激光主动成像图像边缘检测算法研究

Detection algorithm of image edge for laser active imaging

针对传统Canny边缘检测算法存在的不足,结合激光主动成像末制导的实际特点,提出了一种改进的Canny边缘检测算法。该算法在Canny边缘检测算法的基础上,采用同态滤波和提升小波变换级联的方法代替传统的高斯滤波器;采用3×3邻域的权值梯度计算方法代替原有的2×2邻域差分运算;利用类间方差最大值和梯度幅值直方图来进行高低阈值的自适应取值,使算法可根据图像自身特点选择最合适的阈值。实验证明:该算法提升了传统Canny算子的抗干扰能力和鲁棒性,能够较好地保存边缘信息,准确检测出激光主动成像图像边缘,具有较强的自适应性。

For the shortcoming of the traditional Canny algorithm,an improved Canny edge detection algorithm is proposed combining with the characteristics of laser active imaging. Based on Canny edge detection algorithm, the homo-morphic filtering and lifting wavelet transform are adopted to replace the traditional Gaussian filter;a weight gradient method of 3 ×3 adjacent regions is used to replace difference method of 2× 2 adjacent regions. Finally,Otsu algorithm and gradient histogram are used to get appropriate thresholds adaptively based on the features of the image. Experimental results show that the algorithm can improve the anti-interference ability and robustness of traditional Canny operator and keep the edge information, and it can detect the image edge of laser active imaging, so it has strong adaptability.