针对超宽带探地雷达BP成像算法计算量大、运算速度慢的缺陷,利用对折射点近似求解算法进行优化与处理,提出一种基于信息熵特征的快速BP成像算法。算法以信息熵为统计量来判定目标回波在B-scan数据列上的分布区域,利用目标区域数据参与BP成像。实测数据对比验证结果表明,优化后的BP算法运算速度快,成像精度高。
BP algorithm is widely used in ultra-wideband ground penetrating radar imaging,but it has a drawback with large computation and lower speed.In this paper,through optimizing and processing refraction point approximation algorithm,the rapid BP imaging algorithm based on information entropy characteristics is presented,which determines the target echo in B-scan data columns distribution area by information entropy statistics with the region data participating in BP imaging data.The experimental result shows that the optimized BP algorithm computes fast and the image precision is high.