【摘要】：作為新一波科技浪潮的排頭兵,人工智能正以前所未有的速度滲透到人類生活的方方面面。其中,人機(jī)交互技術(shù)作為人工智能領(lǐng)域的重要組成部分,受到廣泛的關(guān)注。在眾多的人機(jī)交互手段中,手勢(shì)交互是最接近人類交流習(xí)慣也是最自然的一種交互方式,相關(guān)手勢(shì)識(shí)別技術(shù)可以被用于聾啞人教學(xué)、智能家居和虛擬現(xiàn)實(shí)等應(yīng)用場(chǎng)合,具有廣泛的應(yīng)用前景。在上述背景下,本文對(duì)基于視覺(jué)的靜態(tài)及動(dòng)態(tài)手勢(shì)識(shí)別問(wèn)題進(jìn)行了重點(diǎn)研究,取得了一些富有實(shí)際意義的研究成果。本文的主要工作與創(chuàng)新點(diǎn)如下:1.深入研究了靜態(tài)手勢(shì)識(shí)別問(wèn)題。針對(duì)傳統(tǒng)的手勢(shì)檢測(cè)方法不能對(duì)前臂、手掌和手指區(qū)域進(jìn)行很好的區(qū)分,導(dǎo)致手勢(shì)識(shí)別效果低下的問(wèn)題,提出了一種有效的、基于直線檢測(cè)的冗余手臂去除方法。實(shí)驗(yàn)結(jié)果驗(yàn)證了方法的有效性。2.現(xiàn)有的靜態(tài)手勢(shì)識(shí)別算法大都首先利用形狀分解方法提取手指特征,然后利用模板匹配技術(shù)實(shí)現(xiàn)對(duì)手勢(shì)的分類。因此,手指檢測(cè)算法性能的好壞會(huì)對(duì)整個(gè)系統(tǒng)的識(shí)別性能產(chǎn)生直接影響。為此,本文從以下三個(gè)方面對(duì)手指檢測(cè)與識(shí)別算法進(jìn)行了改進(jìn):(1)提出了一種新的融合形態(tài)學(xué)處理和曲率信息的手指區(qū)域分割算法:(2)提出了一種基于多參數(shù)的改進(jìn)相似性度量方法;(3)提出了一種基于分層模板匹配的手勢(shì)識(shí)別方法。實(shí)驗(yàn)結(jié)果表明,本文所提出的手勢(shì)檢測(cè)與識(shí)別方法能有效克服雜亂背景、類膚色區(qū)域等不利因素的影響,取得較為理想的檢測(cè)與識(shí)別效果。3.提出了一種基于多卷積神經(jīng)網(wǎng)絡(luò)融合的動(dòng)態(tài)手勢(shì)識(shí)別方法。該方法從給定的深度圖像序列出發(fā),首先提取運(yùn)動(dòng)信息,然后將其送入到不同結(jié)構(gòu)的卷積神經(jīng)網(wǎng)絡(luò)以預(yù)測(cè)相關(guān)的三維時(shí)序信息,據(jù)此可以從空間和時(shí)間的維度去捕捉連續(xù)運(yùn)動(dòng)特征,進(jìn)而實(shí)現(xiàn)對(duì)動(dòng)態(tài)手勢(shì)的分類。定性和定量的實(shí)驗(yàn)結(jié)果驗(yàn)證了本文所提出的動(dòng)態(tài)手勢(shì)識(shí)別算法的性能。
[Abstract]:As the vanguard of a new wave of science and technology, artificial intelligence is permeating every aspect of human life at an unprecedented speed. Among them, as an important part of artificial intelligence field, human-computer interaction technology has received extensive attention. Among the many human-computer interaction methods, gesture interaction is the most close to human communication habits and the most natural way of interaction. Related gesture recognition technology can be used in deaf and mute people teaching, smart home and virtual reality and other applications. It has wide application prospect. Under the above background, this paper focuses on static and dynamic gesture recognition based on vision, and obtains some meaningful research results. The main work and innovation of this paper are as follows: 1. The problem of static gesture recognition is studied in depth. Aiming at the problem that the traditional hand gesture detection method can not distinguish the forearm, palm and finger regions well, which leads to the low performance of gesture recognition, this paper proposes an effective method for removing redundant arms based on line detection. The experimental results show that the method is effective. Most of the existing static gesture recognition algorithms first use shape decomposition method to extract finger features and then use template matching technology to achieve gesture classification. Therefore, the performance of finger detection algorithm will have a direct impact on the recognition performance of the whole system. To that end, This paper improves the algorithm of finger detection and recognition in the following three aspects: (1) A new algorithm of finger region segmentation based on morphological processing and curvature information is proposed; (2) an improved similarity based on multiple parameters is proposed. (3) A method of gesture recognition based on hierarchical template matching is proposed. The experimental results show that the proposed gesture detection and recognition method can effectively overcome the influence of clutter background, skin color region and other adverse factors, and achieve a more ideal detection and recognition effect. 3. A dynamic gesture recognition method based on multi-convolution neural network fusion is proposed. In this method, the motion information is extracted from the given depth image sequence, and then sent to the convolutional neural network with different structures to predict the related 3D temporal information. The continuous motion features can be captured from the dimension of space and time, and the classification of dynamic gestures can be realized. The qualitative and quantitative experimental results verify the performance of the proposed dynamic gesture recognition algorithm.
【學(xué)位授予單位】：中國(guó)科學(xué)技術(shù)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 劉淑萍;劉羽;於俊;汪增福;;結(jié)合手指檢測(cè)和HOG特征的分層靜態(tài)手勢(shì)識(shí)別[J];中國(guó)圖象圖形學(xué)報(bào);2015年06期

相關(guān)博士學(xué)位論文 前1條

1 覃文軍;基于視覺(jué)信息的手勢(shì)識(shí)別算法與模型研究[D];東北大學(xué);2010年

相關(guān)碩士學(xué)位論文 前1條

1 趙亞飛;基于視覺(jué)的手勢(shì)識(shí)別技術(shù)研究[D];浙江大學(xué);2011年

，

本文編號(hào)：2168768