Recognition and classification of leaf patterns using support vector machine.

When it comes to the field of image processing and analysis, pattern recognition has the greatest potential. Boundaries provide a clear demarcation between different parts of an image pattern; as a result, characteristics linked to boundaries are often included in a feature vector used for pattern recognition. Because they play such a crucial part in photosynthesis, leaves are sometimes referred to as the power house of a plant. In a similar vein, the pattern of a leaf may be used to trace its origin, both manually and automatically. To identify a leave family, it's necessary to match leaves with their related plants. In this study, an algorithm is developed using inputs such as leaf color, shape/morphology, texture, etc. In this study, we focus largely on radial characteristics which are taken from the leaf binary representation. Besides color moments, color histogram based features are covered in this feature domain. For the classification purpose SVM is being used. At last the proposed approach is analyzed for the different feature set for computing the classification accuracy with respect to the existing methods. [ABSTRACT FROM AUTHOR]