Artificial learning based on convolution neural networks

Numerical recognition of bank cheques presents a major challenge and plays an important role in today's world, as machines must be able to learn like humans and solve complex problems such as recognizing the digits of bank cheques. Despite attempts to make machines learn like humans, no machine has yet been able to recognize 100% of handwritten digits. Despite attempts to make machines capable of learning like humans, no machine is yet 100% capable of recognizing handwritten digits. It aims to build a prediction model called a classifier that will facilitate this recognition from data in the MNIST database, with a view to possibly helping banks to speed up the processing of banking transactions by cheque.The approach proposed here essentially consists of two steps: feature extraction and classification of image pixels using a convolutional neural network, one of the deep learning algorithms with a proven track record in image processing.

Autorentext

Welcome Bukasa Muepu is a Data Scientist with a solid academic background, including a Master's degree in Mathematical Sciences, specializing in Data Science, and a Bachelor's degree in Computer Engineering. Passionate about the application of AI and mathematical modeling to medical, biological and industrial problems.