基于改进的ResNet网络模型的静态手势识别  

Static Gesture Recognition Based on Improved ResNet Network Model

作　　者：汝昊 RU Hao(School of Optics and Electronic Information,Suzhou City University,Suzhou Jiangsu 215104,China)

机构地区：[1]苏州城市学院光学与电子信息学院,江苏苏州215104

出　　处：《信息与电脑》2024年第5期125-129,共5页Information & Computer

基　　金：江苏省高等学校大学生创新创业训练计划项目“手语姿势识别方法研究”(项目编号:202213983049Y)。

摘　　要：为有效解决卷积神经网络提取特征遗漏以及特征信息利用效率低的问题,文章提出基于改进残差网络的静态手势识别方法。在残差网络(Residual Network,ResNet)模型中,改进了残差块,优化了模型,增强了模型的特征提取能力和训练稳定性。实验结果表明,与传统ResNet34模型相比,改进ResNet模型具有更好的性能,能够提高手势图像的识别精度。To effectively solve the problems of feature omission and low efficiency of feature information utilization in convolutional neural networks,this article proposes a static gesture recognition method based on an improved residual network.In the Residual Network(ResNet)model,residual blocks were improved,the model was optimized,and the feature extraction ability and training stability of the model were enhanced.The experimental results show that compared with the traditional ResNet34 model,the improved ResNet model has better performance and can improve the recognition accuracy of gesture images.

关 键 词：手势识别 特征提取 识别精度 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]