Dear Students, Kindly note, we do not have any external representatives, branches or agents for BillTech Labs, Mohali. We neither have the official associative partnership with any college, university in India and abroad areas. We donot issue any kind of certificate for M.Tech and Ph.D thesis guidance. Billtech Labs, Mohali is a technical research and development laboratory support run by innovative, creative, passionate and dedicated research and development team. All works are carried out in our lab premises located in mohali. We do not outsource your works. If you find any deviations or misuse of our brand (BillTech Labs). Kindly intimate us at info@billtechcoding.com Thank you.

Convolutional Neural Network (CNN) is the class of deep learning networks especially used image recognition. CNNs are a version of multilayer perceptrons and it refers to fully connected networks. In CNN each neuron in one layer is connected to all neurons in the next layer. The fully connected layers of these networks prevent from overfitting.CNN is known by it’s another name shift invariant or space invariant artificial neural network (SIANN) based on their shared weights architecture characteristics.CNN networks are basically inspired from natural biological processes in that neurons resembles the actual visual cortex. Applications of CNN are image and video recognition, image classification, medical image analysis and natural language processing. A CNN network consists of an input, hidden and output layers. It is a series of convolutional layers that uses multiplication or dot product. The activation function may be sigmoid, softmax in ReLU layer and connected to pooling layers, fully connected layers and normalization layers. In CNN programming, each convolutional layer in neural network performs this step number of images, image width, image height and image depth. CNNs networks may include local and global pooling layers for computation. Pooling layers reduces the data by combining the outputs of neuron clusters at one layer into a single neuron in the next layer. Local pooling combines small clusters and global pooling combines large clusters. Pooling may compute maximum or average. Max pooling used maximum value and average pooling uses average value from each of the cluster of neurons at the layer. Fully connected layers connect neuron in one layer to every neuron in another layer same as already implemented in Multi-layer perceptron neural network (MLP). In neural network, each neuron received input from number of locations in the previous layer. Each neuron in a network calculates an output value by applying a specific function to the input values coming from the receptive field in the previous layer. The function that is applied to input values is determined by a weights values and a bias. Training in a neural network by making adjustments in these biases and weights. The weights and bias are called filters and represent particular features of the input.