The objective of this paper is to identify the diseases in the leaves of the all plants. The detection of disease is one of the crucial importance to increase the productivity of plants. Plant pathology is the scientific study of plant diseases, based on this study, to identify the pathogen, and classification under different kind of circumstances. In existing, the automatic plant disease detection without training set using only prior knowledge. The proposed method, to use detects the diseases by using the ELM (Extreme Learning Machine) techniques. In the above method, initially the ABC Clustering has approached for segments the affected area. Secondly, using TELP (Transform Encoded Local Pattern), Gradient Features, Color Histogram Techniques for extract the features from that affected area. Finally, these extracted features are given to the ELM classifier to get the exact type of disease affected on a leaves.