through layer-by-layer propagation, but the sparse target information often diminishes. To address this, we propose DWTFreqNet, a network that splits input data to enhance both local saliency and global contextual differences. It incorporates complementary feature extraction modules designed to match the data distribution characteristics. Specifically, it first utilizes the discrete wavelet transform (DWT) to decompose the input data into low- and high-frequency components. For the low-frequency part, which carries key target information, we apply component-differential dense connections and DWT-based downsampling to maintain feature integrity. For the high-frequency part, rich in target-background contrast, an Adaptive Wavelet Guidance Mechanism optimizes multi-component fusion via adaptive weighting, while a Layer-wide Discrepancy Relationship Capture Module enhances target discrimination by linking multi-scale feature maps. Comparative experiments on public datasets demonstrate its superiority over state-of-the-art methods. The code will be available at https://github.com/Kingwin97/DWTFreqNet.Keywords
Funding Information
- National Key Research and Development Program of China (2023YFC3108500)
- National Natural Science Foundation of China (62205243)
- Open Project Funds for the Key Laboratory of Space Photoelectric Detection and Perception (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology (NJ2024027-1)
- Fundamental Research Funds for the Central Universities (NJ2024027)
- Tianjin Municipal Science and Technology Bureau (23YFZCSN00230)
- Special Project for Enterprise Research and Development in Tiankai Higher Education Innovation Park (23YFZXYC00012)
This publication has 61 references indexed in Scilit:
- Cascaded frameworks in underwater optical image restorationInformation Fusion, 2024
- IRSAM: Advancing Segment Anything Model for Infrared Small Target Detection2024
- PODB: A learning-based polarimetric object detection benchmark for road scenes in adverse weather conditionsInformation Fusion, 2024
- Image Compression Using Discrete Wavelet Transform and Convolution Neural NetworksJournal of Electrical Engineering & Technology, 2024
- An improved deep learning-based optimal object detection system from imagesMultimedia Tools and Applications, 2023
- LDWS-net: A learnable deep wavelet scattering network for RGB salient object detectionImage and Vision Computing, 2023
- MFIF-DWT-CNN: Multi-focus ımage fusion based on discrete wavelet transform with deep convolutional neural networkMultimedia Tools and Applications, 2023
- WaveSNet: Wavelet Integrated Deep Networks for Image Segmentation2022
- Wavelet U-Net for Medical Image Segmentation2020
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015