面向人机交互的手势识别系统设计

在人机交互领域(Human-Computer Interaction,HCI)中,基于视觉的手势识别因其直观、高效的特点拥有广阔的应用前景。为了改善传统手势识别算法识别率低、鲁棒性差的缺点,基于OpenCV和Keras深度学习框架提出一种简单、快速的手势识别方法作为人机交互的接口。手势图像经过3个处理阶段:预处理、特征提取和分类。对输入图像进行预处理,使用YCbCr肤色模型提取出手部肤色区域,将其转化为灰度图像。使用卷积神经网络对手势图像进行特征提取和分类。实验结果表明:提出的手势识别方法识别率很高,达到99.43%,且具有较好的鲁棒性。与传统的人工选取特征相比,卷积神经网络能够有效地进行特征学习。

Design of Gesture Recognition System Towards Human Computer Interaction

This paper proposes a simple and fast gesture recognition method as the interface of human-computer interaction based on OpenCV and Keras deep learning framework.This paper proposes a simple and fast gesture recognition method as an interface for human-computer interaction.The gesture image goes through three processing stages:preprocessing,feature extraction and classification.The input image is preprocessed,using the YCbCr skin color model to extract the hand skin color region and convert it into a grayscale image.Feature extraction and classification of gesture images using convolutional neural networks.The experimental results show that the recognition rate of the gesture recognition method proposed in this paper is very high,reaching 99.43%,and it has good robustness.