基于增强算子的污染土雷达图像特征提取仿真  

Feature Extraction Simulation of Contaminated Soil Radar Image Based on Enhancing Operator

作　　者：任燕红 郭幸丽 马丽[2] REN Yan-hong;GUO Xing-li;MA Li(Hebei University of Water Resources and Electric Engineering,Cangzhou Hebei 061001,China;Shenyang University of Chemical Technology,Shenyang Liaoning 110142,China)

机构地区：[1]河北水利电力学院,河北沧州061001 [2]沈阳化工大学,辽宁沈阳110142

出　　处：《计算机仿真》2020年第4期5-8,61,共5页Computer Simulation

基　　金：河北水利电力学院基金项目(SYKJ1904)。

摘　　要：针对雷达图像特征提取效率低、精度不高等问题,提出一种基于增强算子的污染土雷达图像特征提取方法。首先使用多尺度非均匀滤波把含有噪声与不含噪声的像素点灰度值、构成因素及像素加权灰度密度三个特征进行划分,完成雷达图像去噪。其次提出增强算子方法,将图像数据转变至图像的空间域,以此判断污染土雷达图像内是否存在对比度不高区域和细微区域。最后采用灰色关联分析计算超像素之间的相似度,运用MLS增强算子对曲面进行拟合,通过高斯曲率与平均曲率构建出特征判断准则,提高图像特征提取完整性,实现污染土雷达图像特征的精准提取。仿真结果表明,与传统方法相比所提方法计算速度较快,雷达特征提取效率较高,可大幅提升污染土雷达图像特征提取准确率。Due to low efficiency and low precision of radar image feature extraction,a method of feature extraction for radar image based on enhancement operator was presented.Firstly,the multi-scale non-uniform filtering was used to divide the gray value of pixel with and without noise,the constituent elements and the weighted gray density of pixel,and thus to remove the noise from radar image.Secondly,an enhancement operator method was proposed to transform the image data into the spatial domain,so as to judge whether there were low contrast regions and fine regions in the contaminated soil radar image.Finally,the gray relation analysis was used to calculate the similarity between super pixels,and then MLS operator was used to fit the curved surface.Through Gaussian curvature and average curvature,the feature criterion was constructed to improve the integrity of image feature extraction.Thus,the accurate extraction for image features of contaminated soil radar was realized.Simulation results show that,compared with the traditional method,the proposed method has faster computation and higher efficiency of radar feature extraction,which can greatly improve the accuracy of radar image feature extraction.

关 键 词：增强算子 特征提取 移动最小二乘法 曲面拟合 

分 类 号：TP399[自动化与计算机技术—计算机应用技术]