kodekarim / convolution-neural-network Goto Github PK

Convolution-Neural-Network

Convolution Neural Network for Image Recognition

The convolutional neural network is a class of deep learning, feed-forward artificial neural networks, most commonly applied to analyzing visual imagery. CNN use a variation of multilayer perceptrons designed to require minimal preprocessing.

The CNN Model is using Tensor flow in the backend the main library used is Keras and sub modules:

• Sequential
• Convolution2D
• MaxPooling2D
• Flatten
• Dense
• Image Data Generator 

The image processing is done on two-part and the first part is subcategorize in four step Part 1 - Building the CNN

1. Convolution
2. Max Pooling
3. Flattering
4. Full Connection

Part 2 - Fitting the CNN to the images