基于视觉的手势识别系统研究-控制理论与控制工程专业论文.docx

内蒙 内蒙古科技大学硕士学位论文 IV IV Implements the gesture tracking method based on the center of gravity and gesture tracking method based on CamShift algorithm. Finally, the median filtering, space transformation, Canny edge detection, feature extraction based on Hu moment, shape matching, focus tracking method and CamShift tracking method is verified. The results of gesture recognition and tracking are analyzed. The following conclusions: Collected in the natural scene gestures can complete the segmentation, and remove most of the image noise. About the ten different gesture of 1 to 10, recognition rate is different. Gestures 1, gesture 7 and 10 recognition rate is low. The rest of the gesture recognition rate is about 80%. Gesture tracking effect is stable, can realize tracking lost back. But there are still has some problems. Standard gesture library is not very perfect. Not training samples to a lot of gestures. The representation of the sample gestures is not strong. For more complex scenarios of gesture recognition remains to be improved. Key Words：Machine vision； Hu moment；Shape matching；CamShift algorithm PAGE PAGE VI 目 录 摘 要 I Abstract III HYPERLINK \l _bookmark0 引 言 1 HYPERLINK \l _bookmark1 1 手势识别综述 2 HYPERLINK \l _bookmark2 1.1 手势识别的应用领域 2 HYPERLINK \l _bookmark3 1.2 国内外研究现状 3 HYPERLINK \l _bookmark4 1.2.1 国外手势识别及相关领域的研究成果 3 HYPERLINK \l _bookmark5 1.2.2 国内手势识别领域的研究成果 4 HYPERLINK \l _bookmark6 1.3 手势识别理论基础 4 HYPERLINK \l _bookmark7 1.3.1 手势的定义和分类 4 HYPERLINK \l _bookmark8 1.3.2 手势识别分类 5 HYPERLINK \l _bookmark9 1.3.3 手势建模 6 HYPERLINK \l _bookmark10 1.3.4 手势分割与跟踪 6 HYPERLINK \l _bookmark11 1.3.5 手势识别 7 HYPERLINK \l _bookmark12 1.4 本文研究内容与结构安排 8 HYPERLINK \l _bookmark13 1.5 本章小结 10 HYPERLINK \l _bookmark14 2 手势分割与特征提取 11 HYPERLINK \l _bookmark15 2.1 图像预处理 11 HYPERLINK \l _bookmark16 2.1.1 图像平滑 11 HYPERLINK \l _bookmark17 2.1.2 色彩空间的转换 14 HYPERLINK \l _bookmark18 2.2 阈值分割 17 HYPERLINK \l _bookmark19 2.3 形态学处理 19 HYPERLINK \l _bookmark20