Gesture Recognition is one of the most popular and viable solution for improving Human Computer Interaction and it has become very popular in the recent years due to its use in gaming devices like Xbox, PS4 and other devices like laptops, smartphones etc. Gesture Recognition and more specifically hand gesture recognition has usage in various application like medicine, accessibility support etc. Hand gesture recognition system can be used for interfacing between computer and human using hand gesture. This work defines a method for a human computer interface through hand gesture recognition that is able to recognize 25 static gestures from the American Sign Language hand alphabet. In this research work, proposed there arises a great demand of overcoming the existing problems to achieve greater accuracy, faster response time and the difficulties faced in handling the training data inputs and testing outputs. Hence, gives the need to use such algorithms each of which is unique in its own effective and creative way and try to bring out the best possible results. Our proposed work focuses on segment the gesture using FCM algorithm, extracting the features using (SIFT) Scale Invariant Feature Transform algorithm and classifying the gestures using deep neural network. By doing this our system obtains better performance in terms of classification parameters (Accuracy, FAR, FRR and Recognize Speed) and faster response time or delayed outputs.