基于小波变换的图像边缘检测技术

【作者】 孙琰；

【导师】 任章；

【作者基本信息】 西北工业大学 ， 控制理论与控制工程， 2004， 硕士

【摘要】 现代战争正在向信息化发展，作为典型信息化兵器的精确制导武器已经被广泛的应用于战争。成像制导技术是当代精确制导技术的发展主流和方向。在成像制导技术中，首先要进行目标图像的目标提取和识别。在目标图像的目标提取和识别中最基本的图像处理技术就是图像边缘检测技术。 本文以基于小波变换的边缘检测技术为基础，在小波多分辨分析的框架下构造基于边缘检测的两类B样条单正交小波基函数，并改进现有的基于小波变换的边缘检测算法。 主要工作如下： (1)深入研究图像边缘检测技术和小波变换理论。 (2)在分析Canny边缘检测算法和LoG边缘检测算法的基础上，针对高斯滤波器存在过度光滑图像和丢失缓变边缘的问题，采用基于小波变换的边缘检测算法。 (3)根据边缘检测的评价准则，参照最佳边缘滤波器的设计要求，确定用于边缘检测的小波基函数的一般准则，并在此基础上构造两类B样条小波函数。采用自适应平滑滤波锐化图像边缘，再进行小波边缘检测算法，提出改进的基于B样条小波变换的边缘检测算法。 (4)提出基于改进的B样条小波变换的自适应阈值图像边缘检测算法。更多还原

【Abstract】 With further requirement of the modern war, precision guided weapon has been widely used. Imaging-based guided technology is the important development direction of precision guided technology. The fundamental image processing technology of target extract and recognition is image edge detection technology.By means of Mallat’ s multiscale analysis, we construct two type of arbitrary order cardinal B-spline wavelet. In order to avoid the loss of weak edge, take adaptive filter to sharp image edge. This is the improved image edge detection based on wavelet transform.The important work of this paper:1. Deeply study image detection technology and wavelet analysis theory.2. Based on studying Canny operator and LoG operator, we take wavelet edge detection in allusion to smoothing image extremely and losing weak edge existed in Gaussian filter.3. According to the criterion of edge detection and consulting the design-aim of optimal edge-filter, this paper discussed the general rules of selecting the mother function of wavelet used in edge detection, construct two type B-spline wavelet function. Though avoiding noise disturbance, this method will lose some weak edge. So we take adaptive filter to sharp image edge in order to avoid losing image weak edge.4. Adaptive thresholds edge detection for image based on wavelet transform was presented to avoid losing weak edge.更多还原