MATLAB数字图像处理在根系固土中的应用研究
2014-10-21阎梦晴于司杭王培亮王逸飞李明宝
阎梦晴 于司杭 王培亮 王逸飞 李明宝
摘要 在水土防治中采用植物根系固定土壤的措施,引入基于MATLAB的数字图像处理技术,为植物根系固土机制的研究提供根系和土壤等的形态特征参数。对比Sobel、Prewitt、Roberts、LoG、Canny 5种边缘检测器的效果,对比基于梯度的边缘信息改进全局阈值处理方法和Otsu方法进行最佳全局阈值处理的效果,确定Canny边缘检测器和Otsu阈值处理方法相结合的最优算法。该数字图像处理技术可为植物根系固土机制的研究提供可靠的分析手段。
关键词 根系固土;数字图像处理;Canny边缘检测器;Otsu阈值处理
中图分类号 S157.1 文献标识码 A 文章编号 0517-6611(2014)31-11039-04
Study of Soil Reinforcement by Plant Roots Using MATLAB Digital Image Processing
YAN Mengqing, YU Sihang, WANG Peiliang*, LI Mingbao* et al
(School of Civil Engineering, Northeast Forestry University, Harbin, Heilongjiang 150040)
Abstract For soil conservation, adopting the root reinforcement measure and using digital image processing based on MATLAB can provide morphological character parameters of plant roots and soil for the study of mechanism of soil reinforcement by plant roots. After constrasting the effects of five edge detectors including Sobel edge detector, Prewitt edge detector, Roberts edge detector, LoG edge detector and Canny edge detector, and the effects of two methods of threshold processing, including overall threshold processing which improved the edge information and based on gradient and Otsu threshold processing, the Canny edge detector and the Otsu threshold processing are the best algorithm. This Digital Image processing based on MATLAB provides a dependable method in the study of mechanism of soil reinforcement by plant roots.
Key words Soil reinforcement by plant roots; Digital image processing; Canny edge detector; Otsu threshold processing
我國是世界上水土流失最严重的国家之一[1]。水土流失会破坏土壤的肥力,造成土壤干层,淤积水库、阻塞河道、抬高河床,恶化生态环境。虽然自新中国成立以来,我国的水土流失治理已经逐渐成为国建的生态建设重点工程,但是由于我国国土面积大,自然地理条件特殊,以及长期以来对水土资源的过度利用,我国水土流失仍然面积大、分布广,防治任务艰巨[2]。防止水土流失,采用科学、环保的植物固土措施,引入数字图像处理技术研究植物根系固土机制,能够为我国农业水土保持增添新的技术手段,促进我国生态文明建设的发展。
水土流失的防治措施有工程措施和植物措施,其中植物措施不仅可以防治水土流失,还可改善局部的生态环境,节约经济成本,具有生态可持续性。近年来,由于生态环境建设的需要及各学科科学技术的发展,植物根系固定土体机理的研究成为根系研究的焦点[3]。……
