一种基于方差的自适应火星图像阈值选取算法

在基于光学成像的深空探测自主导航中,图像边缘处理直接影响视线矢量的提取,从而对自主导航的精度产生较大影响。传统的Otsu算法更侧重同区域灰度的均匀性,适用于图像中目标区域和背景区域面积相差不大的情况,因此在自主导航初始阶段,导航精度较低。根据深空探测巡航段拍摄到的火星图像的特点,基于已有火星探测任务的实拍图像,在Otsu算法的基础上,重新设计准则函数,提出了一种基于方差的火星图像阈值自适应选取算法。该算法将灰度值高于待定阈值的区域的方差表示为待定阈值的函数,以函数一阶微分的最大值对应的灰度值作为最优阈值。该方法具有计算量小、图像分割精确的优点。仿真结果表明,相较于传统的Otsu算法,通过该算法得到的图像阈值能够实现更高的视线矢量提取精度及自主导航精度。

An Adaptive Image Threshold Selection Algorithm Based on Variance

In the autonomous navigation of deep space exploration based on optical imaging, image edge processing directly affects the extraction of visual vector, which has a great influence on the accuracy of autonomous navigation. Otsu scheme is one of the most widely used image thresholding techniques. The traditional Otsu algorithm focuses more on the uniformity of the pixels in the same area, and it is suitable for situations where the target area and the background area in the image have little difference. Therefore, in the initial stage of autonomous navigation, navigation accuracy is low. In the new algorithm, according to the characteristics of the Mars image captured in the cruise section, then on the basis of Otsu scheme, an improved algorithm is proposed to redesign the criterion function, the variance of area where the gray value is higher than the undetermined threshold is expressed as a function of it, and the gray value correspond to maximum of this function''s first order differential is taken for the optimal threshold. This method has the advantages of small computation and accurate image segmentation. The simulation results show that compared with Otsu scheme, the image threshold obtained by the improved algorithm can realize higher visual vector extraction accuracy and autonomous navigation accuracy.