Studying The Edge Detection Of Images Of Plant Roots Based On Gabor Wavelet Theory

While the plant root has many advantages such as water conservation, preventing of cropland desertification, conservation of soil and water, preventing wind and consolidating soil of winter, using vegetation for slope consolidation which plays as the role of slope greening and slope reinforce has been widely used in slope treatment for the recent decades. The research of root has become the main research object among many disciplines such as plant nutrition and plant physiology, root morphology analysis is the most important content of root research. The accurate determination of the basic parameters and architecture parameters of root morphology is the premise and key.Therefore, the research of this paper studies the array to use distribution endoscopic type image means to gain the solid slope in the process of plant growth and root morphology which based on the rhizosphere physical factors image information, on the basis of the achieved solid slope plant root image information make a series of related processing, such as according to the image preprocessing, division of root system and soil particle, edge detection and separately extraction root length, diameter root morphology parameter values. With the research into image preprocessing and image segmentation, this paper proposed Gabor using wavelet transformation to capture the edge character of root images. This algorithm can offset the flaw that the traditional algorithms may miss detecting the edge information, present the contour of the root perfectly and achieve a better result in the later analysis research of roots image. Main job of this paper includes:(1)To study the processing of root image deeply using digital image processing technique. That includes emendation of image geometric aberrance of short focus lens, image enhancement and image of the smooth and root image segmentation of root conformational parameters and etc.(2)Proposed Gabor using wavelet transformation for edge detection, this algorithm applied Gabor transformation into wavelet transform fusion algorithm, and the algorithm is compared with some traditional edge detection algorithms, the experimental results show that since Gabor wavelet transform of edge detection algorithm is based on the description of the characteristics and gray, therefore, it is not sensitive to geometric distortion and brightness changes, easy to observe the clarity of image edge detection, have certain resist the ability of noise. And the details that are hard to test out for classic operators can now be well tested. Experiment results show the feasibility of this algorithm in the root system image edge detection.This thesis is originated from the national natural science funds project that research on the mechanism of vegetation in the slope based on the analysis of the dynamic model of root microstructure.