In India and other Asian nations, the prevalence of ulcerative colitis (UC) is rising. It is an idiopathic, chronic, inflammatory condition influenced by genes, the environment, and the immune system. Endoscopy and colonoscopy are used to support gastroenterologists' diagnoses. However, UC detection is a very challenging process. This study aims to use computational techniques to determine Ulcerative Colitis remission without the support of medical experts. Finding the optimal classifier for multiclass image classification issues is more difficult due to the high-level, in-depth properties of the images. An innovative pre-trained Inception V3 model is used in this case for better classification UC images into many categories. The major goal of this research is to automatically identify and effectively learn key aspects from the UC picture. In this research, diverse stages of the UC images are classfied. The proposed method is contrasted with current state-of-the-art procedures, and the results and analysis show that the research is quite successful by achieving a 95% accuracy rate.