CN106897706A - 一种情绪识别装置 - Google Patents

一种情绪识别装置 Download PDF

Info

Publication number
CN106897706A
CN106897706A CN201710122249.3A CN201710122249A CN106897706A CN 106897706 A CN106897706 A CN 106897706A CN 201710122249 A CN201710122249 A CN 201710122249A CN 106897706 A CN106897706 A CN 106897706A
Authority
CN
China
Prior art keywords
expression
micro
histogram
module
identification device
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201710122249.3A
Other languages
English (en)
Other versions
CN106897706B (zh
Inventor
不公告发明人
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Jianxin Intelligent Technology Co ltd
Original Assignee
Shanghai Sail Automation Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Sail Automation Technology Co Ltd filed Critical Shanghai Sail Automation Technology Co Ltd
Priority to CN201710122249.3A priority Critical patent/CN106897706B/zh
Publication of CN106897706A publication Critical patent/CN106897706A/zh
Application granted granted Critical
Publication of CN106897706B publication Critical patent/CN106897706B/zh
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/174Facial expression recognition
    • G06V40/175Static expression
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/21Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/217Validation; Performance evaluation; Active pattern learning techniques
    • G06F18/2193Validation; Performance evaluation; Active pattern learning techniques based on specific statistical tests
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/25Fusion techniques
    • G06F18/254Fusion techniques of classification results, e.g. of results related to same input data
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/443Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
    • G06V10/446Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering using Haar-like filters, e.g. using integral image techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/50Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/46Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/467Encoded features or binary features, e.g. local binary patterns [LBP]

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Human Computer Interaction (AREA)
  • User Interface Of Digital Computer (AREA)

Abstract

一种情绪识别装置,包括表情认知模块、微表情认知模块和加权融合模块,所述表情认知模块用于将表情情绪映射到连续的情感空间进行识别,所述微表情认知模块用于对细微的情绪变化进行识别,所述加权融合模块用于确定情绪状态。本发明的有益效果为:能够对情绪进行准确识别。

Description

一种情绪识别装置
技术领域
本发明创造涉及情绪识别技术领域,具体涉及一种情绪识别装置。
背景技术
随着人工智能科学的飞速发展,如何使计算机能够识别人类的表情进而得到人类的情感状态,己经越来越多地受到计算机科学、心理学等学科的关注。
目前在情感计算领域己出现了诸多情感模型,但大多仅适用于离散状态下的情感计算,对于人与服务机器人的自然交互过程中,认知情感状态的连续时空特性仍无法满足。
发明内容
针对上述问题,本发明旨在提供一种情绪识别装置。
本发明创造的目的通过以下技术方案实现:
一种情绪识别装置,包括表情认知模块、微表情认知模块和加权融合模块,所述表情认知模块用于将表情情绪映射到连续的情感空间进行识别,所述微表情认知模块用于对细微的情绪变化进行识别,所述加权融合模块用于确定情绪状态。
本发明的有益效果为:能够对情绪进行准确识别。
附图说明
利用附图对发明创造作进一步说明,但附图中的实施例不构成对本发明创造的任何限制,对于本领域的普通技术人员,在不付出创造性劳动的前提下,还可以根据以下附图获得其它的附图。
图1是本发明结构示意图。
附图标记:
表情认知模块1、微表情认知模块2、加权融合模块3。
具体实施方式
结合以下实施例对本发明作进一步描述。
参见图1,本实施例的一种情绪识别装置,包括表情认知模块1、微表情认知模块2和加权融合模块3,所述表情认知模块1用于将表情情绪映射到连续的情感空间进行识别,所述微表情认知模块2用于对细微的情绪变化进行识别,所述加权融合模块3用于确定情绪状态;
其中,表情认知模块1:
采用方向梯度直方图对输入图像表情特征进行提取,根据特征维数建立多维表情空间,设表情空间中基本表情Bxi的中心点为Bxic,i=1,2,…,n,表情空间中某表情点s处的该基本表情的势能定义为:
式中,‖·‖表示输入表情与基本表情的欧式距离,α为基本表情的衰减因子;
表情空间中,定义表情点s的势能为:
E(s)=[E(s,Bx1),E(s,Bx2),…,E(s,Bxn)]
式中,Bxi表示第i个基本表情,i=1,2,…,n,E(s)为由表情点s相对各基本表情势能组成的向量,由此确定表情点s的表情。
本优选实施例在实际交流过程中,基于多种基本表情的混合表情不可避免,惊恐的表情便同时具备惊讶和恐惧两种表情的特征,表情识别模块将表情情绪映射到的连续情感空间中,更加符合表情认知的实际情况,这种连续性同时可以使装置情感认知能力得到极大的提高。
优选的,微表情认知模块2:
采用3个尺度,4个方向的Gabor小波对输入图像特征区域特征进行提取,将特征区域划分为m个区域Q0,Q1,…,Qm-1,用直方图统计每个区域的灰度分布属性,具体为:
设图像p(x,y)具有为H个灰度级别,定义图像的直方图为:
zj=ln[∑x,ya×(I{p(x,y)=j}+1)],j=0,1,…,H-1
式中,I{·}表示满足括号中条件则记1,否则记0,a表示直方图放大因子,j代表第j个灰度级别,zj是灰度级为j的像素点的个数;
则从每个区域提取H个灰度级别的直方图可表示为:
其中,GLBP(x,y,α,β)表示Gabor小波提取的特征值采用局部二值算子进行运算的值,j=0,1,…,H-1,k=0,1,…,m-1,α=0,1,2,β=0,1,2,3;
微表情的最终描述可表示为m个区域的直方图序列:
Z=(Z0,0,0,Z0,0,m-1,Z0,1,0,…,Z2,3,m-1)
式中,Z为12×m×H维的特征向量;
假设样本Zi都有其对应的微表情类别,计算待分类微表情直方图序列与已知类别微表情直方图序列的欧式距离,与已知类别微表情直方图序列欧式距离最近的确定为待分类微表情类别C。
微表情既可能包含普通表情的全部肌肉动作,也可能只包含普通表情肌肉动作的一部分,识别过程具有较大的困难,本优选实施例微表情识别模块通过特征区域划分和直方图计算,减少了计算量,提高了识别精度。
优选的,加权融合模块3:
用情绪值来反映情绪,情绪值定义为:
式中,δ1、δ2为权重,δ12=1,E(s,Bxi)表示表情空间中某表情点s处的基本表情Bxi的势能,C(Bxi)=1,表示微表情类别C所属基本表情类别为Bxi
本优选实施例采用基于表情认知结果和微表情认知结果相结合的方法,既从宏观上把握了表情类别,提高了识别效率,又获取了更为细微的情绪变化,识别结果更为准确。
采用本发明情绪识别装置对情绪进行识别,在δ1、δ2不同情况下对200个人的情绪识别情况进行了统计,同现有技术相比,情绪识别准确率和情绪识别速度都有不同程度的提高,产生的有益效果如下表所示:
情绪识别准确率提高 情绪识别速度提高
50% 30%
40% 35%
35% 45%
30% 50%
最后应当说明的是,以上实施例仅用以说明本发明的技术方案,而非对本发明保护范围的限制,尽管参照较佳实施例对本发明作了详细地说明,本领域的普通技术人员应当理解,可以对本发明的技术方案进行修改或者等同替换,而不脱离本发明技术方案的实质和范围。

Claims (6)

1.一种情绪识别装置,其特征在于,包括表情认知模块、微表情认知模块和加权融合模块,所述表情认知模块用于将表情情绪映射到连续的情感空间进行识别,所述微表情认知模块用于对细微的情绪变化进行识别,所述加权融合模块用于确定情绪状态。
2.根据权利要求1所述的一种情绪识别装置,其特征在于,表情认知模块:
采用方向梯度直方图对输入图像表情特征进行提取,根据特征维数建立多维表情空间,设表情空间中基本表情Bxi的中心点为Bxic,i=1,2,…,n,表情空间中某表情点s处的该基本表情的势能定义为:
E ( s , Bx i ) = e - α | | s - Bx i c | |
式中,‖·‖表示输入表情与基本表情的欧式距离,α为基本表情的衰减因子;
表情空间中,定义表情点s的势能为:
E(s)=[E(s,Bx1),E(s,Bx2),…,E(s,Bxn)]
式中,Bxi表示第i个基本表情,i=1,2,…,n,E(s)为由表情点s相对各基本表情势能组成的向量,由此确定表情点s的表情。
3.根据权利要求2所述的一种情绪识别装置,其特征在于,微表情认知模块采用直方图序列对微表情进行描述,进而确定微表情类别。
4.根据权利要求3所述的一种情绪识别装置,其特征在于,所述直方图序列具体为:
采用3个尺度,4个方向的Gabor小波对输入图像特征区域特征进行提取,将特征区域划分为m个区域Q0,Q1,…,Qm-1,用直方图统计每个区域的灰度分布属性,具体为:
设图像p(x,y)具有为H个灰度级别,定义图像的直方图为:
zj=ln[∑x,ya×(I{p(x,y)=j}+1)],j=0,1,…,H-1
式中,I{·}表示满足括号中条件则记1,否则记0,a表示直方图放大因子,j代表第j个灰度级别,zj是灰度级为j的像素点的个数;
则从每个区域提取H个灰度级别的直方图可表示为:
Z α , β , Q k = l n [ Σ ( x , y ) ∈ Q k a × ( I { G L B P ( x , y , α , β ) = j } + 1 ) ]
其中,GLBP(x,y,α,β)表示Gabor小波提取的特征值采用局部二值算子进行运算的值,j=0,1,…,H-1,k=0,1,…,m-1,α=0,1,2,β=0,1,2,3;
微表情的最终描述可表示为m个区域的直方图序列:
Z=(Z0,0,0,Z0,0,m-1,Z0,1,0,…,Z2,3,m-1)
式中,Z为12×m×H维的特征向量;
假设样本Zi都有其对应的微表情类别,计算待分类微表情直方图序列与已知类别微表情直方图序列的欧式距离,与已知类别微表情直方图序列欧式距离最近的确定为待分类微表情类别C。
5.根据权利要求4所述的一种情绪识别装置,其特征在于,加权融合模块采用情绪值确定情绪状态。
6.根据权利要求5所述的一种情绪识别装置,其特征在于,所述情绪值定义为:
Y = δ 1 × Σ i = 1 n E ( s , Bx i ) Σ i = 1 n [ E ( s , Bx i ) ] [ E ( s , Bx i ) ] + δ 2 × E ( s , Bx i ) C ( Bx i )
式中,δ1、δ2为权重,E(s,Bxi)表示表情空间中某表情点s处的基本表情Bxi的势能,C(Bxi)=1,表示微表情类别C所属基本表情类别为Bxi
CN201710122249.3A 2017-03-02 2017-03-02 一种情绪识别装置 Active CN106897706B (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710122249.3A CN106897706B (zh) 2017-03-02 2017-03-02 一种情绪识别装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710122249.3A CN106897706B (zh) 2017-03-02 2017-03-02 一种情绪识别装置

Publications (2)

Publication Number Publication Date
CN106897706A true CN106897706A (zh) 2017-06-27
CN106897706B CN106897706B (zh) 2019-11-22

Family

ID=59184527

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710122249.3A Active CN106897706B (zh) 2017-03-02 2017-03-02 一种情绪识别装置

Country Status (1)

Country Link
CN (1) CN106897706B (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107705808A (zh) * 2017-11-20 2018-02-16 合光正锦(盘锦)机器人技术有限公司 一种基于面部特征与语音特征的情绪识别方法
CN108261178A (zh) * 2018-01-12 2018-07-10 平安科技(深圳)有限公司 动物疼痛指数判断方法、装置及存储介质
CN108577866A (zh) * 2018-04-03 2018-09-28 中国地质大学(武汉) 一种多维情感识别与缓解的系统及方法
CN109830280A (zh) * 2018-12-18 2019-05-31 深圳壹账通智能科技有限公司 心理辅助分析方法、装置、计算机设备和存储介质
CN111143615A (zh) * 2019-12-12 2020-05-12 浙江大学 一种短视频情感类别的识别装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007156688A (ja) * 2005-12-02 2007-06-21 Mitsubishi Heavy Ind Ltd ユーザ認証装置およびその方法
CN102629321A (zh) * 2012-03-29 2012-08-08 天津理工大学 基于证据理论的人脸表情识别方法
CN105740948A (zh) * 2016-02-04 2016-07-06 北京光年无限科技有限公司 一种面向智能机器人的交互方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007156688A (ja) * 2005-12-02 2007-06-21 Mitsubishi Heavy Ind Ltd ユーザ認証装置およびその方法
CN102629321A (zh) * 2012-03-29 2012-08-08 天津理工大学 基于证据理论的人脸表情识别方法
CN105740948A (zh) * 2016-02-04 2016-07-06 北京光年无限科技有限公司 一种面向智能机器人的交互方法及装置

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107705808A (zh) * 2017-11-20 2018-02-16 合光正锦(盘锦)机器人技术有限公司 一种基于面部特征与语音特征的情绪识别方法
CN108261178A (zh) * 2018-01-12 2018-07-10 平安科技(深圳)有限公司 动物疼痛指数判断方法、装置及存储介质
CN108261178B (zh) * 2018-01-12 2020-08-28 平安科技(深圳)有限公司 动物疼痛指数判断方法、装置及存储介质
CN108577866A (zh) * 2018-04-03 2018-09-28 中国地质大学(武汉) 一种多维情感识别与缓解的系统及方法
CN109830280A (zh) * 2018-12-18 2019-05-31 深圳壹账通智能科技有限公司 心理辅助分析方法、装置、计算机设备和存储介质
CN111143615A (zh) * 2019-12-12 2020-05-12 浙江大学 一种短视频情感类别的识别装置
CN111143615B (zh) * 2019-12-12 2022-12-06 浙江大学 一种短视频情感类别的识别装置

Also Published As

Publication number Publication date
CN106897706B (zh) 2019-11-22

Similar Documents

Publication Publication Date Title
CN106897706A (zh) 一种情绪识别装置
Day et al. Deep learning for sentiment analysis on google play consumer review
CN106529503B (zh) 一种集成卷积神经网络人脸情感识别方法
CN109992779B (zh) 一种基于cnn的情感分析方法、装置、设备及存储介质
Chen et al. A SA-ANN-based modeling method for human cognition mechanism and the PSACO cognition algorithm
Chen et al. Adaptive feature selection-based AdaBoost-KNN with direct optimization for dynamic emotion recognition in human–robot interaction
CN110472042A (zh) 一种细粒度情感分类方法
Jiang et al. An eight-layer convolutional neural network with stochastic pooling, batch normalization and dropout for fingerspelling recognition of Chinese sign language
Ruiz-Garcia et al. Deep learning for emotion recognition in faces
CN110083770A (zh) 一种基于特征层面更深的自注意力网络的序列推荐方法
Gaddam et al. Human facial emotion detection using deep learning
CN110399821A (zh) 基于人脸表情识别的顾客满意度获取方法
Zhang et al. Sparse Auto-encoder with Smoothed l 1 Regularization
Li et al. Mining online reviews for ranking products: A novel method based on multiple classifiers and interval-valued intuitionistic fuzzy TOPSIS
CN111985532A (zh) 一种场景级上下文感知的情感识别深度网络方法
Bhattacharya A survey on: facial expression recognition using various deep learning techniques
Yang et al. Facial expression recognition based on arousal-valence emotion model and deep learning method
CN117764227A (zh) 一种用于加油站的客户流失预测装置
Zhong et al. Face expression recognition based on NGO-BILSTM model
Liu Art painting image classification based on neural network
Padalkar et al. Using Logic Programming and Kernel-Grouping for Improving Interpretability of Convolutional Neural Networks
Zhou et al. Conditional generative adversarial networks for domain transfer: a survey
Lin et al. Real emotion seeker: recalibrating annotation for facial expression recognition
CN106914903A (zh) 一种面向智能机器人的交互系统
Rahman et al. Convolutional Neural Networks based multi-object recognition from a RGB image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right
TA01 Transfer of patent application right

Effective date of registration: 20191024

Address after: 236700 No. 12 Liu Xiaozhai, Chundian Community, Chengguan Town, Lixin County, Bozhou City, Anhui Province

Applicant after: Lixin chengchuang Technology Intermediary Service Co.,Ltd.

Address before: 200000 Shanghai, Songjiang District Road, No. 139 Mei Rong, building 1

Applicant before: SHANGHAI FANYU AUTOMATION TECHNOLOGY CO.,LTD.

GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20221019

Address after: 236700 Liu Xiaozhai 12, Chun Dian community, Chengguan Town, Lixin County, Bozhou, Anhui

Patentee after: Zhiyunbao (Bozhou) Technology Development Co.,Ltd.

Address before: 236700 Liu Xiaozhai 12, Chun Dian community, Chengguan Town, Lixin County, Bozhou, Anhui

Patentee before: Lixin chengchuang Technology Intermediary Service Co.,Ltd.

TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20231228

Address after: Room 311, Accelerator Production Workshop, Building B-2, Haichuang Technology Industrial Park, No. 627 Lugu Avenue, Changsha High tech Development Zone, Changsha City, Hunan Province, 410000

Patentee after: HUNAN JIANXIN INTELLIGENT TECHNOLOGY Co.,Ltd.

Address before: 236700 Liu Xiaozhai 12, Chun Dian community, Chengguan Town, Lixin County, Bozhou, Anhui

Patentee before: Zhiyunbao (Bozhou) Technology Development Co.,Ltd.