Vehicle Targets Detection Of The Ground Background Synthetic Aperture Radar Image

Synthetic Aperture Radar(SAR)is able to provide clear images throughout the day,and therefore it has been widely used in civil and military fields.However,due to the coherence of the device,there are a large number of speckle noises in the provided image,which increases the difficulty in image processing,image analysis,and image interpretation.Moreover,due to the shooting environment terrain changes and the complex situation,the background conditions of the acquired images are extremely complicated,composed of a large number of redundant data within the image.Therefore,the key problem of current SAR image target detection algorithm optimization is to improve the accuracy and timeliness of related processing algorithms.Due to the coherent imaging mechanism of SAR,coherent speckle noise is inevitably generated.Therefore,this paper firstly studies the coherent speckle distribution in SAR images and explores the applicability of related models.For the SAR images with different resolutions and different scenes,the distribution model which can accurately describe the clutter background is studied to provide support for the constant false alarm rate(CFAR)processing technology.For the detection of land targets,this paper mainly analyzes the images obtained by land target monitoring,analyzes the distribution and characteristics of various types of clutter,and studies the automatic detection algorithm based on the synthetic aperture radar(SAR)image targets of the principle of constant false alarm rate(CFAR).The vehicle targets detection scheme is based on CFAR principle using the clutter model of Rayleigh distribution,Gaussian distribution,and Weibull distribution.The detection performance under different clutter models was analyzed and compared.The experiment results show the feasibility of Rayleigh distribution,Gaussian distribution and Weibull distribution for the target detection in land clutter.In addition,According to the distribution characteristics of the actual SAR data,the corresponding clutter model is reasonably selected.