An Intelligent Discovery and Error Correction Algorithm for Misunderstanding Gesture Based on Probabilistic Statistics Model

Abstract

Numerous experiments have shown that there are similar gestures in visual-based gesture recognition. In order to solve the problem, this paper proposes a new algorithm based on Convolution neural network. According to the model test results, the confusion matrix is established. And according to the correspondence between each gesture and the predicted result, misjudgment probability matrix is established. Based on the misjudgment probability matrix, we correct the gestures that have been incorrectly identified by the Convolution neural network model. After this algorithm, the recognition rate of similar gestures is increased by 5% to 12%. The innovation of this paper lies in the secondary error correction of the wrong gesture of Convolution neural network structure.