Spatial-Spectral Binarized Neural Network for Panchromatic and Multi-spectral Images Fusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiang, Yizhen, Ma, Mengting, Zhu, Anqi, Ma, Xiaowen, Li, Jiaxin, Zhang, Wei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914066595840000
author Jiang, Yizhen
Ma, Mengting
Zhu, Anqi
Ma, Xiaowen
Li, Jiaxin
Zhang, Wei
author_facet Jiang, Yizhen
Ma, Mengting
Zhu, Anqi
Ma, Xiaowen
Li, Jiaxin
Zhang, Wei
contents Remote sensing pansharpening aims to reconstruct spatial-spectral properties during the fusion of panchromatic (PAN) images and low-resolution multi-spectral (LR-MS) images, finally generating the high-resolution multi-spectral (HR-MS) images. Although deep learning-based models have achieved excellent performance, they often come with high computational complexity, which hinder their applications on resource-limited devices. In this paper, we explore the feasibility of applying the binary neural network (BNN) to pan-sharpening. Nevertheless, there are two main issues with binarizing pan-sharpening models: (i) the binarization will cause serious spectral distortion due to the inconsistent spectral distribution of the PAN/LR-MS images; (ii) the common binary convolution kernel is difficult to adapt to the multi-scale and anisotropic spatial features of remote sensing objects, resulting in serious degradation of contours. To address the above issues, we design the customized spatial-spectral binarized convolution (S2B-Conv), which is composed of the Spectral-Redistribution Mechanism (SRM) and Gabor Spatial Feature Amplifier (GSFA). Specifically, SRM employs an affine transformation, generating its scaling and bias parameters through a dynamic learning process. GSFA, which randomly selects different frequencies and angles within a preset range, enables to better handle multi-scale and-directional spatial features. A series of S2B-Conv form a brand-new binary network for pan-sharpening, dubbed as S2BNet. Extensive quantitative and qualitative experiments have shown our high-efficiency binarized pan-sharpening method can attain a promising performance.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23321
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Spatial-Spectral Binarized Neural Network for Panchromatic and Multi-spectral Images Fusion
Jiang, Yizhen
Ma, Mengting
Zhu, Anqi
Ma, Xiaowen
Li, Jiaxin
Zhang, Wei
Computer Vision and Pattern Recognition
Remote sensing pansharpening aims to reconstruct spatial-spectral properties during the fusion of panchromatic (PAN) images and low-resolution multi-spectral (LR-MS) images, finally generating the high-resolution multi-spectral (HR-MS) images. Although deep learning-based models have achieved excellent performance, they often come with high computational complexity, which hinder their applications on resource-limited devices. In this paper, we explore the feasibility of applying the binary neural network (BNN) to pan-sharpening. Nevertheless, there are two main issues with binarizing pan-sharpening models: (i) the binarization will cause serious spectral distortion due to the inconsistent spectral distribution of the PAN/LR-MS images; (ii) the common binary convolution kernel is difficult to adapt to the multi-scale and anisotropic spatial features of remote sensing objects, resulting in serious degradation of contours. To address the above issues, we design the customized spatial-spectral binarized convolution (S2B-Conv), which is composed of the Spectral-Redistribution Mechanism (SRM) and Gabor Spatial Feature Amplifier (GSFA). Specifically, SRM employs an affine transformation, generating its scaling and bias parameters through a dynamic learning process. GSFA, which randomly selects different frequencies and angles within a preset range, enables to better handle multi-scale and-directional spatial features. A series of S2B-Conv form a brand-new binary network for pan-sharpening, dubbed as S2BNet. Extensive quantitative and qualitative experiments have shown our high-efficiency binarized pan-sharpening method can attain a promising performance.
title Spatial-Spectral Binarized Neural Network for Panchromatic and Multi-spectral Images Fusion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.23321