一种用于桥梁识别的水域分割提取方法

提出一种新的水域分割提取算法.采用灰度图像减去加权梯度图像的方法拉大水域与田地的差异,利用水域平均灰度高于图像平均灰度的特点,对常规OSTU阈值分割算法进行改进,图像分割后通过水域标记排除小面积干扰区城,通过寻找组成河流的水域方法排除大面积田地干扰.仿真实验结果证明该算法可实现水域的合理有效分割提取.

A Water Area Segmentation Method for Bridge Recognition

A new algorithm was proposed for water area segmentation and extraction.In order to enlarge the difference between the water area and the field,the method of subtracting weighted gradient image from the gray image was adopted.The characteristic that the average gray intensity of water area is higher than that of image was used.The traditional OSTU threshold segmentation algorithm was improved.After image segmentation and water area labeling,small false areas were removed.Besides,the large-area field was also eliminated by searching water areas that making up a river.The results of experimentation prove that the method is valid and practical for segmentation and extraction of water area.