Sum and Difference Histograms for Texture Classification
IEEE Transactions on Pattern Analysis and Machine Intelligence
Texture features based on texture spectrum
Pattern Recognition
Digital image processing (2nd ed.)
Digital image processing (2nd ed.)
A Tutorial on Support Vector Machines for Pattern Recognition
Data Mining and Knowledge Discovery
Generalized Fourier Descriptors with Applications to Objects Recognition in SVM Context
Journal of Mathematical Imaging and Vision
| Hi-index | 0.00 |
For activities of agronomical research institute, land experimentations are essential and provide relevant information on crops. Generally accurate, they are manually done and present numerous drawbacks, such as penibility, notably for wheat ear counting. In this last case, the use of image processing methods to estimate the number of ears per square metre can be an improvement. In this paper we compare three different texture image segmentation techniques based on feature extraction by first and higher order statistical methods which have been applied on our images. The extracted features are used for unsupervised pixel classification with k-means algorithm to obtain the different classes in the image. The run length technique for which the results are closed to the manual countings with an average error of 3% seems to be the best solution. Nevertheless, a fine justification of hypothesis made on the values of the classification and description parameters is necessary and is currently done, especially for the number of classes and the size of analysis windows, through the estimation of a cluster validity index. Moreover, other statistical methods of higher order will be implemented in the future jointly with methods based on spatio-frequential transforms and specific filtering.