Gabor滤波与数学形态学结合下激光图像亚像素边缘检测

Laser image sub-pixel edge detection combined with Gabor filter and mathematical morphology

针对当前图像亚像素识别定位易受到噪声干扰,导致边缘检测完整性较差的问题,提出基于Gabor滤波与数学形态学的激光图像亚像素边缘检测方法。采用指定不同方向与尺度Gabor滤波器和图像卷积,对卷积激光图像实行规格化操作,利用取极小值与平均值实现激光图像Gabor滤波器结果自适应性融合,获取增强后的图像。依据图像处理,根据图像灰度与空间结构特征,通过数学形态学实现边缘点粗定位,在图像像素级上得到边缘点坐标与梯度方向。依据边缘点向量与参考阈值,通过Zernike矩实现边缘点亚像素边缘精定位,完成激光图像亚像素边缘检测。实验结果表明,所提方法具备较高的检测精度和效率,可实践性能较强。

In view of the problem that the current image sub-pixel recognition and location are easy to be disturbed by noise,which leads to poor integrity of edge detection,a sub-pixel edge detection method of laser image based on Gabor filter and mathematical morphology is proposed.The convolution laser image is normalized by specifying Gabor filters and image convolution in different directions and scale.The laser image Gabor filter results are adaptively fused by taking the minimum value and the average value to obtain the enhanced image.According to image processing,gray level and spatial structure characteristics of the image,the edge point coarse positioning is realized by mathematical morphology,and the edge point coordinates and the gradient direction are obtained at image pixel level.According to the edge point vector and the reference threshold,the sub-pixel edge of edge point is finely positioned by Zernike moment,and the sub-pixel edge detection of laser image is completed.The experimental results show that the proposed method has higher detection accuracy and efficiency,and has better performance.