Large-Scale Investigation of Weed Seeds Identification by Machine Vision Techniques

We explore the feasibility of implementing fast and reliable computer-based systems for the automatic identification of weed seeds from color and black and white images. Seeds size, shape, color and texture characteristics are obtained by standard image-processing techniques, and their discriminating power as classification features is assessed. These investigations are performed on a database much larger than those used in previous studies, containing 10310 images of 236 different weed species. We consider the implementation of a simple Bayesian approach (näıve Bayes classifier) and (single and bagged) artificial neural network systems for seed identification. Our results indicate that the näıve Bayes classifier based on an adequately selected set of classification features has an excellent performance, competitive with that of the comparatively more sophisticated neural network approach. In addition, we discuss the possibility of using only morphological and textural characteristics as classification features, which would reduce the operational complexity and hardware cost of a commercial system since they can be obtained from black and white images. We find that, under particular operational conditions, this would result in a relatively small loss in performance when compared to the implementation based on color images.

[1]  D. Bertrand,et al.  Identification of Seeds by Colour Imaging: Comparison of Discriminant Analysis and Artificial Neural Network , 1996 .

[2]  F. Lai,et al.  Discrimination of wheat and nonwheat components in grain samples by image analysis , 1989 .

[3]  Y. Pomeranz,et al.  Image analysis and characterization of cereal grains with a laser range finder and camera contour extractor , 1989 .

[4]  E. Shwedyk,et al.  An instrumental system for cereal grain classification using digital image analysis , 1987 .

[5]  John B. Shoven,et al.  I , Edinburgh Medical and Surgical Journal.

[6]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[7]  E. Shwedyk,et al.  Wheat grain colour analysis by digital image processing II. Wheat class discrimination , 1989 .

[8]  Robert M. Haralick,et al.  Textural Features for Image Classification , 1973, IEEE Trans. Syst. Man Cybern..

[9]  F. S. Lai,et al.  Discrimination between wheat classes and varieties by image analysis , 1986 .

[10]  S. R. Draper,et al.  The measurement of new characters for cultivar identification in wheat using machine vision , 1986 .

[11]  P. I. Jansen.  Seed production quality in Trifolium balansae and T. resupinatum: The role of seed colour , 1995 .

[12]  Mohan M. Trivedi,et al.  Segmentation of a high-resolution urban scene using texture operators , 1984, Comput. Vis. Graph. Image Process..

[13]  J. F. Reid,et al.  Color Classifier for Symptomatic Soybean Seeds Using Image Processing. , 1999, Plant disease.

[14]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[15]  G. W. Krutz.  Automatic identification of weed seeds by colour machine vision , 1992 .

[16]  Anil K. Jain,et al.  Feature Selection: Evaluation, Application, and Small Sample Performance , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[17]  Stephen J. Symons,et al.  Determination of wheat kernel morphological variation by digital image analysis: I. Variation in Eastern Canadian milling quality wheats , 1988 .

[18]  D. Bertrand,et al.  Feature selection by a genetic algorithm. Application to seed discrimination by artificial vision , 1998 .

[19]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[20]  E. Shwedyk,et al.  Discrimination of wheat class and variety by digital image analysis of whole grain samples , 1987 .

[21]  Pablo M. Granitto,et al.  Weed seeds identification by machine vision , 2002 .

[22]  E. Shwedyk,et al.  Wheat grain colour analysis by digital image processing I. Methodology , 1989 .