TY - JOUR
T1 - A pattern recognition strategy for visual grape bunch detection in vineyards
AU - Pérez-Zavala, Rodrigo
AU - Torres-Torriti, Miguel
AU - Auat Cheein, Fernando
AU - Troni, Giancarlo
N1 - Funding Information:
This project has been supported by the National Commission for Science and Technology Research of Chile (Conicyt) under grants Fondecyt 1140575 and Basal FB008 . We thank the Center for Research and Innovation of Concha y Toro Winery for making available their data. The authors would like to thank Manuel Reis and Carlos Pereira from UTAD, Portugal, for sharing their dataset.
Publisher Copyright:
© 2018 Elsevier B.V.
PY - 2018/8
Y1 - 2018/8
N2 - Automating grapevine growth monitoring, spraying, leaf thinning and harvesting tasks, as well as improving yield estimation and plant phenotyping, requires reliable methods for detecting grape bunches across different vineyard environmental and plant variety conditions, in which illumination, occlusions, colors and contrast are the main challenges to computer vision techniques. This work presents a method that employs visible spectrum cameras for robust grape berries recognition and grape bunch detection that does not require artificial illumination nor is limited to red or purple grape varieties. The proposed approach relies on shape and texture information together with a strategy to separate regions of clustered pixels into grape bunches. The approach employs histograms of oriented gradients (HOG) as shape descriptor and local binary patterns (LBP) to obtain texture information. A review of the existing methods and comparative analysis of different feature vectors (DAISY, DSIFT, HOG, LBP) and support vector classifiers (SVM-RBF, SVDD) is also presented. Datasets from four countries containing 163 images of different grapevine varieties acquired under different vineyard illumination and occlusion levels were employed to assess the approach. Grapes bunches are detected with an average precision of 88.61% and average recall of 80.34%. Single berries are detected with precision rates above 99% and recall rates between 84.0% and 92.5% on average. The proposed approach should facilitate the estimation of yield, crop thinning measurements and the computation of leaf removal indicators, as well as the implementation guidance strategies for precise robotic harvesters.
AB - Automating grapevine growth monitoring, spraying, leaf thinning and harvesting tasks, as well as improving yield estimation and plant phenotyping, requires reliable methods for detecting grape bunches across different vineyard environmental and plant variety conditions, in which illumination, occlusions, colors and contrast are the main challenges to computer vision techniques. This work presents a method that employs visible spectrum cameras for robust grape berries recognition and grape bunch detection that does not require artificial illumination nor is limited to red or purple grape varieties. The proposed approach relies on shape and texture information together with a strategy to separate regions of clustered pixels into grape bunches. The approach employs histograms of oriented gradients (HOG) as shape descriptor and local binary patterns (LBP) to obtain texture information. A review of the existing methods and comparative analysis of different feature vectors (DAISY, DSIFT, HOG, LBP) and support vector classifiers (SVM-RBF, SVDD) is also presented. Datasets from four countries containing 163 images of different grapevine varieties acquired under different vineyard illumination and occlusion levels were employed to assess the approach. Grapes bunches are detected with an average precision of 88.61% and average recall of 80.34%. Single berries are detected with precision rates above 99% and recall rates between 84.0% and 92.5% on average. The proposed approach should facilitate the estimation of yield, crop thinning measurements and the computation of leaf removal indicators, as well as the implementation guidance strategies for precise robotic harvesters.
KW - Grape bunch detection
KW - Grape recognition
KW - Histogram of oriented gradients
KW - Local binary pattern
KW - Precision viticulture
KW - Support vector machine
UR - http://www.scopus.com/inward/record.url?scp=85048134520&partnerID=8YFLogxK
U2 - 10.1016/j.compag.2018.05.019
DO - 10.1016/j.compag.2018.05.019
M3 - Article
AN - SCOPUS:85048134520
SN - 0168-1699
VL - 151
SP - 136
EP - 149
JO - Computers and Electronics in Agriculture
JF - Computers and Electronics in Agriculture
ER -