改进Canny算子在水面目标边缘检测中的研究

Research on improved Canny operator in edge detection of the target on water surface

基于边缘特征的提取方法被广泛应用于水面目标识别当中。针对传统方法抗噪性较差且易误判和漏判的缺点,文章对Canny算法进行了改进,基于Canny边缘检测思想,采用双边滤波,以及通过Sobel算子计算梯度幅值方向,并在梯度非极大值抑制过程中,将极大值点作为候选边缘点,将极大值二次刷选,保留8邻域内候选边缘点不相邻的非极大值点,最后利用Otsu计算图像高低阈值。结果证明,该方法能更准确的检测边缘信息,且具备较强的自适应性。

The edge feature based extraction method is widely used in water surface target recognition.Aiming at the disadvantages of traditional methods such as poor noise resistance and easy misjudgment and missed judgment,this paper makes the Canny algorithm improved.By using bilateral filtering,and using Sobel operator to calculate the gradient amplitude direction,and in the gradient non-maximum value suppression process,the maximum value point is used as the candidate edge point,and the maximum value is twice selected,leaving 8 neighbors.The non-maximal points of the candidate edge points in the domain are not adjacent,and finally the Otsu is used to calculate the image high and low thresholds.The result proves that the method can detect edge information more accurately and has strong adaptability.