Gesture recognition method based on fused skin color region segmentation and machine learning algorithm and application thereof

Abstract

The invention discloses a gesture recognition method based on fused skin color region segmentation and a machine learning algorithm and an application thereof. The method comprises the following steps: after capturing and pre-processing a gesture image, using an Otsu adaptive threshold algorithm to segment a skin color region under an YCbCr skin color space; after segmentation, segmenting a gesture by setting a gesture region decision condition, and extracting an Hu moment character and the fingertip number as feature vectors on a gesture contour; and using an SVM classifier to classify and recognize six kinds of commonly used static gestures. According to the gesture recognition method based on the fused skin color region segmentation and the machine learning algorithm provided by the invention, the gesture can be accurately located and segmented through a skin color setting gesture decision condition; and the extracted gesture contour Hu moment character and the fingertip number provide more accurate feature vectors for gesture classification, and classification and recognition are carried out on the gesture by utilizing the mature SVM classifier, thus the gesture recognition rate is guaranteed.

Description

technical field [0001] The invention relates to the technical field of gesture recognition, in particular to a gesture recognition method based on the fusion of skin color region segmentation and machine learning algorithms and its application. Background technique [0002] With the rapid development of information technology, human-computer interaction technology plays an increasingly important role in people's life. Nowadays, in order to meet the needs of people's lives, gesture recognition is being used more and more as a natural and humanized human-computer interaction method. At present, many scholars have also done a lot of work on gesture recognition technology. For example, Yang Xuewen and others combined gesture main direction and Hausdorff-like distance to gesture recognition, and solved the problem that gesture recognition is affected by gesture rotation, translation and scaling, but their experiment It can only be carried out under the conditions of stable light...

AI Technical Summary

Problems solved by technology

At present, many scholars have also done a lot of work on gesture recognition technology. For example, Yang Xuewen and others combined gesture main direction and Hausdorff-like distance to gesture recognition, and solved the problem that gesture recognition is affected by gesture rotation, translation and scaling, but their experiment It can only be carried out under the conditions of stable lighting, less noise and no face interference

Dardas et al. extracted image scale-invariant features and vectorized features, and then used feature packages and multi-class support vector machines to recognize gestures, and the recognition effect was good; however, due to the high computational complexity of the SIFT algorithm, the recognition speed was slow. Poor real-time performance

Tao Meiping and others trained image patches through an unsupervised sparse self-encoder neural network, extracted edge features of gesture images as input to the training classifier, and finally tuned the parameters of the trained classifier to improve accuracy, but only It can be realized in a limited background, but it cannot be recognized in a real background

[0003] In practice, in order to improve the recognition rate, gestures are usually recognized by means of limited backgrounds or background sketches, which cannot exclude interference such as faces, lighting, and similar skin colors, so it is not conducive to natural human-computer interaction. At present, in real environments, there are still No mature gesture recognition system is widely available

Images