Gesture classification method based on transfer learning

Abstract

The invention discloses a gesture classification method based on transfer learning. The gesture classification method is suitable for gesture image classification. The method comprises the steps of 1,converting a gesture video V into a gesture frame data set G0; 2, performing noise removal, binarization and background segmentation processing on the G0 through a Gaussian filtering method, an OTSUalgorithm and image AND operation to obtain a gesture frame data set G1, and setting a label for the G1 to obtain a frame label data set L; 3, carrying out transfer learning by using a MobileNet convolutional neural network architecture and the weight file, and creating and training a model M1; 4, extracting features of the frame data set G1 through the model M1 to obtain a frame feature vector set F0; and 5, classifying the test set by taking the XGBoost as a classification model to obtain a final classification result. According to the method, the weight of the trained MobileNet convolutional neural network is migrated to a gesture image data set for feature extraction, and XGBoost is adopted as a classification model, so that the model calculation amount is reduced while the classification accuracy is improved.

Description

technical field [0001] The invention belongs to the technical field of supervisory algorithms and image recognition, in particular to a gesture classification method based on transfer learning. Background technique [0002] When facing the problem of gesture classification, some recent literatures only use a single model to autonomously learn features and classify images, for example: Zhang Jiangxin, Wu Xiaofeng, Xu Xinchen. A gesture detection and recognition method based on Faster R-CNN. China Patent publication number: CN107239731A, 2017.10.10; Cheng Shuying, Lin Peijie, Lu Xiaoyang. A static sign language recognition system based on XGBoost. Chinese patent publication number: CN109086699A, 2018.12.25; Wang Wei, Zou Ting, Wang Xin. A Image classification method based on D-MobileNet neural network. Chinese patent publication number: CN 109214406A, 2019.01.15. This type of method takes a lot of time to train. Some literature proposes an improved method of extracting featu...

AI Technical Summary

Problems solved by technology

But it should be noted that this algorithm is a binary tree, that is, each non-leaf node can only lead to two branches, so when a non-leaf node is a multi-level (more than 2) discrete variable, the variable has may be used multiple times

Images