Building Change Detection For High-Resolution Remote Sensing Data Based On W-Net

With the development of aerospace technology and computer technology,remote sensing earth observation has broken through the boundage of data acquisition,transmission and storage,and has gradually moved towards a new stage of comprehensiveness,automation and intelligence.The spatial resolution of remote sensing data continues to increase,and data types continue to be enriched,providing high-quality and diversified data sources for the research of remote sensing change detection.Building change detection is a research hotspot in the field of change detection,and it is also one of the core issues in earth observation.It has received extensive attention in recent years.Buildings are one of the most dynamic structures in cities,and their changes can reflect the process of urbanization to a large extent.Accurate and effective evaluation of building changes is a powerful means to obtain reliable urban change information,and it is also an urgent need in some fields such as government management,economic construction,and sociological research.High-resolution remote sensing images contain richer and more detailed feature information of ground objects,which brings a new scientific perspective to the study of building change detection.But there are problems such as confusion of image spectral features,low statistical separability of spectral domain,and difficult information extraction.Due to the dual effects of the complexity of the image spectrum and the differences of buildings,building change detection based on high-resolution remote sensing images and multi-source data urgently requires efficient,accurate and intelligent theoretical research and method innovation.In addition,the acceleration of urbanization has put forward new requirements for building change detection.The theoretical method that can simultaneously perform 2D and 3D change detection has high research value.Based on the theory of deep learning,this paper designs a new 2D and 3D change detection method for buildings that takes into account multi-source remote sensing data and data characteristics by constructing a new network model.In this article,based on U-Net,we designed a new type of bilateral end-to-end W network.It can simultaneously input multi-source/multi-feature homogeneous and heterogeneous data,and consider the internal relationship of input data through the squeeze-and-excitation strategy.We named it squeeze-and-excitation W-Net.It has two-sided input and singleoutput,independent weights on both sides can take into account the data on both sides(homogeneous and heterogeneous data),and can be used for change detection tasks in the field of remote sensing.The main contributions of this article are concluded as follows:(1)The proposed squeeze-and-excitation W-Net is a powerful and universal network structure,which can learn the abstract features contained in homogeneous and heterogeneous data through a structured symmetric system.(2)The form of two-sided input not only satisfies the input of multi-source data,but also is suitable for multiple features derived from multi-source data.We innovatively introduce the squeeze-and-excitation module as a strategy for explicit modeling of the interdependence between channels,which makes the network more directional and can recalibrate the feature channels,emphasize essential features,and suppress secondary features.Moreover,the squeeze-and-excitation module is embedded between each convolution operation,which can overcome the insufficiency of the convolution operation that can only take into account the features information in the local receptive field and improve the global reception ability of the network.(3)The idea of multi-source and multi-feature combination as model input integrates information advantages such as spectrum,texture,and structure,which can significantly improve the robustness of the model.For buildings,which present complex spatial patterns,have multi-scale features,and have large differences between individuals,they are more targeted,and the detection accuracy of the model is significantly higher.