| Agricultural engineering |
|
|
|
|
| Research on recognition methods for red tomato image in the natural environment |
Xiaohui WANG( ),Kunpeng ZHOU() |
| College of Engineering, Inner Mongolia University for Nationalities, Tongliao 028000, Inner Mongolia, China |
|
|
|
Abstract In view of actual situations such as light change, soil, branch and leaf background and fruit overlap in the natural environment, which causing the problem of red tomato recognition during the robotic picking process was not accurate, a tomato image recognition method based on circle fitting algorithm was proposed. We collected the images of tomato by camera, used the red, green, blue (RGB) color space based Matlab as simulation experiment, and preprocessed the tomato images with red-green (R-G) color component. Then, edge detection algorithm, threshold segmentation and watershed segmentation methods were adopted to segment tomato target and the background, respectively. The Otsu segmentation method of threshold segmentation was adopted, which was best to segment target. We used the back propagation-artificial neural network (BP-ANN) and circle fitting algorithm to recognize the tomato fruit. Finally, the contour, centroid and radius of the red tomato were obtained. The results of red tomato images were statistically analyzed, and the recognition rate of circle fitting algorithm was as high as 90.07%. This algorithm not only has a high recognition rate for single fruit, but also solves the problem of multiple fruit overlapping in a complex environment, which lays a good foundation for the following robotic picking work.
|
|
Received: 10 September 2020
Published: 25 June 2021
|
|
|
|
Corresponding Authors:
Kunpeng ZHOU
E-mail: 171428343@qq.com;kunpeng032@126.com
|
自然环境中的红色番茄图像识别方法研究
针对机器人采摘过程中因对自然环境中的光照变化、土壤及枝叶等背景和果实间重叠等实际情况造成的红色番茄识别不准确的难题，提出了一种基于圆拟合算法的番茄图像识别方法。使用照相机采集番茄图像，在Matlab软件平台中选择三原色（red, green, blue, RGB）彩色空间进行实验；利用红-绿（red-green, R-G）色差分量对番茄图像进行预处理，然后分别采用边缘检测算法、阈值分割和分水岭分割方法对果实目标和背景进行分割，最终选用阈值分割中的最大类间方差法进行图像分割，并基于反向传播人工神经网络（back propagation-artificial neural network, BP-ANN）和圆拟合算法进行番茄果实的识别，最终得到红色番茄果实的轮廓、质心和半径，即定位果实目标。对红色番茄图像的识别结果进行统计，圆拟合算法的识别率高达90.07%。此算法不仅对单个果实的识别率高，还较好地解决了复杂环境下多个果实重叠的识别问题，为后续的机器人采摘工作打下了良好的理论基础。
关键词：
自然环境,
番茄图像,
预处理,
分割,
识别
|
|
| [1] |
ILLINGWORTH J, KITTLER J. A survey of the Hough transform. Computer Vision, Graphics, and Image Processing, 1988,44:87-116.
|
|
|
| [2] |
KONDO N, NISHITSUJI Y, LING P P. Visual feedback guided robotic cherry tomato harvesting. Transactions of the ASAE, 1996,39(6):2331-2338. DOI:10.13031/2013.27744
doi: 10.13031/2013.27744
|
|
|
| [3] |
赵杰文,刘木华,杨国彬.基于HIS颜色特征的田间成熟番茄识别技术.农业机械学报,2004,35(5):122-124, 135.
ZHAO J W, LIU M H, YANG G B. Discrimination of mature tomato based on HIS color space in nature outdoor scenes. Transactions of the Chinese Society for Agricultural Machinery, 2004,35(5):122-124, 135. (in Chinese with English abstract)
|
|
|
| [4] |
李印清,郑小东,王晓洁.基于区域特征的果实识别技术.河南科学,2008,26(3):339-342.
LI Y Q, ZHENG X D, WANG X J. Technology of discriminating fruit based on region feature. Henan Science, 2008,26(3):339-342. (in Chinese with English abstract)
|
|
|
| [5] |
孙建桐,孙意凡,赵然,等.基于几何形态学与迭代随机圆的番茄识别方法.农业机械学报,2019,50():22-26, 61. DOI:10.6041/j.issn.1000-1298.2019.S0.004
SUN J T, SUN Y F, ZHAO R, et al. Tomato recognition method based on iterative random circle and geometric morphology. Transactions of the Chinese Society for Agricul-tural Machinery, 2019,50():22-26, 61. (in Chinese with English abstract)
doi: 10.6041/j.issn.1000-1298.2019.S0.004
|
|
|
| [6] |
皮伟强,杜健民,陈程,等.基于归一化植被指数对荒漠化草原地面高光谱影像中植被的识别.内蒙古农业大学学报(自然科学版),2018,39(4):75-79. DOI:10.16853/j.cnki.1009-3575.2018.04.012
PI W Q, DU J M, CHEN C, et al. Identification of vegetation in high-spectral images of desertification grassland based on normalized vegetation index. Journal of Inner Mongolia Agricultural University (Natural Science Edition), 2018,39(4):75-79. (in Chinese with English abstract)
doi: 10.16853/j.cnki.1009-3575.2018.04.012
|
|
|
| [7] |
纪姚林,黄慧明.基于边缘信息的二维Otsu阈值分割算法的改进.信息技术,2017(11):43-45, 48. DOI:10.13274/j.cnki.hdzj.2017.11.011
JI Y L, HUANG H M. Improvement of 2-D Otsu threshold segmentation algorithm based on edge information. Information Technology, 2017(11):43-45, 48. (in Chinese with English abstract)
doi: 10.13274/j.cnki.hdzj.2017.11.011
|
|
|
| [8] |
王晓慧,周立丽,张红娜.自然环境中的红色苹果图像分割研究.内蒙古民族大学学报(自然科学版),2016,31(3):199-202. DOI:10.14045/j.cnki.15-1220.2016.03.005
WANG X H, ZHOU L L, ZHANG H N. Red apples’ image segmentation in natural environment. Journal of Inner Mongolia University for Nationalities (Natural Sciences), 2016,31(3):199-202. (in Chinese with English abstract)
doi: 10.14045/j.cnki.15-1220.2016.03.005
|
|
|
| [9] |
桂江生,吴子娴,李凯.基于卷积神经网络模型的大豆花叶病初期高光谱检测.浙江大学学报(农业与生命科学版),2019,46(2):256-262. DOI:10.3785/j.issn.1008-9209.2018.05.151
GUI J S, WU Z X, LI K. Hyperspectral imaging for early detection of soybean mosaic disease based on convolutional neural network model. Journal of Zhejiang University (Agriculture and Life Sciences), 2019,46(2):256-262. (in Chinese with English abstract)
doi: 10.3785/j.issn.1008-9209.2018.05.151
|
|
|
| [10] |
余心杰,吴雄飞,沈伟良.基于计算机视觉的岱衢族大黄鱼选育群体外形特征模式识别方法.浙江大学学报(农业与生命科学版),2018,44(4):490-498. DOI:10.3785/j.issn.1008-9209.2018.07.030
YU X J, WU X F, SHEN W L. Pattern recognition method for the identification of Daiqu large yellow croaker based on computer vision. Journal of Zhejiang University (Agriculture and Life Sciences), 2018,44(4):490-498. (in Chinese with English abstract)
doi: 10.3785/j.issn.1008-9209.2018.07.030
|
|
|
| [11] |
ZHANG Z Y. A flexible new technique for camera calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000,22(11):1330-1334. DOI:10.1109/34.888718
doi: 10.1109/34.888718
|
|
|
| [12] |
FAN J P, ZENG G H, MATHURIN B, et al.Seeded region growing:an extensive and comparative study. Pattern Recognition Letters, 2005,26:1139-1156. DOI:10.1016/j.patrec.2004.10.010
doi: 10.1016/j.patrec.2004.10.010
|
|
|
| [13] |
陈磊,薛河儒,郜晓晶,等.基于二分法与BP神经网络的牛乳体细胞识别.内蒙古农业大学学报(自然科学版),2020,41(2):69-74. DOI:10.16853/j.cnki.1009-3575.2020.02.013
CHEN L, XUE H R, GAO X J, et al. Milk somatic cells recognition based on dichotomy method and BP neural network. Journal of Inner Mongolia Agricultural University (Natural Science Edition), 2020,41(2):69-74. (in Chinese with English abstract)
doi: 10.16853/j.cnki.1009-3575.2020.02.013
|
|
|
| [14] |
王吉权.BP神经网络的理论及其在农业机械化中的应用研究.沈阳:沈阳农业大学,2011:15-20.
WANG J Q. Research on BP neural network theory and its application in agricultural mechanization. Shenyang: Shenyang Agricultural University, 2011:15-20. (in Chinese with English abstract)
|
|
|
| [15] |
丁金婷,臧泽林,黄敏.模糊方法改进的反向传输神经网络预测南美白对虾养殖的水质.浙江大学学报(农业与生命科学版),2017,43(1):128-136. DOI:10.3785/j.issn.1008-9209.2016.03.291
DING J T, ZANG Z L,HUANG M. Penaeus vannamei aquaculture water quality prediction based on the improved back propagation neural network. Journal of Zhejiang University (Agriculture and Life Sciences), 2017,43(1):128-136. (in Chinese with English abstract)
doi: 10.3785/j.issn.1008-9209.2016.03.291
|
|
|
| [16] |
陈雪艳.改进的遗传神经网络特征提取和分类应用.内蒙古民族大学学报(自然科学版),2016,31(1):31-35. DOI:10.14045/j.cnki.15-1220.2016.01.009
CHEN X Y. Improved genetic neural network for feature extraction and classification. Journal of Inner Mongolia University for Nationalities (Natural Sciences), 2016,31(1):31-35. (in Chinese with English abstract)
doi: 10.14045/j.cnki.15-1220.2016.01.009
|
|
|
| [17] |
张萌,李光辉.基于RELIEF算法和极限学习机的苹果轻微损伤高光谱检测方法.浙江大学学报(农业与生命科学版),2019,45(1):126-134. DOI:10.3785/j.issn.1008-9209.2017.09.043
ZHANG M, LI G H. Detection method of slight bruises of apples based on hyperspectral imaging and RELIEF-extreme learning machine. Journal of Zhejiang University (Agriculture and Life Sciences), 2019,45(1):126-134. (in Chinese with English abstract)
doi: 10.3785/j.issn.1008-9209.2017.09.043
|
|
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
