期刊名称：International Journal of Advanced Computer Research
印刷版ISSN：2249-7277
电子版ISSN：2277-7970
出版年度：2014
卷号：4
期号：15
页码：653-659
出版社：Association of Computer Communication Education for National Triumph (ACCENT)
摘要：In this research, identification and classification of cotton diseases is done. The pattern of disease is important part where some features like the colour of actual infected image are extracted from image. There are so many diseases occurred on cotton leaf so the leaf color is different for different diseases. This paper uses k-mean clustering with Discrete Wavelet Transform for efficient plant leaf image segmentation and classification between normal & diseased images using neural network technique. Segmentation is basic pre-processing task in image processing applications and it is required to extract diseased plant leaf from normal plant leaf image and image background. Image segmentation is necessary to detect objects and borderlines in images. Even though different methods are already proposed, it is still hard to accurately segment a random image by one specific method. In last years, additional attention has been given to merge segmentation algorithm and feature extraction algorithm to enhance segmentation results.
