Modelling agronomic images for weed detection and comparison of crop/weed discrimination algorithm performance

A new method for weed detection based on modelling agronomic images taken from a virtual camera placed in a virtual field is proposed. The aim was to measure and compare the effectiveness of the developed algorithms. Two sets of images with and without perspective effects were simulated. For images with no perspective, based on Gabor filtering and on the Hough transform, the performance of two crop/inter-row weed discrimination algorithms were tested and compared. The method based on the Hough transform is, in any case, better than the one based on Gabor filtering. For images with perspective effects only, an algorithm based on the Hough transform was tested and an extension to real images is discussed. These tests were done by a comparison between the weed infestation rate detected by these algorithms and the true one. This evaluation was completed with a crop/weed pixel classification and it demonstrated that the algorithm based on a Hough transform gave the best results (up to 90%).

[1]  N. D. Tillett,et al.  A Robotic System for Plant-Scale Husbandry , 1998 .

[2]  O. Faugeras.  Three-dimensional computer vision: a geometric viewpoint , 1993 .

[3]  J. F. Reid,et al.  Texture-Based Weed Classification Using Gabor Wavelets and Neural Network for Real-time Selective Herbicide Applications , 2000 .

[4]  François Goreaud,et al.  Apports de l'analyse de la structure spatiale en forêt tempérée à l'étude et la modélisation des peuplements complexes , 2000 .

[5]  B. Ripley.  Computer generation of random variables: a tutorial , 1983 .

[6]  R. B. Brown,et al.  Prescription Maps for Spatially Variable Herbicide Application in No-till Corn , 1995 .

[7]  Jaakko Astola,et al.  Image Processing: Algorithms and Systems II , 2003 .

[8]  Frédéric Truchetet,et al.  Development of a multispectral imagery device devoted to weed detection , 2004, J. Electronic Imaging.

[9]  Albert-Jan Baerveldt,et al.  An Agricultural Mobile Robot with Vision-Based Perception for Mechanical Weed Control , 2002, Auton. Robots.

[10]  Frederic Truchetet,et al.  Development of methods based on double Hough transform or Gabor filtering to discriminate between crop and weed in agronomic images , 2006, Electronic Imaging.

[11]  R. E. Miles.  On the homogeneous planar Poisson point process , 1970 .

[12]  R. A. Fischer,et al.  The role of spatial pattern in the competition between crop plants and weeds. a theoretical analysis , 1973 .

[13]  B. Hapke.  Bidirectional reflectance spectroscopy: 1. Theory , 1981 .

[14]  F. Baret,et al.  PROSPECT: A model of leaf optical properties spectra , 1990 .

[15]  Frédéric Truchetet,et al.  Spatial and Spectral Methods for Weed Detection and Localization , 2002, EURASIP J. Adv. Signal Process..

[16]  John F. Reid,et al.  DEVELOPMENT OF A PRECISION SPRAYER FOR SITE-SPECIFIC WEED MANAGEMENT , 1999 .

[17]  J. V. Stafford,et al.  Simulation of perspective agronomic images for weed detection. , 2007 .

[18]  Christine Onyango,et al.  Image Processing Performance Assessment Using Crop Weed Competition Models , 2005, Precision Agriculture.

[19]  Dennis Gabor,et al.  Theory of communication , 1946 .

[20]  Richard O. Duda,et al.  Use of the Hough transformation to detect lines and curves in pictures , 1972, CACM.

[21]  J. M. van Groenendael,et al.  Patchy distribution of weeds and some implications for modelling population dynamics: a short literature review , 1988 .

[22]  J. Cardina,et al.  The nature and consequence of weed spatial distribution , 1997, Weed Science.

[23]  Jeffery R. Price,et al.  Machine Vision Applications in Industrial Inspection XII , 2004 .

[24]  Mats Rudemo,et al.  Assessment of weed density at an early stage by use of image processing , 1997 .

[25]  Frédéric Truchetet,et al.  Crop/weed discrimination in simulated images , 2007, Electronic Imaging.