[1]胡维炜,张武,刘连忠,等.利用图像处理技术计算大豆叶片相对病斑面积[J].江苏农业学报,2016,(04):774-779.[doi:10.3969/j.issn.100-4440.2016.04.010]
 HU Wei-wei,ZHANG Wu,LIU Lian-zhong,et al.Measurement of relative lesion area on soybean leaf using image processing technology[J].,2016,(04):774-779.[doi:10.3969/j.issn.100-4440.2016.04.010]
点击复制

利用图像处理技术计算大豆叶片相对病斑面积()
分享到:

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2016年04期
页码:
774-779
栏目:
植物保护
出版日期:
2016-10-31

文章信息/Info

Title:
Measurement of relative lesion area on soybean leaf using image processing technology
作者:
胡维炜1张武12刘连忠12蔡芮莹1朱小倩1
1.安徽农业大学信息与计算机学院,安徽 合肥 230036;2.农业部农业物联网技术集成与应用重点实验室,安徽 合肥 230036
Author(s):
HU Wei-wei1ZHANG Wu12LIU Lian-zhong12CAI Rui-ying1ZHU Xiao-qian1
1.College of Information and Computer Science, Anhui Agricultural University, Hefei 230036, China;2. Key Laboratory of Technology Integration and Application in Agricultural Internet of Things, Ministry of Agriculture, Hefei 230036, China
关键词:
相对病斑面积病斑区域图像分割 K-means聚类
Keywords:
relative lesion areadiseased spotimage segmentationK-means clustering
分类号:
TP391
DOI:
10.3969/j.issn.100-4440.2016.04.010
文献标志码:
A
摘要:
为了定量化评估大豆作物的病害程度,提出一种基于图像处理测量大豆叶片相对病斑面积的方法。通过HSV、L*a*b*特征空间聚类法逐步分离目标叶片和病斑区域,并采用区域填充法减少叶面水珠、菌碟和菌丝等干扰,最后根据目标叶片和病斑区域像素数计算相对病斑面积。利用该方法分别对培养皿和背光板叶片图像进行试验,结果表明,该方法处理一幅800×800大小的彩色图像只需约20 s,叶片病斑区域与健康区域分割准确度达97%以上,相对病斑面积计算准确度达99%以上;与传统阈值分割方法相比,该方法能有效分割目标叶片与病斑区域,快速准确地计算相对病斑面积。
Abstract:
To quickly and accurately estimate the disease level of soybean leaf, a measurement based on image processing technology was proposed for relative lesion area (RLA) on soybean leaf. Firstly, HSV and L*a*b* color models were employed to deal with the leaf images of the dishes and blacklights, and K-means clustering algorithm was adopted to split the images. Then, area filling was applied to decrease the interference of water drops and hyphae. Finally, RLA was measured according to the pixels of diseased spots and leafs. The experiments on multiple soybean leaves showed that the time for processing a 800×800 pixel color image only took 20 seconds, and the segmentation precision and accuracy of RLA calculation reached 97% and 99% respectively. The method is effective and accurate in calculating RLA when compared with OTSU thresholding method and photoshop method.

参考文献/References:

[1]郑燕,吴为人.利用稻米垩白度分析软件测量叶片相对病斑面积[J].中国农业科学,2008,41(10):3405-3409.
[2]周丽娜.基于叶绿素荧光光谱分析的稻米瘟病害识别与预警[D].长春:吉林大学,2014.
[3]陈丁山.稻米外观品质性状快速检测系统的研究与应用[D].长沙:湖南农业大学,2011.
[4]韩殿元,黄心渊,付慧.基于彩色通道相似性图像分割方法的植物叶面积计算[J].农业工程学报,2012,28(6):179-183.
[5]张善文,张云龙,尚怡君. 1 种基于Otsu 算法的植物病害叶片图像分割方法[J]. 江苏农业科学,2014,42(4): 337-339.
[6]何应德.基于图像分析的树木叶片像素面积计算[D].北京:北京林业大学,2011.
[7]栗娜,李萍,张善文. 基于改进遗传算法的作物叶片病斑分割算法[J]. 江苏农业科学,2014,42(7): 14-142.
[8]郭文川,周超超,韩文霆.基于Android 手机的植物叶片面积快速无损测量系统[J].农业机械学报,2014,45(1):275-280.
[9]刘丽娟,刘仲鹏. 北方旱育稀植水稻病害图像识别预处理研究[J]. 江苏农业科学,2014,42(1): 92-94.
[10] 夏营威,徐大勇,堵劲松,等.基于机器视觉的烟叶面积在线测量[J].农业机械学报,2012,43(10):167-173.
[11] 龚爱平,吴武豪,裘正军,等.基于Android系统手机的叶面积测量方法[J].农业机械学报,2013,44(9):203-208.
[12] JI Z, XIA Y, CHEN Q, et al. Fuzzy c-means clustering with weighted image patch for image segmentation [J]. Applied Soft Computing, 2012, 12(6):1659-1667.
[13] JIN R, KOU C, LIU R, et al. A color image segmentation method based on improved k-means clustering algorithm [C]//Zhong Z. Proceedings of the International Conference on Information Engineering and Applications. London:Springer-Verlag,2013:499-505.
[14] 朱征宇,王丽敏.基于聚类和局部区域的彩色图像分割方法[J].计算机工程与设计,2015,36(1):201-205.
[15] SELIM S Z, ISMAIL M A.K-means-type algorithm[J].IEEE Trans Pattern Anal Mach Intell, 1994, 6(1):81-87.
[16] 齐文斌,毛秉毅.主色调颜色特征的图像检索与分类[J].计算机工程与应用,2011,47(24):191-192.
[17] 陈昌涛,仇国庆,杨平,等.空间色彩分割在快速车牌定位中的应用[J].计算机应用研究, 2010, 27(8):3191-3193.
[18] 陈丽雪, 陈昭炯.基于Lab空间的图像检索算法[J].计算机工程, 2008,34(13):224-226.
[19] 张强,王正林.精通MATLAB 图像处理[M].北京:电子工业出版社,2009:272-279.
[20] 李冠林,马占鸿,黄冲,等.基于K_means硬聚类算法的葡萄病害彩色图像分割方法[J].农业工程学报,2010,26(S2):32-37.
[21] SEZGIN M, SANKUR B. Survey over image thresholding techniques and quantitative performance evaluation[J].Journal of Electronic Imaging, 2004, 13(1): 146-165.

备注/Memo

备注/Memo:
收稿日期:2015-12-28基金项目:农业部引进国际先进科学技术“948”项目(2015-Z44);农业部农业物联网技术集成与应用重点实验室开放基金项目(2016KL05);安徽农业大学引进与稳定人才科研资助项目(wd2015-05)作者简介:胡维炜(1993-),女,安徽马鞍山人,硕士研究生,研究方向为图像处理、计算机应用技术。通讯作者:张武,(E-mail)zhangwu@edu.cn.com
更新日期/Last Update: 2016-11-01