[1]李建军,史春梅,单琪凯,等.基于SURF算法的大豆灰斑病视觉识别系统[J].大豆科学,2019,38(01):90-96.[doi:10.11861/j.issn.1000-9841.2019.01.0090]
　LI Jian-jun,SHI Chun-mei,SHAN Qi-kai,et al.Visual Identification System of Soybean Frogeye Leaf Spot Based on SURF Feature Extraction[J].Soybean Science,2019,38(01):90-96.[doi:10.11861/j.issn.1000-9841.2019.01.0090]
点击复制

基于SURF算法的大豆灰斑病视觉识别系统

《大豆科学》[ISSN:1000-9841/CN:23-1227/S] 卷: 第38卷 期数: 2019年01期 页码: 90-96 栏目: 出版日期: 2019-01-20
Title:
Visual Identification System of Soybean Frogeye Leaf Spot Based on SURF Feature Extraction
作者:
李建军1; 史春梅2; 单琪凯1; 华秀萍1; 孟庆祥1; 王岩1; 王丽丽1; 姜永成1
（1.佳木斯大学 智能检测与控制研究所，黑龙江 佳木斯 154007； 2.黑龙江省水利学校，黑龙江 大庆 163000）
Author(s):
LI Jian-jun1;  SHI Chun-mei2;  SHAN Qi-kai1;  HUA Xiu-ping1;  MENG Qing-xiang1;  WANG Yan1;  WANG Li-li1;  JIANG Yong-cheng1
(1.Institute of Intelligent Detection and Control, Jiamusi University, Jiamusi 154007, China; 2.Heilongjiang Water Conservancy School, Daqing 163000, China)
关键词:
机器视觉; 大豆灰斑病; SURF; SIFT; Open CV
Keywords:
Machine vision;  Soybean gray spot;  SURF;  SIFT;  Open CV
DOI:
10.11861/j.issn.1000-9841.2019.01.0090
文献标志码:
A
摘要:
机器视觉技术是农田信息采集系统的关键技术之一，在精细农业中有广泛应用，农作物病害部位的精准识别作为精准施药的前提和关键，其识别精准度对病害防治效果影响巨大，当前农田信息采集系统识别精度差的问题急需解决。本研究结合机器视觉与计算机图像处理技术，以Visual Studio 2010为开发平台，搭建大豆灰斑病计算机视觉识别系统。由计算机摄像头采集大豆叶片图像，通过对彩色图像灰度化预处理，分别对比Open CV（Open Source Computer Vision Library）计算机视觉开源库中两个图像特征检测识别方法—SURF（Speeded Up Robust Features）法和SIFT（Scale-Invariant Feature Transform）法，对图像灰斑特征点进行检测，两种算法在输出帧率上差别明显，SIFT算法输出帧率为0.3~0.5 fps，SURF算法输出帧率为0.6~0.9 fps，考虑设备性能和灰斑识别的精准性，最终选用SURF算法；建立图像像素点海森矩阵，经高斯滤波，利用非极大值抑制法确定特征点，再由立体空间差值法定位极值点，根据哈尔小波响应值选取特征点主方向，构造SURF特征点描述算子以提取特征点，编写相关程序代码，分别对黑农44分枝期和结荚期大豆叶片灰斑进行检测。结果显示：分枝期叶片纹理少，特征点少，检测效果好，检测正确率97.28%，耗时0.97 s，结荚期叶片纹理增多，绘制特征点较多，检测正确率89.49%，耗时1.19 s，基本满足大豆灰斑病识别系统功能需求。通过田间试验，利用FLANN算法对分枝期视频图像进行特征点提取，实现对视频图像帧检测并匹配特征点的目的，检测率为90.7%，匹配率93.8%，该大豆灰斑病视觉识别系统的构建能够为下一步精准施药及相关农田信息精准采集系统设计提供思路与参考。
Abstract:
Machine vision technology is one of the key technologies of farmland information collection system, which is widely used in precision agriculture. As premises and keys of precision pesticide application, recognition accuracy of crop disease site has strong influences on the diseases control. The badness of recognition accuracy is urgent to be solved. Combining machine vision and computer image technology, visual studio 2010 is used as the experimental platform to build a computer vision identification system of soybean gray spot. The image of soybean leaf was collected by computer camera, and the color image was preprocessed by graying. Two image feature detection methods from Open CV (Open Source Computer Vision Library)-SURF(Speeded Up Robust Features) algorithm and SIFT(Scale-Invariant Feature Transform) algorithm were used to detect gray spot feature points. There were obvious differences between two methods in output frame rate: The range of SIFT algorithm output frame rate was between 0.3 and 0.5 fps, and SURF algorithm was between 0.6 and 0.9 fps. Considering the efficiency and equipment performance, SURF algorithm was selected. The Hessian Matrix of pixel points in the image was established. The feature points were determined by using the Gauss filter and the non-maximum suppression, and then the extreme points were located by the spatial interpolation. The main direction of the feature points was selected according to the Hal wavelet. The descriptor of SURF feature point was constructed to extract feature points, and the program codes were compiled to detect the gray spot in soybean leaves of Heinong 44 in branching stage and pod bearing stages respectively. The results showed that the leaf texture and the feature points in branching stage were few, detection effect was good, the correct rate of detection was 97.3%, and detected time was 0.97 s. The leaf texture and the feature points increase in pod bearing stages, the correct rate of detection was 89.49% and detected time was 1.19 s, which basically meets the functional requirements of soybean gray spot recognition system. Through field trials and using FLANN algorithm to extract feature points of video images, detecting of video image frame rate and matching of feature points were achieved. The detection rate was 90.7% and the matching rate was 93.8%, which provides ideas and references for the next research on precision pesticide application and the design of other farmland information collection system.

参考文献/References:

［1］赵忠志. 大豆灰斑病预防措施［J］.现代农业研究, 2018(1): 117-118. (Zhao Z Z . Prevention measures for soybean gray spot disease［J］. Modern Agricultural Research, 2018(1): 117-118.)
［2］李沐慧. 东北地区大豆田病害调查研究与鉴定［D］. 沈阳：沈阳农业大学, 2016. (Li M H. Investigation and identification of soybean diseases in northeast China［D］. Shenyang: Shenyang Agricultural University, 2016.)
［3］马淑梅, 韩新华, 邵红涛. 大豆主要病害多抗性资源筛选鉴定［J］. 中国农学通报, 2014, 30(27): 58-65. (Ma S M, Han X H, Shao H T. Evaluation of selected soybean cultivars and breeding lines for multi-resistance to major pathogens［J］. Chinese Agricultural Science Bulletin, 2014, 30(27): 58-65.)
［4］柳美玲. 北方大豆灰斑病发病规律与防治［J］. 农民致富之友, 2017(9): 102. (Liu M L. Regularity and prevention of soybean grey spot disease in north China［J］. Friends of Farmers Becoming Rich, 2017(9): 102.)
［5］李博宇. 黑龙江省大豆品种对灰斑病抗病性丧失原因分析［D］. 哈尔滨: 黑龙江大学, 2014. (Li B Y. Analysis on the reason of losing resistance to gray spot disease of soybean varieties in Heilongjiang province［D］. Harbin: Heilongjiang University, 2014.)
［6］董志敏, 厉志, 刘佳, 等.大豆抗灰斑病研究进展［J］. 作物杂志, 2017(3): 1-5. (Dong Z M, Li Z, Liu J, et al. Progress on resistance to frogeye leaf spot in soybean［J］. Crops, 2017(3): 1-5.)
［7］程伟, 常芳国, 赵团结, 等. 大豆灰斑病发生特点及抗病遗传育种研究进展［J］. 安徽农业科学, 2015, 43(30): 22-25,50. (Cheng W, Chang F G, Zhao T J, et al. Advance in the disease occurrence and resistance breeding of frogeye leaf spot in soybean［J］. Journal of Anhui Agricultural Sciences, 2015, 43(30): 22-25, 50.)
［8］王风云, 封文杰, 郑纪业, 等. 基于机器视觉的双孢蘑菇在线自动分级系统设计与试验［J］. 农业工程学报, 2018, 34(7): 256-263. (Wang F Y, Feng W J, Zheng J Y, et al. Design and experiment of automatic sorting and grading system based on machine vision for white agar［J］. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(7): 256-263.)
［9］张志斌, 赵帅领, 罗锡文, 等. 基于SURF算法的绿色作物特征提取与图像匹配方法［J］. 农业工程学报, 2015, 31(14): 172-178. (Zhang Z B, Zhao S L, Luo X W, et al. Matching method of green crops based on SURF feature extraction［J］. Transactions of the Chinese Society of Agricultural Engineering, 2015, 31(14): 172-178.)
［10］张帆, 李绍明, 刘哲, 等. 基于机器视觉的玉米异常果穗筛分方法［J］. 农业机械学报, 2015, 46(S1): 45-49. (Zhang F, Li S M, Liu Z, et al. Screening method of abnormal corn ears based on machine vision［J］. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(S1): 45-49.)
［11］陈兵旗, 吴召恒, 李红业, 等. 机器视觉技术的农业应用研究进展［J］. 科技导报, 2018, 36(11): 54-65. (Chen B Q, Wu S H, Li H Y, et al. Research of machine vision technology in agricultural application: Today and the future［J］. Science & Technology Review, 2018, 36(11): 54-65.)
［12］刁智华, 王会丹, 魏伟. 机器视觉在农业生产中的应用研究［J］. 农机化研究, 2014, 36(3): 206-211. (Diao Z H, Wang H D, Wei W. Summary of research on machine vision application in agricultural production［J］. Journal of Agricultural Mechanization Research, 2014, 36(3): 206-211.)
［13］赵世达, 马力. 机器视觉在农业生产中的运用与技术［J］. 电子技术与软件工程, 2018(9): 55. (Zhao S D, Ma L. Application and technology of machine vision in agricultural production［J］. Electronic Technology & Software Engineering, 2018(9): 55.)
［14］王风云, 郑纪业, 唐研, 等. 机器视觉在我国农业中的应用研究进展分析［J］. 山东农业科学, 2016, 48(4): 139-144. (Wang F Y, Zheng J Y, Tang Y, et al. Analysis on application and research progress of machine vision in agriculture in China［J］. Shandong Agricultural Sciences, 2016, 48(4): 139-144.)
［15］Hong P T T, Hai T T T, Lan L T, et al. Comparative study on vision based rice seed varieties identification［C］//Seventh International Conference on Knowledge and Systems Engineering. New Jersey: IEEE Computer Society, 2015: 377-382.
［16］Hasankhani R, Navid H. Potato sorting based on size and color in machine vision system［J］. Journal of Agricultural Science, 2012, 4(5): 235-244.
［17］陈胜勇, 刘盛. 基于Open CV的计算机视觉技术实现［M］. 北京：科学出版社, 2008. (Chen S Y, Liu S. The realization of computer vision technology in Open CV［M］. Beijing: Science Press, 2008.)

相似文献/References:

[1]尹俊琦,王 楠,周 莹,等.两个大豆品种转hrpZPsta基因后代对大豆灰斑病的抗性分析[J].大豆科学,2013,32(02):238.[doi:10.3969/j.issn.1000-9841.2013.02.022]
　YIN Jun-qi,WANG Nan,ZHOU Ying,et al.Analysis on Resistance to Frogeye Leaf Spot between Two Species of Transgenic Soybean Progenies with HrpZpsta Gene[J].Soybean Science,2013,32(01):238.[doi:10.3969/j.issn.1000-9841.2013.02.022]
[2]马淑梅.2006~2010年黑龙江省大豆灰斑病菌生理小种监测及部分主栽品种抗性鉴定[J].大豆科学,2011,30(03):450.[doi:10.11861/j.issn.1000-9841.2011.03.0450]
　MA Shu-mei.Monitoring of Physiological Races of Cercospora sojina in Heilongjiang Province from 2006 to 2010 and Resistance Identification of Partial Main Cultivars[J].Soybean Science,2011,30(01):450.[doi:10.11861/j.issn.1000-9841.2011.03.0450]
[3]顾鑫,丁俊杰,杨晓贺,等.2008~2009年黑龙江省大豆灰斑病生理小种的监测[J].大豆科学,2010,29(03):540.[doi:10.11861/j.issn.1000-9841.2010.03.0540]
　GU Xin,DING Jun-jie,YANG Xiao-he,et al.Monitoring of Physiological Race of Cercospora sojina in Heilongjiang Province between 2008 and 2009[J].Soybean Science,2010,29(01):540.[doi:10.11861/j.issn.1000-9841.2010.03.0540]
[4]丁俊杰.影响大豆灰斑病主要气象因子的通径分析[J].大豆科学,2010,29(04):727.[doi:10.11861/j.issn.1000-9841.2010.04.0727]
　DING Jun-jie.Path Analysis on Main Meteorological Factors Affecting Soybean Frogeye Leaf Spot[J].Soybean Science,2010,29(01):727.[doi:10.11861/j.issn.1000-9841.2010.04.0727]
[5]丁俊杰,文景芝,胡国华,等.黑龙江省大豆灰斑病生理小种监测及主栽品种抗性分析[J].大豆科学,2009,28(01):178.[doi:10.11861/j.issn.1000-9841.2009.01.0178]
　DING Jun-jie,WEN Jing-zhi,HU Guo-hua,et al.Monitoring of Physiological Race of Soybean Frogeye Spot and Analysis of Variety Resistance in Heilongjiang Province[J].Soybean Science,2009,28(01):178.[doi:10.11861/j.issn.1000-9841.2009.01.0178]
[6]姜翠兰,胡国华,丁俊杰,等.气象因子对黑龙江省大豆灰斑病发生的影响[J].大豆科学,2009,28(02):276.[doi:10.11861/j.issn.1000-9841.2009.02.0276]
　JIANG Cui-lan,HU Guo-hua,DING Jun-jie,et al.Effect of Meteorological Factors on Frogeye Leafspot of Soybean in Heilongjiang Province[J].Soybean Science,2009,28(01):276.[doi:10.11861/j.issn.1000-9841.2009.02.0276]
[7]刘亚光,赵滨,马超.水杨酸和壳聚糖诱导大豆对灰斑病的抗性[J].大豆科学,2008,27(02):298.[doi:10.11861/j.issn.1000-9841.2008.02.0298]
　LIU Ya-guang,ZHAO Bin,MA Chao.Induction effect of Salicylic acid and Chitosan on Frogeye Leaf Spot in Soybean[J].Soybean Science,2008,27(01):298.[doi:10.11861/j.issn.1000-9841.2008.02.0298]
[8]王德亮杨丹霞姜玉久阎晓东井旭源.大豆不同株龄及不同的叶龄、荚龄对灰斑病的反应[J].大豆科学,2000,19(02):126.[doi:10.11861/j.issn.1000-9841.2000.02.0126]
　Wang Delia ngYang Danx iaJiang YujouYan Xiaodo ngJing Xuyuan.THE REACTION OF VARIOUS AGE OF NODES LEAVESAND PODS OF SOYBEAN TO FROGEYE LEAF SPOT[J].Soybean Science,2000,19(01):126.[doi:10.11861/j.issn.1000-9841.2000.02.0126]
[9]曹越平杨庆凯.大豆灰斑病抗感标准划分的研究[J].大豆科学,2002,21(02):113.[doi:10.11861/j.issn.1000-9841.2002.02.0113]
　Cao YuepingYang Qing kai.STUDY ON RESISTANCE STANDARD OF SOYBEAN PLANTS TOCERCOSPOR A SOJIN A Hara[J].Soybean Science,2002,21(01):113.[doi:10.11861/j.issn.1000-9841.2002.02.0113]
[10]刘亚光李海英杨庆凯.大豆品种的抗病性与叶片内苯丙氨酸解氨酶活性关系的研究[J].大豆科学,2002,21(03):195.[doi:10.11861/j.issn.1000-9841.2002.03.0195]
　Liu Yag uangLi HaiyingYang Qingkai.STUDY ON THE RELATIONSHIP BETWEEN RESISTANCE OF SOYBEAN AND ACTIVITYOF PAL IN LEAVES OF SOYBEAN INFECTED BY CERCOSPORA SOJINA HARA[J].Soybean Science,2002,21(01):195.[doi:10.11861/j.issn.1000-9841.2002.03.0195]

备注/Memo

收稿日期：2018-09-11

基金项目：教育部留学回国人员科研启动基金(020450)；黑龙江省教育厅研究生教学改革项目（JGXM_HLJ_2016094）；十三五国家重点专项(2016yfd0701704-02)；佳木斯大学校长创新创业基金（xzyf2017-06）。
第一作者简介：李建军(1991-)，男，硕士，主要从事农业电气化研究。E-mail：18249242716@139.com。
通讯作者：姜永成(1976-)，男，博士，教授，主要从事智能农业装备研究。E-mail：jiangyongcheng@jmsu.edu.cn。

更新日期/Last Update: 2019-01-22