Virtual Alphabet Recognition using Deep Convolution Neural Networks

Air-based visual perception plays an important role in human interactions with computers. With the advent of many new technological advances such as biometric authentication, which is now regularly used in smartphones, similarly touch recognition is a modern way of interacting between a computer and humans i.e. the system can be controlled by showing or moving the hands in front of a webcam. A hand touch detection can be helpful for all types of people. Spiritual writing recognition plays a major role in intelligent programming and application. To date, some of the most common isolation challenges have not been successfully resolved. In this case, the user drawn character in the air is captured by a computer camera, following the prediction of drawn character. Therefore, while capturing aerial footage, the video camera should be turned ON. Now, an object is defined based on its color to determine the movement of the user. The color is captured by the lower and upper limit of HSV (hue saturation value), which leads to the final acquisition of the object at all times. It uses open CV and machine learning algorithms to capture and predict the written text.