摘要
人体动作的精确识别面临多方面的挑战,特别是动作采集易受光照强度的影响、动作特征描述不清楚和易物理变形。为了降低这些不利因素的影响,提高动作的识别精确度,该文从3个步骤展开研究:首先,对Kinect提取的人体关节数据进行预处理,从而克服光照问题;随后,使用针对性编码方法对人体动作数据进行编码,进而利用卷积神经网络(convolutional neural network,CNN)自动提取人体动作特征,解决动作特征描述的难题;最后,使用SoftMax完成复杂动作的识别。实验表明,该文算法具有较高的识别准确率和泛化能力,其F1值普遍在0.8以上;在单一属性测试中,复合属性数据比被复合的单一属性数据更有优势,F1值可达0.916;混合属性测试的F1值相比单一属性测试有所下降,下降幅度最高可达约25%。
It is affected by many negative factors to recognize human motions accurately,such as the affecting of the light intensity during motion data collecting,the vagueness and transformation of the human motion features.To decrease the effect of these adverse factors and improve the accuracy of motion recognition,the following contents are investigated in this paper.Firstly,the human joint data collected by Kinect are preprocessed to overcome the illumination problem.Secondly,encoding methods are proposed to encode the preprocessed data,and then the encoded data are inputted into CNN to extract human motion features automatically,making the description of motion feature easier.Finally,the CNN completes motion classification with SoftMax.These experiments show the proposed algorithm can achieve a high recognition accuracy(most of the F1 values are larger than 0.8)and can adapt to different data settings;the compound property data are better than single property data in single property tests and the F1 value can be 0.916;the F1 values in compound property tests are smaller than those of single property tests and the maximal decrease percentage can be 25%.
作者
胡青松
张亮
丁娟
李世银
HU Qing-song;ZHANG Liang;DING Juan;LI Shi-yin(School of Information and Control Engineering,China University of Mining&Technology Xuzhou Jiangsu 221116;Zhuhai College,Jilin University Zhuhai Guangdong 519041)
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
2020年第3期473-480,共8页
Journal of University of Electronic Science and Technology of China
基金
国家自然科学基金(51874299,61771474)
江苏省科技成果转化专项资金(BA2016016)。
关键词
卷积神经网络
特征提取
人体动作识别
动作数据编码
convolutional neural network
feature extraction
human motion recognition
motion data encoding
