Gesture Recognition Based On Particle Swarm Optimization Algorithm And Self Generating Neural Network

Human-computer interaction is becoming more and more important in human daily life.Gesture recognition has always been an important research field in the field of human-computer interaction,and it is also a research hotspot in recent years.With the development of human-machine interaction and computer vision,the research of gesture recognition has made great progress.With the development of computer vision,the problem has been transformed into the research of image processing,and the gesture recognition algorithm has been studied by image processing technology.Gesture recognition technology uses gestures instead of various keys,and users can directly use simple gestures to control electronic devices.It aims to use gesture as a more natural,faster and more efficient means of interaction.The image information collected by the image acquisition equipment,because of the environment information other than the gesture,the gesture recognition algorithm needs to divide the gesture information from the image.In the recognition of the gestures based on the computer,the detection and segmentation of the hand area is the precondition for accurate recognition.Then the extracted gesture image features are input into the network to train andidentify,calculate the matching degree of the gesture and all categories,take the gesture category corresponding to the maximum value as the recognition result,get the gesture recognition results and feedback the meaning of the gesture to the system,and make the response,and then complete the whole gesture recognition.Cheng.Therefore,this paper aims to study how to design a more effective gesture segmentation and recognition algorithm.Self generating neural network is a kind of self-organizing neural network.It has the advantages of high learning autonomy and no need to adjust network structure and parameters artificially.In view of the characteristics of hand gesture interaction,using the advantages of self generating neural network,this paper uses intelligent optimization algorithm to optimize the self generated neural network,and applies it to hand gesture segmentation in gesture interaction,and the recognition of gesture type is realized by generating classified neural tree.In this paper,the self generating neural network is developed and the particle swarm optimization algorithm is used to generate the network structure based on particle swarm optimization algorithm and self generated neural network.This method is mainly divided into three parts:gesture segmentation,feature extraction and gesture recognition.First,the self generated neural network algorithm based on particle swarm optimization is used to detect and segment the part of the gesture in the image.Then the features of the gestures are extracted and the feature vectors are constructed.On the basis of the extraction of Hu moments and multi-layer features,the depth of the method is used.The convolution neural network in learning is used for mining hidden information.The features are selected using embedded feature selection.Finally,the selected features are put into the network to train and classify and identify gestures.In order to get good segmentation results,a sparse representation method based on simulated annealing algorithm is also used to denoise images quickly.Experimental results show that the proposed methodcan achieve high recognition accuracy and is a feasible and efficient gesture detection and recognition method.