Conception and Appreciation of Convolution Neural Network Using PyTorch for Pattern Recognition
Showkat Ahmad Dar · SSRN Electronic Journal · 2019
In recent years, deep learning has been used in image classification, object tracking, pose estimation, text detection and recognition, visual saliency detection, action recognition and scene labeling. Auto Encoder, sparse coding, Restricted Boltzmann Machine, Deep Belief Networks (DBN) and Convolutional neural networks (CNN) is commonly used models in deep learning. In present era, machines have successfully achieved 99% accuracy in understanding, identifying the features and objects in the images. This state of the art performance of the machine was possible because of a specific type of neural network called Convolutional Neural Network, also known as convent. CNN is a special type of neural network which works exceptionally well on images. Before Convolutional Neural Networks, multilayer perceptrons have been used in building image classifiers. Image classification refers to the task of extracting information classes from a multiband (Color, Black White) raster image. Modern deep-learning frameworks like Tensorflow and PyTorch make it easy to teach machines about images as how data flows through a neural network and how the Deep layer is able to understand the data and capable of extracting the features from the data. This paper provides a clear idea how an image is passed as an input to the Convolution neural network and how CNN layers are capable of understanding and recognizing the Patterns from the Image using PyTorch.