摘要
研究了一种基于视频的手势识别算法,该算法利用均值法去除噪声,根据物体颜色配合最大类间方差法对目标和背景进行区分,并使用BP神经网络进行手势的分类.实验表明,该算法对部分典型手势识别的准确率达到74.7%,具有较高的实际应用价值.
We study a kind of hand gesture recognition algorithm based on video. Average method is used in re- moving noise, the algorithm based on object color togelher with the OTSU is used in distinguishing target and background, and the BP neuron network is used in classifying hand gestures. Experiments show that the accu- racy of the algorithm is 74.7%, so its application value is high.
出处
《延边大学学报(自然科学版)》
CAS
2013年第3期211-214,共4页
Journal of Yanbian University(Natural Science Edition)
关键词
手势识别
神经网络
视频
算法
gesture recognition
artificial neuron network
video
algorithm