A Combination Of Dynamic Characteristics And Trajectories Of The Hand Gesture Recognition

Hand gestures play a natural and intuitive communication mode for all human dialogs. The ability for computer to visually recognize hand gestures is essential for future human-computer interaction. However, vision-based recognition of hand gestures reasons is an extremely challenging interdisciplinary project for the following hand gestures are rich in diversities, mufti-meanings and space-time varieties; human hands are complex non-rigid objects; computer vision itself is an ill-posed problem.This paper discussed the research of vision-based Dynamic Gesture Recognition based in 3 aspects: gesture image preprocessing, feature extraction and the design of classifier.In the process of image preprocessing we extract the frames that we are interested in from the video file first, then there are several image operations we should do. We turn the RGB color images into gray-scaled images, take the smoothing measure to reduce the noise in the images, then we get the binary version of the images by the means of adaptive shrinkage value.Following the image preprocessing, it's turn to extract the right feature from the gesture. We grossly classified the gesture according to the track of the hand and the shape of the hand, and then we calculate the value of the feature.In the classifier designing, we split the 14 sets of gesture images into 10 testing sets and 4 designing sets. The system gets the feature from the designing sets and save it as a template, and then it calculates the feature of the input gesture of the testing sets. Comparing with the template using Euclideam Distence, it gets the result by nearest neighbor rule.The recognition ratio is proved 93.3%, the result shows our system is totally effective.