An Image Registration Method Using Deep Residual Network Features for Multisource High-Resolution Remote Sensing Images

Xin Zhao,Hui Li,Ping Wang,Linhai Jing

doi:10.3390/rs13173425

Abstract

Accurate registration for multisource high-resolution remote sensing images is an essential step for various remote sensing applications. Due to the complexity of the feature and texture information of high-resolution remote sensing images, especially for images covering earthquake disasters, feature-based image registration methods need a more helpful feature descriptor to improve the accuracy. However, traditional image registration methods that only use local features at low levels have difficulty representing the features of the matching points. To improve the accuracy of matching features for multisource high-resolution remote sensing images, an image registration method based on a deep residual network (ResNet) and scale-invariant feature transform (SIFT) was proposed. It used the fusion of SIFT features and ResNet features on the basis of the traditional algorithm to achieve image registration. The proposed method consists of two parts: model construction and training and image registration using a combination of SIFT and ResNet34 features. First, a registration sample set constructed from high-resolution satellite remote sensing images was used to fine-tune the network to obtain the ResNet model. Then, for the image to be registered, the Shi_Tomas algorithm and the combination of SIFT and ResNet features were used for feature extraction to complete the image registration. Considering the difference in image sizes and scenes, five pairs of images were used to conduct experiments to verify the effectiveness of the method in different practical applications. The experimental results showed that the proposed method can achieve higher accuracies and more tie points than traditional feature-based methods.

Highlights

Introduction iationsWith the development of remote sensing technology, multisource remote sensing images, which provide richer information for the same region [1], have been applied in remote sensing tasks such as earthquake disaster monitoring, change detection, and ground target identification
scale-invariant feature transform (SIFT) + ResNet34 and SIFT + ResNet50 represent the proposed method using a combination of SIFT and ResNet34 features and a combination of SIFT and ResNet50 features, respectively
A new registration method based on a deep residual network (ResNet) and SIFT was proposed to improve multisource remote sensing image registration accuracy

Summary

Introduction

With the development of remote sensing technology, multisource remote sensing images, which provide richer information for the same region [1], have been applied in remote sensing tasks such as earthquake disaster monitoring, change detection, and ground target identification. The spatial resolution of remote sensing images is continuously improving, making the details of ground objects more prominent [2]. The size and amount of image data are increasing, which increases the difficulty of multisource high-resolution remote sensing data preprocessing and analysis. As an essential preprocessing step of remote sensing imagery, image registration is a method to map one or more remote sensing images (local) to the target image optimally by using some algorithm and based on some evaluation criteria [3]. In various remote sensing applications, the size of the image, differences between different sensors, and Licensee MDPI, Basel, Switzerland.

Methods

Results

Conclusion