Convolutional Neural Network for Dummies

27 Nov 2019

---

Convolutional Neural Nets (ConvNets or CNN) are a class of neural network specialized for image processing. Like other neural networks, they transform input to output through many layers. In CNN layers have a Convolution step, a pooling step (optional) and a non-linear activation. Each layer in a neural network transforms the input tensor into the output tensor through linear and non-linear operations. All these intermediate tensors (including the network input and output) are called activations and they are all different representations of the input.

Why CNN?

We know that Convolutional Neural Networks (CNN) are good at Image Recognition tasks. Now if we have the Image Recognition task then it also can be done with Simple Neural Networks. But if theme images are of large pixels then the problem arises as the number of parameters of the Neural Network will also increase. That makes Neural Networks very slow and computationally expensive.