Hand Gesture Recognition System Design By Using 5GHz Radar

Hand gesture recognition has long been a study topic in the field of Human Computer Inter-action.Traditional camera-based hand gesture recognition systems can not work properly under dark circumstances.In this paper,a radar-based hand gesture recognition system using 5GHz microwave signal and machine learning algorithm is proposed.There are two part of this work.First of all,a complete hand gestures system had been proposed.We use a double sub-carrier modulation centered at 5GHz band IF architecture to collect gesture data samples.Then,the re-ceived hand gesture signals are then processed with time-frequency analysis.Convolutional neural networks are used to classify different gestures.Experimental results verify the effectiveness of the system.Besides,related factors such as recognition distance and gesture scale are investigated.In the second part,we proposed a new algorithm called deformable deep generative adversar-ial networks to mitigate the two main problems,over-fitting and longer testing time,caused by the small database and increased categories of targeted gesture.And detailed experiments verifies the effectiveness of the algorithm.