In recent years, deep artificial neural networks is swiftly emerging in pattern recognition and machine learning. Convolutional Neural Networks (CNNs) is one of the most remarkable forms of Artificial Neural Network (ANN) architecture. This paper focuses on Convolutional neural networks and its visualization technique. Compared to other techniques Convolutional neural networks (CNNs) are mostly used in image and pattern recognition problems since they have a number of advantages compared to other techniques. This paper emphasizes on the fundamental of CNNs, functioning and description of the different layers used in Convolutional neural networks.