ABSTRACT

A novel method for detecting rice leaf disease using image processing technique called fractal dimension and chaos theory is proposed in this paper. The analysis of a diseased leaf is carried out according to its image pattern and fractal dimension, and especially box-counting ratio calculation, and chaos, are applied to be able to identify the disease pattern’s self-similarity and to recreate the fractal. The image’s self-similarity is the disease infected one which is same as when it is fully infected. This method is proposed as preliminary information for the development of an early detection system or for developing knowledge based expert system or decision support system.