摘要
图像的边缘是图像特征提取与分析理解的基础,其检测质量直接决定后期理解的效果。为了更有效地检测出图像边缘信息,提出了一种基于灰色预测模型的图像边缘检测新方法。该方法向GM(1,1)求解模型的指数中加入了一个调整参数p,通过选择象素周围不同方向的象素数据点以及2次调整参数p的取值对图像进行预测处理,从而得到1幅轮廓增强和1幅未增强的预测图像,将这2幅图像做差,便获得了1幅图像的边缘位置信息。使用改进后的方法对1组图像进行了预测处理,其结果表明,该算法能够有效地检测出图像的边缘信息,并且图像的细节部分也能够清楚地获得,说明是一种有效的图像边缘检测新算法,同时为灰色理论应用于图像边缘检测进行了尝试与探讨。
Aim. Image edge is the foundation of feature extraction, analysis and comprehension; its quality determines the quality of image processing results. In order to detect the image edge more effectively, we put forward a new method based on grey model （GM）. In the full paper, we explain in some detail how to improve the GM and use it to improve the effect of image edge detection. In this abstract, we just add some pertinent remarks to listing the two topics of explanation. The first topic is. GM（1,1） theory. The second topic is： how to improve the GM（1, 1） and apply it to image edge detection. In the second topic, the improved GM （1, 1） contains an added parameter p as can be seen in eq. （6） in the full paper. With the improved GM （1,1） and non-improved GM （1,1）, we obtain two prediction images, enhance the edge of one image and compare it with the edge of another which is not. Finally, we did experiments and analyze their results. The analysis of the experimental results, given in Fig. 3, shows preliminarily that our improved GM can not only effectively detect the information of image edges but also keep their details.
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2008年第5期603-606,共4页
Journal of Northwestern Polytechnical University
基金
国家自然科学基金(30470459)
西北工业大学基础研究基金(W018102)资助
