Cotton is one of the most important fibers and a major play in the world's industrial and agricultural economies. Maintaining agricultural productivity depends critically on the health of cotton and disease identification. In this paper, an Inception V3 classification is proposed to quickly identify diseases in cotton plant. Firstly, adaptive wiener filter is applied to cotton diseased detection dataset to reduce the mean squared error between the intended signal and the filter's output are detected to get high quality image. Next, the processed image is given to the watershed segmentation here a topographic surface, divided according to watershed boundaries. After that cotton plant is featured using Discrete Fourier Transform (DFT) for further analysis. Finally, Inception V3 framework is processed to enhance the classification of cotton plant. Using python software the proposed framework is examined and the developed classifier results with accuracy of 92% in detecting disease when compared to other techniques.