Remote Sensing Multimodal Image Matching Based on Structure Feature and Learnable Matching Network

Matching remotely sensed multimodal images is a crucial process that poses significant challenges due to nonlinear radiometric differences and substantial image noise. To overcome these difficulties, this study presents novel practical template-matching algorithm specifically designed for purpose. Unlike traditional approaches rely on intensity, the proposed focuses matching based their geometric structure information. This approach enables method effectively adapt variations in grayscale caused by differences. enhance performance, principal component analysis calculation log-Gabor filter estimate structural feature of image. The can accurately even under severe noise distortion. In addition, learnable network similarity measuring gradient reversal difference among images. Infrared, visible light, synthetic aperture radar are adopted evaluation, verify performance algorithm. Based results, has distinct advantage over other state-of-the-art algorithms.