本文提出一种首先进行平滑滤波，然后检测模糊边缘的方法来进行复杂景物图像边缘检测。
In the paper, a way of smooth filtering first, then detect edge using fuzzy sets in complex scene image is provided.
景物边缘信息是进行图象分析和识别的重要属性，如何有效地从噪声图象中提取边缘是这些领域中的难点。
Edge is an important attribute for image analysis and recognition, but it is difficult to effectively extract edge from noisy image in the domain.
实验结果表明，改进后的颜色估计模型在图像边缘附近颜色相差较大的自然景物提取中有较好表现。
The improved color estimate model has better behavior in the natural image matting for the image which the edge color is very different.
模糊技术早已应用于图像边缘检测，但是目前的方法对于复杂景物图像的边缘检测，效果不理想。
Fuzzy method has been used to detect edge, but it is not effective on complex scene image.
然而，边缘检测又是图像处理中一个困难的问题，因为实际景物图像中的边缘往往是各种类型的边缘以及它们模糊化后结果的组合，实际图像信号存在着噪声。
Because of the noise in the image signal and all kinds of fuzzy edge combined in the image, the edge detection is one of the difficult problem in the image processing.
通常这种半透明部分出现在前景物体的边缘。
Usually, such translucent portion appears on the fringe of the foreground object.
通常这种半透明部分出现在前景物体的边缘。
Usually, such translucent portion appears on the fringe of the foreground object.
应用推荐