本文關(guān)鍵詞： 語音特征 表情特征 融合算法 支持向量機(jī) 參數(shù)優(yōu)化　出處：《華東理工大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：單模態(tài)情感識別由于受到單一模態(tài)情感特征的限制,識別率并沒有得到較大的提高。近年來,多模態(tài)情感識別突破了這一限制,在情感識別過程中,引入了多種模態(tài)的情感特征進(jìn)行融合,從而在識別率上有了較大的提高。 目前,多模態(tài)情感識別的方法和思路主要有判決層融合與特征層融合。本文采用特征層融合的方式,提取人臉表情特征和語音情感特征,然后根據(jù)兩種模態(tài)情感特征的特點,進(jìn)行特征優(yōu)化處理,最后設(shè)計分類器進(jìn)行情感的分類。本文選擇本課題組自建的情感數(shù)據(jù)庫作為課題研究的數(shù)據(jù),該數(shù)據(jù)庫包含語音、表情和腦電三種模態(tài)的情感數(shù)據(jù),情感類別有7種,即生氣、厭惡、害怕、高興、中性、悲傷和驚奇。 本文的主要研究工作有： (1)語音情感特征提取,本文采用不同的語音特征提取方法(14維特征和74維特征),提取包括短時能量、基音頻率、第一共振峰、美爾頻率倒譜系數(shù)(MFCC)和語音持續(xù)時間等特征類別,同時計算了這些特征類別相關(guān)的統(tǒng)計參數(shù),并以這些特征作為語音情感特征數(shù)據(jù)用于情感識別。 (2)人臉表情特征提取,本文提出了改進(jìn)的局部二值模式(LBP)表情特征提取算法主要提取人臉的眼睛和嘴巴兩個部位的紋理特征。該算法的目的在于保證表情識別率,同時盡可能地降低特征數(shù)據(jù)的維數(shù),減少計算量。 (3)語音與表情特征的融合,本文根據(jù)語音和表情的情感特征,提出了語音與表情特征的直接融合算法和語音與表情特征的融合優(yōu)化算法。語音與表情特征的直接融合算法主要解決兩種模態(tài)特征維數(shù)上差異；語音與表情特征的融合優(yōu)化算法考慮兩種模態(tài)特征的聯(lián)系與差異,提出先融合,后利用主成分分析(PCA)方法進(jìn)行降維優(yōu)化處理,再進(jìn)行情感分類。 (4)雙模態(tài)情感識別,本文采用支持向量機(jī)(SVM)算法進(jìn)行情感識別的仿真實驗。該算法對小樣本、非線性分類問題具有很強(qiáng)的分類能力。在SVM參數(shù)優(yōu)化問題中,本文提出了改進(jìn)的網(wǎng)格搜索參數(shù)優(yōu)化算法,該算法基本思想是先通過基本的網(wǎng)格搜索算法進(jìn)行粗搜,確定參數(shù)的范圍,然后再在此范圍內(nèi)進(jìn)行精搜,找到最優(yōu)識別率的參數(shù)組合。仿真實驗驗證了上述算法的有效性。
[Abstract]:Single modal emotion recognition due to single modal emotion feature constraints, the recognition rate has not been greatly improved. In recent years, multimodal emotion recognition breaks through the limit, in the emotion recognition process, introduces the affective characteristics of multi modality fusion, and the recognition rate is improved greatly.
At present, methods and ideas of multimodal emotion recognition are the main decision level fusion and feature level. This paper uses the feature fusion method, facial expression feature extraction and speech emotion features, and then according to the characteristics of two kinds of modal emotion features, feature classification optimization, finally classifier is designed. This paper chooses this topic of emotion group self emotion database as the research data, the database contains data of three modes of speech, emotion expression and EEG, emotion category has 7 kinds, namely, anger, disgust, fear, happy, neutral, sadness and surprise.
The main research work of this article is as follows:
(1) speech feature extraction, the method for extracting speech features different (14 dimension and 74 dimension), including short-time energy, pitch frequency extraction, first formant, Mel frequency cepstrum coefficient (MFCC) and the duration of speech feature categories, statistical parameters. These features also related categories, and use these features as speech emotion feature data for emotion recognition.
(2) facial expression feature extraction, this paper proposes a local two value model (LBP) improved facial feature extraction texture feature extraction algorithm mainly face the eyes and mouth of two parts. The purpose of the algorithm is to ensure the recognition rate, at the same time as far as possible to reduce the dimension of data, reduce the amount of computation.
(3) the integration of voice and facial expression features, according to the characteristics of speech and emotion expression, proposed fusion algorithm directly based on voice and expression characteristics of the algorithm and speech and expression characteristics. Direct speech and expression feature fusion algorithm mainly solves two modal feature dimension difference; speech and expression feature fusion optimization considering the links and differences between the two algorithms of modal characteristics of the proposed fusion, after using principal component analysis (PCA) to reduce the dimension optimization processing method, then sentiment classification.
(4) bimodal emotion recognition, this paper uses support vector machine (SVM) algorithm simulation of emotion recognition. The algorithm of small sample, nonlinear classification problem has strong ability of classification. In the SVM parameter optimization problems, this paper proposes parameter optimization algorithm of improved grid search algorithm is the basic idea. The basic grid search algorithm for rough search, to determine the range of parameters, and then this range of precision search, find the optimal parameters recognition rate. Simulation results verify the effectiveness of the algorithm.

【學(xué)位授予單位】：華東理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TN912.34;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王玉潔,王志良,陳鋒軍,王國江,王玉鋒;基于隱馬爾可夫模型的情感建模[J];北京農(nóng)學(xué)院學(xué)報;2005年01期

2 韓一;王國胤;楊勇;;基于MFCC的語音情感識別[J];重慶郵電大學(xué)學(xué)報(自然科學(xué)版);2008年05期

3 余伶俐;蔡自興;陳明義;;語音信號的情感特征分析與識別研究綜述[J];電路與系統(tǒng)學(xué)報;2007年04期

4 黃程韋;金峗;王青云;趙力;鄒采榮;;基于語音信號與心電信號的多模態(tài)情感識別[J];東南大學(xué)學(xué)報(自然科學(xué)版);2010年05期

5 余華,王治平,趙力;語音信號中情感特征的分析和識別[J];電聲技術(shù);2004年03期

6 秦宇強(qiáng);張雪英;;基于SVM的語音信號情感識別[J];電路與系統(tǒng)學(xué)報;2012年05期

7 薛為民,石志國,谷學(xué)靜,王志良;基于Agent的人機(jī)情感交互系統(tǒng)研究[J];計算機(jī)工程與應(yīng)用;2002年19期

8 張迎輝;林學(xué)娋;;情感可以計算——情感計算綜述[J];計算機(jī)科學(xué);2008年05期

9 王志良,解侖,董平;情感計算數(shù)學(xué)模型的研究初探[J];計算機(jī)工程;2004年21期

10 谷學(xué)靜,石志國,王志良;基于BDI Agent技術(shù)的情感機(jī)器人語音識別技術(shù)研究[J];計算機(jī)應(yīng)用研究;2003年04期

，

本文編號：1511078