海面目标雷达图像多尺度轮廓段特征提取方法

Multi-scale contour segment feature extraction method for sea target radar images

为了有效提高舰船雷达图像特征识别度,基于当前灰度投影拟合相关技术,提出舰船雷达图像轮廓特征提取方法。根据雷达感应电磁波输入信号排列,通过预设灰度级和拟合高度,根据图像三原色组成比例,获取灰度图像,利用二维分割法,根据灰度级转换特征,将一维灰度图分割为二维灰度图,通过模糊积累特征识别方法,设置二维灰度图的特征函数,求取隶属函数,建立模糊特征集,实现雷达图像轮廓特征提取。实验数据表明,应用该设计特征提取方法,图像灰度回波提高了32%,灰度纵波提高了29%,图像特征识别精度具有明显提高。

in order to effectively improve the recognition degree of ship radar image features,a method of extracting ship radar image contour features was proposed based on the current gray projection fitting technology.Based on column,radar induction electromagnetic wave input signal through the preset grayscale and fitting height,according to the image color composition proportion,get gray image,using the method of two-dimensional segmentation,according to the characters of gray level transformation,one-dimensional gray image segmentation into two-dimensional grayscale,through the fuzzy accumulation feature recognition method,set the characteristic function of two-dimensional gray image,calculate the membership function,the feature set and fuzzy,realize the radar image contour feature extraction.The experimental results show that the method can improve the recognition accuracy of image features by 32%and 29%,respectively.