This post explains the application of convolutional neural networks (CNN) in visual recognition problems. It covers the basics of CNN, including convolution, padding, strides, pooling, and the design of a CNN architecture. It also introduces the LeNet-5 architecture and provides a step-by-step guide to implementing a CNN with TensorFlow 2.0.
Table of contents
Technical requirementsIntroduction to Conventional Visual RecognitionWhat is Digital Image ProcessingArtificial Neural Network for Digital Image ProcessingBuilding Blocks of Convolutional Neural NetworkConvolution over 2D ImageConvolution over 3D ImagePaddingConvolution StridesPooling OperationSort: