Primary Pitfalls in Detection and Analysis of Diseased Chunk of Leaf Images

In Agriculture, the plant leafs may be affected by various diseases and they can be detected by image processing techniques based on color, shape and its texture. The increase in differences from a healthy to diseased leaf shows the increase in severity of the disease. But, the image which is undergoing the analysis for the detection of the diseased chunk may have busy or complex background or presence of shadow in the leaf image, inclusion of sunlight reflection which turns out to be a difficult task in applying segmentation/ clustering techniques. The diseased Plant leaf can undergo various techniques such as Image Preprocessing, Histogram, Thresholding, Segmentation, Feature extraction and Classification to identify the diseased fragment. The challenging aspect in establishing these techniques is the image condition. Certain affected leaf images may also contain more than one symptom of diseased spot which is also a pitfall in identifying the disease. This study reveals the complications of the diseased images and how can be they dealt with.