Convolutional Neural Network (CNN) for Image Detection and Recognition

TLDR

Deep learning algorithms, modeled after the human cortex, use deep neural networks with many hidden layers, and convolutional neural networks train large image datasets with millions of parameters by convolving 2D inputs with filters to produce desired outputs. The study builds CNN models to evaluate their performance on image recognition and detection datasets. The algorithm was implemented on MNIST and CIFAR‑10 datasets, and its performance was evaluated. The models achieved 99.6% accuracy on MNIST, and on CIFAR‑10 they employed real‑time data augmentation and dropout while running on a CPU.

Abstract

Deep Learning algorithms are designed in such a way that they mimic the function of the human cerebral cortex. These algorithms are representations of deep neural networks i.e. neural networks with many hidden layers. Convolutional neural networks are deep learning algorithms that can train large datasets with millions of parameters, in form of 2D images as input and convolve it with filters to produce the desired outputs. In this article, CNN models are built to evaluate its performance on image recognition and detection datasets. The algorithm is implemented on MNIST and CIFAR-10 dataset and its performance are evaluated. The accuracy of models on MNIST is 99.6 %, CIFAR-10 is using real-time data augmentation and dropout on CPU unit.