论文部分内容阅读
在数字图像处理中,理想的图像边缘检测算法可以根据挖空法再结合边缘跟踪技术设计算法来实现。而在实际工程应用中,所获取的图像中的噪声很多,现有的边缘检测算法检测出的轮廓一般粗细不一,边缘不连续之处过多。为此在检测出图像中物体的轮廓后,还要花大量的时间来处理躁声,在实际应用中很难实现,并且实时性也很差。为此提出一种基于连通区域面积阈值化的实现算法,可以同时实现噪声消除与轮廓提取,以更好地定位图像中的目标物体。
In digital image processing, the ideal image edge detection algorithm can be based on the hollowed out method combined with the edge tracking technology to achieve the design algorithm. However, in actual engineering applications, there are many noises in the obtained image. The contours detected by the existing edge detection algorithms generally vary in thickness and edges are not continuous. Therefore, after detecting the contour of the object in the image, it takes a lot of time to deal with the noise, which is hard to be realized in practice and has poor real-time performance. Therefore, an algorithm based on threshold thresholding of connected regions is proposed, which can achieve both noise cancellation and contour extraction to better locate the target object in the image.