[1]刘妤,刘洒,杨长辉,等.无遮挡重叠柑橘目标分割与重建[J].江苏农业学报,2019,(06):1441-1449.[doi:doi:10.3969/j.issn.1000-4440.2019.06.025]
　LIU Yu,LIU Sa,YANG Chang-hui,et al.Segmentation and reconstruction of overlapped citrus without blocking by branches and leaves[J].,2019,(06):1441-1449.[doi:doi:10.3969/j.issn.1000-4440.2019.06.025]
点击复制

无遮挡重叠柑橘目标分割与重建()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2019年06期
页码:
1441-1449
栏目:
园艺
出版日期:
2019-12-31

文章信息/Info

Title:
Segmentation and reconstruction of overlapped citrus without blocking by branches and leaves
作者:
刘妤1; 2; 刘洒2; 杨长辉1; 2; 3; 王卓2; 熊龙烨2
（1.重庆理工大学汽车零部件先进制造技术教育部重点实验室，重庆400054；2.重庆理工大学机械工程学院，重庆400054；3.西安交通大学机械工程学院，陕西西安710049）
Author(s):
LIU Yu1; 2; LIU Sa2; YANG Chang-hui1; 2; 3; WANG Zhuo2; XIONG Long-ye2
(1.Key Laboratory of Advanced Manufacturing Technology for Automobile Parts of Ministry of Education, Chongqing University of Technology, Chongqing 400054, China;2.College of Mechanical Engineering,Chongqing University of Technology, Chongqing 400054, China;3.College of Mechanical Engineering, Xian Jiaotong University, Xian 710049, China)
关键词:
智能采摘; 重叠分割; 轮廓重建; 凹区域提取; 距离分析
Keywords:
intelligent picking; overlapped segmentation; contour reconstruction; concave area extraction; distance analysis
分类号:
TP391；S666
DOI:
doi:10.3969/j.issn.1000-4440.2019.06.025
文献标志码:
A
摘要:
自然环境下重叠果实的精准识别是智能采摘面临的难题之一。本研究针对无遮挡重叠柑橘，提出了一种基于凹区域简化和距离分析的果实分割与重建方法。该方法提取、分割果实轮廓凹区域，对其进行多边形简化，利用角点检测提取多边形顶点，通过分析各顶点到轮廓凸壳曲线的距离确定轮廓分割点，采用最小二乘圆拟合方法对分割后的轮廓进行重建。结果表明，基于凹区域简化和距离分析的无遮挡重叠柑橘重建轮廓的平均误差为3.12%，不重合度为4.55%，时间为0.291 s，优于RANSAC算法和Hough变换算法，能够满足自然环境下无遮挡重叠果实的智能识别需求。
Abstract:
Accurate identification of overlapped fruits in natural environment is one of the problems that need to be solved in intelligent picking. In this study, a new method of segmentation and reconstruction for overlapped citrus without blocking by branches and leaves was proposed based on concave region simplification and distance analysis. After extracting and segmenting the concave area of the fruit, the polygon was simplified, the vertexes of the polygon were extracted by using the corner point detection, and the dividing point of the contour was determined by analyzing the distance from each vertex to the convex curve of the contour. On this basis, the segmentation points of the contour were determined, and the contour was reconstructed successfully by the least-squares ellipse fitting method. The results showed that the average error, non-coincidence and time of contour reconstruction for this kind of overlapped citrus based on concave region simplification and distance analysis were 3.12%, 4.55% and 0.291 s respectively, which were better than those of RANSAC algorithm as well as Hough transform algorithm. In conclusion, this method can meet the intelligent identification requirements of overlapped fruits without blocking by branches and leaves in natural environment.

参考文献/References:

[1]吕强,蔡健荣,赵杰文,等.自然场景下树上柑橘实时识别技术[J].农业机械学报,2010,41(2):185-188.
[2]RIZON M, YUSRI N A N, KADIR M F A , et al. Determination of mango fruit from binary image using randomized Hough transform[C]. Barcelona: SPIE Press, 2015.
[3]谢忠红, 姬长英, 郭小清,等. 基于凹点搜索的重叠果实定位检测算法研究[J]. 农业机械学报, 2011, 42(12):191-196.
[4]XU Y, IMOU K, KAIZU Y, et al. Two-stage approach for detecting slightly overlapping strawberries using HOG descriptor[J]. Biosystems Engineering, 2013, 115(2):144-153.
[5]XIANG R, JIANG H, YING Y. Recognition of clustered tomatoes based on binocular stereo vision[J]. Computers and Electronics in Agriculture, 2014, 106: 75-90.
[6]LU J, SANG N. Detecting citrus fruits and occlusion recovery under natural illumination conditions[J]. Computers and Electronics in Agriculture, 2015, 110: 121-130.
[7]苗中华, 沈一筹, 王小华,等. 自然环境下重叠果实图像识别算法与试验[J]. 农业机械学报, 2016, 47(6):21-26.
[8]FENG J, WANG S, LIU G, et al. A separating method of adjacent apples based on machine vision and chain code information[C]. Heidelberg: Springer Berlin Heidelberg, 2012.
[9]宋怀波, 张传栋, 潘景朋,等. 基于凸壳的重叠苹果目标分割与重建算法[J]. 农业工程学报, 2013，29(3):163-168.
[10]徐越, 李盈慧, 宋怀波,等. 基于Snake模型与角点检测的双果重叠苹果目标分割方法[J]. 农业工程学报, 2015, 31(1):196-203.
[11]彭辉, 吴鹏飞, 翟瑞芳,等. 基于视差图像的重叠果实图像分割算法[J]. 农业机械学报, 2012, 43(6):167-173.
[12]王丹丹, 徐越, 宋怀波,等. 融合K-means与Ncut算法的无遮挡双重叠苹果目标分割与重建[J]. 农业工程学报, 2015, 31(10):227-234.
[13]王丹丹. 重叠及遮挡影响下的苹果目标识别与定位方法研究[D]. 杨凌：西北农林科技大学, 2016.
[14]杨超, 刘本永.基于Lab颜色空间纹理特征的图像前后景分离[J].激光与光电子学进展,2019,56(12):59-64.
[15]李扬,杨长辉,胡友呈,等. 基于凸壳及距离变换的重叠柑橘目标识别与定位方法[J].现代制造工程,2018(9):82-87.
[16]周启海. 论二维点集或线段集凸壳生成算法改进与优化的同构化方向[J]. 计算机科学, 2007, 34(7):216-218.
[17]SHI J, TOMASI. Good features to track[C]. Seattle: IEEE, 1994.
[18]陈明晶, 方源敏, 陈杰. 最小二乘法和迭代法圆曲线拟合[J]. 测绘科学, 2016, 41(1):194-197.
[19]曾庆兵，刘成良，苗玉彬，等. 基于形态学图像处理的重叠葡萄果径无损测量[J]. 农业工程学报，2009，25(9)：356-360.
[20]张艳诚，毛罕平，胡波，等. 作物病害图像中重叠病斑分离算法[J]. 农业机械学报，2008，39(2):112-115.
[21]孙国祥，汪小品，何国敏. 基于边缘链码信息的番茄苗重叠叶面分割算法[J]. 农业工程学报，2010，26(12):206-211.
[22]蒋焕煜，施经挥，任烨，等. 机器视觉在幼苗自动移钵作业中的应用[J]. 农业工程学报，2009，25(5):127-131.
[23]司永胜，乔军，刘刚，等. 基于机器视觉的苹果识别和形状特征提取[J]. 农业机械学报, 2009, 40(8):161-165.
[24]马晓丹, 刘刚, 周薇, 等. 基于量子遗传模糊神经网络的苹果果实识别[J]. 农业机械学报, 2013, 44(12):227-232.
[25]BAI X, SUN C, ZHOU F. Splitting touching cells based on concave points and ellipse fitting[J]. Pattern Recognition, 2009, 42(11): 2434-2446.

备注/Memo

备注/Memo:
收稿日期：2019-02-25 基金项目：重庆市重点产业共性关键技术创新专项(cstc2015zdcy-ztzx70003)；重庆理工大学研究生创新项目（ycx2018213） 作者简介：刘妤（1974-），女，四川泸州人，博士，教授，主要从事山地农业机械研究。(E-mail)liuyu_cq@126.com 通讯作者：杨长辉，(E-mail)yangchanghui@cqut.edu.cn
更新日期/Last Update: 2020-01-09