基于局部二值模式的作物叶部病斑检测
发布时间：2018-09-18 20:53
【摘要】：根据作物叶片症状准确、快速检测作物病害是防治和控制作物病害的基础。为准确检测作物叶部病害,在窗阈值中心对称局部二值模式(WTCSLBP)的基础上,提出了一种作物病斑检测方法。首先利用自适应局部二值模式获取正常叶片图像特征并确定病斑判断阈值,然后将待检测叶片图像分割为大小相同的检测窗,并提取同样特征与阈值进行比较,以判断该检测窗是否有病斑。在三种苹果病害叶片图像数据库上的实验结果表明,该方法能够有效检测作物病斑分布特性。与中心对称LBP(CS-LBP)和WTCSLBP相比,该方法具有更少的特征维数和更高的正确识别率。
[Abstract]:According to the accurate symptom of crop leaves, rapid detection of crop diseases is the basis of preventing and controlling crop diseases. In order to accurately detect crop leaf diseases, a method for detecting crop disease spot was proposed based on window threshold centrosymmetric local binary mode (WTCSLBP). Firstly, the adaptive local binary mode is used to obtain the normal leaf image features and determine the threshold of the disease spot, then the image is divided into the same size detection window, and the same feature is extracted and compared with the threshold value. To determine whether the detection window has disease spots. The experimental results on three apple disease leaf image databases show that this method can effectively detect the distribution characteristics of crop disease spots. Compared with centrosymmetric LBP (CS-LBP) and WTCSLBP, this method has less characteristic dimension and higher correct recognition rate.
【作者单位】： 西北大学信息科学与技术学院;西北工业大学电子信息学院;西京学院工程技术学院;
【基金】：国家自然科学基金(No.61473237) 陕西省自然科学基础研究计划(No.2014JM2-6096)
【分类号】：TP391.41
本文编号：2249079
[Abstract]:According to the accurate symptom of crop leaves, rapid detection of crop diseases is the basis of preventing and controlling crop diseases. In order to accurately detect crop leaf diseases, a method for detecting crop disease spot was proposed based on window threshold centrosymmetric local binary mode (WTCSLBP). Firstly, the adaptive local binary mode is used to obtain the normal leaf image features and determine the threshold of the disease spot, then the image is divided into the same size detection window, and the same feature is extracted and compared with the threshold value. To determine whether the detection window has disease spots. The experimental results on three apple disease leaf image databases show that this method can effectively detect the distribution characteristics of crop disease spots. Compared with centrosymmetric LBP (CS-LBP) and WTCSLBP, this method has less characteristic dimension and higher correct recognition rate.
【作者单位】： 西北大学信息科学与技术学院;西北工业大学电子信息学院;西京学院工程技术学院;
【基金】：国家自然科学基金(No.61473237) 陕西省自然科学基础研究计划(No.2014JM2-6096)
【分类号】：TP391.41
【相似文献】
相关硕士学位论文 前2条
1 陈刚;农作物病害图像处理系统[D];沈阳理工大学;2008年
2 聂林红;基于鲁棒局部二值模式的纹理图像分类算法研究[D];天津大学;2016年
，本文编号：2249079
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2249079.html
