LeNet, developed by French computer scientist Yann Lecun, was the frontrunner to the convolutional neural network (CNN). His breakthrough came when he conceived a neural network modelled on the human visual cortex. He called it a convolutional neural network, inspired by Kunihiko Fukushima, a Japanese computer scientist. CNNs process an image by dividing it into squares, analysing each one separately to find small patterns, and later piecing them together to make sense, just like the human brain’s visual cortex.

The Turing award winner changed computer vision and, for that matter, artificial intelligence forever with CNNs.

Convolutional neural networks are made of multiple layers of artificial neurons that calculate the weighted sum of various inputs and produces an activation value. The primary application areas of CNN include image recognition, image classification, object detection, and face recognition.

Here, we have curated a list of the top best books to learn CNNs.

Java deep learning cookbook

Deeplearning4j is one of the most popular Java libraries for training neural networks efficiently. This book by Rahul Raj offers ways to perform deep learning using the same library. At the very start, the book will help the user install and configure Java and DL4J on their systems. The book delves deep into anomaly detection of unsupervised data and effectively setting up neural networks in distributed systems. It also explains how to import models from Keras and modify the configuration of a pre-trained DL4J model and optimise neural networks for the best results.

Find the book here.

Deep learning from scratch

This book by Seth Weidman explains the inner working of neural networks. The book will teach the user how to apply convolutional neural networks, multilayer neural networks, and recurrent neural networks from scratch. The text is full of working code examples and mathematical explanations to understand neural networks better. The book offers a detailed introduction to data scientists and software engineers with the machine learning experience.

Find the book here

Artificial vision and language processing for robotics

The authors of this book, Alvaro Morena Alberola, Gonzalo Molina Gallego, Unai Garay Maestre, explains how to build a basic convolutional neural network and improve it using a generative model. The book also covers topics like implementing AI and object recognition with the help of deep learning. Artificial Vision and Language Processing for Robotics is the go-to book for engineers who want to learn how to combine computer vision and deep learning techniques to build robotic systems.

Find the book from here.

Hands-on computer vision with TensorFlow 2

This book from Benjamin Planche and Eliot Andres will help readers learn how to use convolutional neural networks (CNNs) for visual tasks. The book shows how to identify images using modern solutions like Inception and ResNet and extract relevant content with YOLO, Mask R-CNN, and U-Net. The book riffs on transfer learning, GANs, and domain adaptation and teach how to use recurrent neural networks for video analysis. 

Find the book here.

Deep learning with R for beginners

The authors, Mark Hodnett, Joshua F. Wiley, Yuxi (Hayden) Liu, Pablo Maldonado, offer the basics of deep learning in this book. The readers can learn how to train convolutional neural networks (CNNs), recurrent neural networks (RNNs), and extended short-term memory networks (LSTMs) in R using real-world projects. The book is ideal for data scientists, data analysts, machine learning developers, and deep learning enthusiasts who want to learn more about the deep learning paradigm using R.

Find the book here.

Deep learning with Python

Written by Mark Graph, the book will provide a basic understanding of deep learning and its applications in the real world. The book deals with creating a CNN, ethical implications of deep learning, an overview of TensorFlow and PyTorch, basics and advanced programming in Python etc. The book is a clinic on how to code in Python and use Python libraries to create neural networks.

Find the book here.

Hands-on neural networks

The authors of the book are Leonardo De Marchi and Laura Mitchell. The book will help you understand and learn neural networks in a more practical way. You will learn to use embeddings to process textual data and use extended short-term memory networks (LSTMs) to solve basic natural language processing (NLP) problems. The book also dives deep into advanced concepts such as transfer learning, generative adversarial networks (GANs), autoencoders, and reinforcement learning.

Find the book here.

Deep learning and CNN for medical imaging and clinical informatics

Editors of the book include Le Lu, Xiaosong Wang, Gustavo Carneiro and Lin Yang. It primarily focuses on convolutional neural networks and recurrent neural networks such as the LSTM, with multiple practical examples. The book discusses how deep neural networks can answer new questions, develop new protocols, and resolve current challenges in medical image computing. It describes the various deep learning approaches for object and landmark detection tasks in 2D and 3D medical imaging.

Find the book here.

Advanced applied deep learning

Author Umberto Michelucci covers basic CNN operations like convolution and pooling and more complex architectures like inception networks, resnets, etc. You can learn about working effectively with Keras, Dataset TensorFlow abstraction, and removing and adding layers to pre-trained networks for customisation. 

Find the book here.