With the increasing significance of high-quality, high-resolution multispectral images (HRMS) in various domains, pansharpening, which fuses low-resolution (LRMS) with panchromatic (PAN), has gained considerable attention. However, current deep learning methods have limitations capturing global long-range dependencies and incorporating spectral characteristics across different bands (MS). Additionally, model-based approaches do not effectively utilize multi-scale information between LRMS...

Abstract. Accurate matching of multimodal remote sensing (RS) images (e.g., optical, infrared, LiDAR, SAR, and rasterized maps) is still an ongoing challenge because nonlinear radiometric differences (NRD) between these images. Considering that structural properties are preserved images, this paper proposes a robust method based on multi-directional multi-scale features, which consist two critical steps. Firstly, novel descriptor named the Steerable Filters first- second-Order Channels...

Pansharpening aims to produce a high-resolution multispectral (HRMS) image by combining low-resolution (LRMS) with panchromatic (PAN) through fusion process. Deep learning (DL)-based pansharpening methods have demonstrated impressive results in generating high-quality HRMS images. However, they suffer from lack of interpretability due their black-box network architectures. Recently, model-based deep unrolling networks been proposed improve the networks. Among these approaches, multi-source...

Pansharpening aims to sharpen low resolution (LR) multispectral (MS) images with the help of corresponding high (HR) panchromatic (PAN) obtain HRMS images. Model-based pansharpening methods manually design objective functions via observation model and hand-crafted priors. However, inevitable performance degradation may occur in case that prior is invalid. Although many deep learning based end-to-end have been proposed recently, they still need be improved due insufficient study on related...

Abstract. This paper presents a fast and robust method for the registration of multimodal remote sensing data (e.g., optical, LiDAR, SAR map). The proposed is based on hypothesis that structural similarity between images preserved across different modalities. In definition method, we first develop pixel-wise feature descriptor named Dense Orientated Gradient Histogram (DOGH), which can be computed effectively at every pixel to non-linear intensity differences images. Then metric DOGH built...

Abstract. Co-Registration of aerial imagery and Light Detection Ranging (LiDAR) data is quilt challenging because the different imaging mechanism causes significant geometric radiometric distortions between such data. To tackle problem, this paper proposes an automatic registration method based on structural features three-dimension (3D) phase correlation. In proposed method, LiDAR point cloud first transformed into intensity map, which used as reference image. Then, we employ Fast operator...

Abstract. With the expansion of optical and SAR image fusion application scenarios, it is necessary to integrate their information in land classification, feature recognition, target tracking. Current methods focus excessively on integrating multimodal enhance richness fused images, while neglecting highly corrupted visual perception results by modal differences speckle noise. To address this problem, paper we propose a novel framework named Visual Saliency Features Fusion (VSFF). We...

Abstract. Automatic registration of optical and synthetic aperture radar (SAR) images is a challenging task due to significant geometric deformation radiometric differences between two images. To address this issue, paper proposes an automatic method for SAR based on spatial constraint structure features. Firstly, the Harris detector with block strategy used extract evenly distributed feature points in Subsequently, local correction performed by using Rational Function Model, which...

Abstract. Image matching is a crucial procedure for multimodal remote sensing image processing. However, the performance of conventional methods often degraded in images due to significant nonlinear intensity differences. To address this problem, letter proposes novel feature representation named Main Structure with Histogram Orientated Phase Congruency (M-HOPC). M-HOPC able precisely capture similar structure properties between by reinforcing main information construction phase congruency...