ANALYSIS ON HAND GESTURE RECOGNITION USING ARTIFICIAL NEURAL NETWORK
Hemant Maurya, Charu Chauhan, Himanshu Tiwari, Anusha Jain · Ethics and Information Technology · 2020
The main goal of gesture recognition research is to create a system that can recognize specific human gestures and use it to transmit information or to control the device.Human-computer interaction needs to use various methods, such as location, voice, gestures, lip development, facial joints, etc., and coordinate them to provide an increasingly vivid customer experience.The visual interpretation of gestures can be used to complete the natural human-computer interaction.In this article, we propose a method for recognizing gestures.We have designed a system that can recognize specific gestures and use them to transmit information.We have used backward propagation and training algorithms based on supervised feeding neural networks to classify gestures.A neural network program was developed in MATLAB, which can recognize the number of fingers and achieve correct results of 95.4% testing accuracy.An effective and effective gesture recognition method is proposed.The hand area is detected by background subtraction.Then spread your palm and fingers apart to identify your fingers.After recognizing fingers, you can classify gestures with a simple ANN classifier.