一种基于图像纹理特征的波浪检测方法

针对无人水面艇视觉系统检测波浪问题,基于图像纹理特征,利用灰度共生矩阵的独立特征量,提出一种识别水面波浪等级的特征融合阈值法。该方法对波浪图像进行灰度级调整,计算灰度共生矩阵,提取其四个独立特征量,然后根据分析结果确定特征量权值并计算不同波浪等级的阈值,根据得到的阈值来检测波浪等级。实验表明,该方法适用于无人水面艇对周围波浪环境的检测,具有受光照条件影响较小的优点。

Water wave detection based on texture characteristics

To solve the problem of water wave detection for unmanned surface vehicle (USV), this paper proposed a method based on independent features of gray level co-occurrence matrix(GLCM) to identify the scale of wave. Firstly, this method adjusted gray-scale of wave images, calculated GLCM and extracted its four independent features. Then determined the weight values of features and obtained the thresholds of different wave scale. Finally, it judged the scale of wave based on the thre-sholds. Experiments show that the method is less affected by light conditions and USV can effectively detect the scale of water surface wave.