Smart Evaluation System

  • Ruchika Shete
  • Dr. Geetanjali Kale


The field of artificial intelligence and machine learning has done tremendous development in recent times. Handwriting recognition has gained a lot of importance in the field of computer vision and pattern recognition. The aim of the paper is smart education, and to adopt a paperless procedure for evaluation, revaluation and preservation of examination answer sheets, digital technology must be used. The proposed system focuses on segmentation of answer sheets. Different techniques can be applied to improve the recognition performance such as pre-processing, feature extraction and classification. All the experiments are conducted on the MNIST dataset. In the proposed approach we used Convolutional Neural Network and KNN for training and classification of the digits. The solution will minimize the human error and save a lot of time.