正态分布转换算法在声纳图像处理中的应用

The Application of Normal Distribution Transformation Algorithm in Sonar Image Processing

根据声纳图像匹配处理的实时性需求,研究了正态分布转换算法。该方法基于对图像中目标点的概率分布进行匹配,不需要提取图像的几何特征或灰度特征,可有效缩短匹配时间。通过对均值滤波法、中值滤波法、小波去噪法、形态学平滑法4种图像去噪和迭代阈值法、最大熵法、一维Otsu法和二维Otsu法4种分割算法,进行仿真并对比,选出符合实时性处理的方法,即中值滤波法和一维Otsui法。基于美国德州仪器公司的数字信号处理芯片,将图像去噪、分割和匹配算法在硬件上实现,硬件实现结果和仿真结果一致,实现了声纳图像匹配的实时处理。

For the requirements of the real-time processing of sonar image matching processing,the normal distributions transform（ NDT） algorithm and its application in the sonar image processing are studied. The NDT algorithm is based on probability model. The coordinate of target point is calculated instead of the gray value of image,which speeds up the calculation speed of the NDT algorithm. Several methods of image denoising and segmentation are compared to give the results and analysis of each method. Based on TI DSP chip,the NDT image matching algorithm is achieved.