Day by day lots of efforts are been taken towards developing an intelligent and natural interface between computer system and users. And looking at the technologies now a day’s it has become possible by means of variety of media information like visualization, audio, paint etc. Gesture has become important part of human communication to convey the information. Thus In this paper we proposed a method for HAND GESTURE RECOGNIZATION which includes Hand Segmentation, Hand Tracking and Edge Traversal Algorithm. We have designed a system which is limited to the hardware parts such as computer and webcam. The system consists of four modules: Hand Tracking and Segmentation, Feature Extraction, Neural Training, and Testing. The objective of this system to explore the utility of a neural network-based approach to the recognition of the hand gestures that create a system that will easily identify the gesture and use them for device control and convey information instead of normal inputs devices such as mouse and keyboard.