約三萬余字

摘 要
人機(jī)交互（Human-Computer Interaction）主要研究人、計(jì)算機(jī)以及彼此之間互相影響的有關(guān)技術(shù)。視覺語言（包括手勢(shì)和面部表情）作為一種自然的人-人信息交互的輔助方式，具有所見即所得、自然方便等優(yōu)點(diǎn)，因此在新一代“以人為中心”的智能人機(jī)交互中得到了廣泛的關(guān)注。此次課題就是在這一方面做的一些嘗試和探索，其中選擇了基于手勢(shì)的人機(jī)交互方式做研究，由于手勢(shì)具有多樣復(fù)雜性以及時(shí)空的差異性，因此，基于視覺的手勢(shì)識(shí)別是一個(gè)極富挑戰(zhàn)性的多學(xué)科交叉難題。
手勢(shì)一般隱藏了一定的信息，我們需要對(duì)手勢(shì)進(jìn)行識(shí)別，得到其表達(dá)的含義，從而控制人機(jī)交互的實(shí)施。本文實(shí)現(xiàn)了一個(gè)利用視頻手勢(shì)提取進(jìn)行人機(jī)交互的原型系統(tǒng)，大致可分為四個(gè)部分：視頻采集和分離、圖像預(yù)處理、手勢(shì)提取和識(shí)別、人機(jī)交互。其中的關(guān)鍵部分是手勢(shì)的提取和識(shí)別，手勢(shì)提取需要采用圖像處理的方法，而手勢(shì)勢(shì)識(shí)別采用的則是決策樹的方法，通過對(duì)手勢(shì)幾何特征的分析和決策來判定手勢(shì)類型。
采用以上方法對(duì)一段視頻進(jìn)行實(shí)時(shí)處理和分析，進(jìn)而實(shí)現(xiàn)人機(jī)交互。其中主要的交互功能可以實(shí)現(xiàn)，實(shí)時(shí)性也較好，準(zhǔn)確性也較高，不過該方法的魯棒性不是很好，對(duì)光照強(qiáng)度較敏感。

關(guān)鍵詞：人機(jī)交互，手勢(shì)識(shí)別，圖像處理，幾何特征

Abstract
Human—Computer Interaction(HCI) is the technology of researching the interaction of human and computer, Visual language(gesture and face look) is a assistant human to human mutual method, it has the advantage of so called "what you see is what you get", and the most important is that it's a natural way. in view of these advantage, the researchers has put much focus on this aspect. the paper aims at doing some work on it. I select the item of HCI based on gesture recognition as my target. the gesture recognition is a challenging problem owe to the complexity and variety of gestures.
Generally speaking, The gestures contain some information in them, what we need to do is founding it. in this paper, it mainly involved the analyse of static gesture, and we must find the inherent clue, if we get it, then we can make use of it to control the HCI. in this paper ,the whole project can be separate into four parts: they are capturing and separating the video、image Preprocess、 acquiring and recognising the gesture 、the human computer interaction. the key part is acquiring and recognising the gesture. the gesture capture applys the theory of image processing, while gesture recognition applys the theory of decision-tree.
In the experiment, I acquired the video stream from a usb camera and processed it just as aboved approach. the final result is that it performed as we want. in a word, It’s a real-time system, the delay is in the control, and it has high accuracy too, But the robustness of the method is not very good, and it's much sensitive to light intensity.

KEY WORDS:human-comuter interaction hand gesture recognition
Image processing Geometry features


目 錄
第一章 緒論
1.1 人機(jī)交互概述……………………………………………………………… 4
1.2 手勢(shì)識(shí)別概述……………………………………………………………… 5
1.3 國(guó)內(nèi)外研究現(xiàn)狀…………………………………………………………… 6
1.4 本論文所做的工作及論文安排…………………………………………… 7
第二章 基于視覺的手勢(shì)識(shí)別技術(shù)基礎(chǔ)
2.1 手勢(shì)建?！?. 8
2.2 手勢(shì)分割…………………………………………………………………… 11
2.3 手勢(shì)分析…………………………………………………………………… 11
2.4 手勢(shì)識(shí)別…………………………………………………………………… 12
2.5 本章小結(jié)…………………………………………………………………… 14
第三章 基于視頻的人機(jī)交互設(shè)計(jì)
3.1 總體框架設(shè)計(jì)……………………………………………………………… 15
3.2 圖像獲取及預(yù)處理………………………………………………………… 16
3.3 手勢(shì)提取和識(shí)別…………………………………………………………… 23
3.4 人機(jī)交互…………………………………………………………………… 28
3.5 本章小結(jié)…………………………………………………………………… 29
第四章 方案實(shí)現(xiàn)與結(jié)果分析
4.1程序運(yùn)行環(huán)境………………………………………………………………. 30
4.2程序界面設(shè)計(jì)和主要功能…………………………………………………. 31
4.3程序運(yùn)行結(jié)果以及分析……………………………………………………. 36
4.4本章小結(jié)……………………………………………………………………. 37
總結(jié)和展望………………………………………………………. 38
致謝………………………………………………………………. 39
參考文獻(xiàn)…………………………………………………………. 40

參考文獻(xiàn)
[1] 董士海．人機(jī)交互的進(jìn)展及面臨的挑戰(zhàn)，計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào) Vol.16,No.1,Jan.2004
[2] 袁景和．用于人機(jī)交互的手勢(shì)指令識(shí)別技術(shù)：[博士學(xué)位論文]，天津：南開大學(xué)，2002.4
[3] 王茂吉．基于視覺的靜態(tài)手勢(shì)識(shí)別系統(tǒng)：[碩士學(xué)位論文]，哈爾濱：哈爾濱工業(yè)大學(xué)，2006.6
[4] T.Takahashi,F. K. Shino．Hand gesture coding based on experiments using a hand gesture interface device，SIGCHl Bulletin．1991, 23(2):67~73
[5] Davis, M. Shah．Visual gesture recognition. In IEEE Proceeding on Vision-Image Signal Processing.1994, 141(2): 321~332-br..