Bayesian classification and unsupervised learning for isolating weeds in row crops

This paper presents a weed/crop classification method using computer vision and morphological analysis. Subsequent supervised and unsupervised learning methods are applied to extract dominant morphological characteristics of weeds present in corn and soybean fields. The novelty of the presented technique resides in the feature extraction process that is based on spatial localization of vegetation in fields. Features from the weed leaf area distribution are extracted from the cultivation inter-rows, then features from the crop are inferred from the mixture model equation. Those extracted features are then passed to a naive bayesian classifier and a gaussian mixture clustering algorithm to discriminate weed from crop plant. The presented technique correctly classifies an average of 94 % of corn and soybean plants and 85 % of the weed (multiple species) without any prior knowledge on the species present in the field.

[1]  Anil K. Jain,et al.  Unsupervised Learning of Finite Mixture Models , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Chun-Chieh Yang,et al.  Development of an Image Processing System and a Fuzzy Algorithm for Site-Specific Herbicide Applications , 2003, Precision Agriculture.

[3]  J. V. Stafford,et al.  Implementing precision agriculture in the 21st century. , 2000 .

[4]  A. J. S. McDonald,et al.  Weed and crop discrimination using image analysis and artificial intelligence methods , 2003 .

[5]  Bernard Panneton,et al.  Colour representation methods for segmentation of vegetation in photographs , 2009 .

[6]  Albert-Jan Baerveldt,et al.  A vision based row-following system for agricultural field machinery , 2005 .

[7]  Ethem Alpaydin,et al.  Introduction to machine learning , 2004, Adaptive computation and machine learning.

[8]  Alberto Tellaeche,et al.  A computer vision approach for weeds identification through Support Vector Machines , 2011, Appl. Soft Comput..

[9]  M. DeLorenzo,et al.  Toxicity of pesticides to aquatic microorganisms: A review , 2001, Environmental toxicology and chemistry.

[10]  Muhammad Asif,et al.  A Vision System for Autonomous Weed Detection Robot , 2010 .

[11]  Louis Longchamps,et al.  Discrimination of corn, grasses and dicot weeds by their UV-induced fluorescence spectral signature , 2010, Precision Agriculture.

[12]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[13]  Roland Gerhards,et al.  The Economic Impact of Site-Specific Weed Control , 2003, Precision Agriculture.

[14]  Frederic Truchetet,et al.  Original paper: Assessment of an inter-row weed infestation rate on simulated agronomic images , 2009 .

[15]  E. Parzen.  On Estimation of a Probability Density Function and Mode , 1962 .

[16]  David C. Slaughter,et al.  Autonomous robotic weed control systems: A review , 2008 .

[17]  Richard D. Deveaux,et al.  Applied Smoothing Techniques for Data Analysis , 1999, Technometrics.

[18]  Gilles D. Leroux,et al.  Validation of an Operator-Assisted Module to Measure Weed and Crop Leaf Cover by Digital Image Analysis , 1998, Weed Technology.

[19]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[20]  K. Freemark,et al.  Impacts of agricultural herbicide use on terrestrial wildlife in temperate landscapes: A review with special reference to North America , 1995 .

[21]  M. Rosenblatt.  Remarks on Some Nonparametric Estimates of a Density Function , 1956 .

[22]  J. Hemming,et al.  PA—Precision Agriculture: Computer-Vision-based Weed Identification under Field Conditions using Controlled Lighting , 2001 .