I am working on detection and classification of rice varieties using hyperspectral imaging. I wants to see how colour, morphological and textural features can be extracted from the image and used for the classification.
See here we are talking about the feature extraction and texture so let us handle it separately to under stand in a better way and texture is a feature used to or in the partition images into regions of interest and then later on to classify various or those selected regions whereas the feature extraction is nothing but a method of capturing the visual contents of the images which will be or may be for indexing and then alter on retrieval. Statistical based or model based and transform based methods are used in case of feature extraction and in case of texture this itself is characterized by the spatial distribution of level in the particular region.