See discussions, stats, and author profiles for this publication at: [Link]
net/publication/362531204
FACE RECOGNITION AND FACE EMOTION DETECTION
Article · May 2018
CITATIONS READS
0 87
4 authors, including:
Rohini Khalkar
bvducoep
26 PUBLICATIONS 92 CITATIONS
SEE PROFILE
All content following this page was uploaded by Rohini Khalkar on 06 August 2022.
The user has requested enhancement of the downloaded file.
International Journal of Computer Engineering and Applications,
Volume XII, Issue V, May 18, [Link] ISSN 2321-3469
FACE RECOGNITION AND FACE EMOTION DETECTION
Priyam Kumar Singh1, Sumit Kumar Mehta2, Samarpit Gupta3, Prof. Rohini Khalkar4
1,2,3,4 Department of Computer Engineering , Bharati Vidyapeeth (Deemed to be) University College of
Engineering Pune
1
kumar.priyam55@[Link]
2
sumit.mehta043@[Link]
3
samarguptag@[Link]
4
[Link]@[Link]
ABSTRACT:
The objective of this paper is to recognize face and detect face emotions. The key elements of face are
to determine the face and their emotions. With the help of machine learning algorithm, we are going
to recognize and classify different images of face emotions. The main objective of this project is to
identify faces and their facial expression. We have used Support Vector Machine algorithm to identify
different emotions. We have used Python programming language for the implementation of this
project.
Keywords: Face recognition; Local Binary Patterns (LBP); Support vector machine.
[1] INTRODUCTION
The study of face and its features is a research topic from past few decades. In this project, we
are performing face recognition and emotion detection. The face recognition is done using local Binary
Priyam Kumar Singh, Sumit Kumar Mehta, Samarpit Gupta, Prof. Rohini Khalkar 57
FACE RECOGNITION AND FACE EMOTION DETECTION
Patterns (LBP) algorithm and emotion detection is performed by Support vector machine. We are going
to recognize the face and detect emotions of people whose facial training data is used to train the
Support vector machine model. The software is built with very high accuracy and completely built and
written in Python Programming language.
Step for make this project:
A. Face recognition
The plan for the project development was divided into different categories:
1. First, we have downloaded sample face images from internet assigned a label(name of the person) to
them. This is a training data for our model.
2. Then some samples are used to test our model.
3. Based on the training, the label of the test samples is predicted.
B. Emotion Detection
The plan for the project development was divided into different categories:
1. First, we have downloaded 10k US Adult Faces Database of face and emotion from the dataset
provided by wilmabainbridge and pitt [4][5].
[Link] contains 2,222 target faces. The data is categorized into different emotion labels.
3. We trained our svm model with this data
4. After training, we have tested our model on live camera for detecting emotions.
[2] FACE RECOGNITION
Face recognition is important process to identify the image of the person face .It requires several step
to identify the image of person. The basic architecture of a face recognition system is shown in figure
1 which includes Face Detection, pre-processing, Feature extraction and face recognition block.
Recognition is based on the stored image data of the different image of persons. Input images can be
used for recognition still images. Input Image allows for face detection to detect the image of person.
After the image is extracted and store in database. Store image is used to compare with input image.
After comparing the input image with store image database by applying Local Binary Patterns (LBP)
algorithm and identify the user image in the database and the mostly image corresponding image is
determined.
The algorithm is designed for if the person is recognizing the image for the first time then the system
has to perform each step of operation and performs the result. OpenCV contains cascade classifiers in
which Viola &Jones face detection algorithm is implemented. By using these classifiers the face region
Priyam Kumar Singh, Sumit Kumar Mehta, Samarpit Gupta, Prof. Rohini Khalkar 58
International Journal of Computer Engineering and Applications,
Volume XII, Issue V, May 18, [Link] ISSN 2321-3469
is detected from the image. Face detection is done before face recognition because the accuracy of the
recognition process is a direct function of the accuracy of the detection system. In pre-processing block,
the face image has to minimize the effect of factors that can adversely influence the face recognition
algorithm. The step for Feature Extraction is used for recognition phase are computed.
The face recognition block consists of two stage:
1. Training Process
2. Evaluation Process
The training process, where the Local Binary Patterns (LBP) Algorithm is fed sample of the subject to
be learned and distinct image for each subject is determined. The Evaluation process, where image of a
newly acquired test subject is compared against existing image in the database
Figure-1: Block diagram of face recognition
3. EMOTION DETECTION
The human being use facial expression as important modality to communicate their emotions and to
interact with society. Face emotion detection is done by machine learning system of support vector
machine (SVM) to the recognition and classification of facial expression in still image. The basic
architecture of a emotion Detection system is shown in figure 2 which include Feature Extraction, SVM
training And SVM classification.
For facial expression, a vector of displacements is calculated by taking the Euclidean distance between
landmark locations in a neutral and a peak frame representative of the expression SVM training is applied
to reduce the image [Link] Classification is done to classifying different emotion like Happy,
Priyam Kumar Singh, Sumit Kumar Mehta, Samarpit Gupta, Prof. Rohini Khalkar 59
FACE RECOGNITION AND FACE EMOTION DETECTION
Sad, Angry, Surprise etc. The action unit of face such eye, nose, lips and face contour are responsible
for creation of expression of face. SVMs are maximal margin hyperplane classifiers that exhibit high
classification accuracy for small training sets and good generalization performance on very variable and
difficult to separate data.
Figure-2: Block diagram of emotion detection
[4] RESULTS
The Proposes system is used tested on datasets which consist of wide range of images and different facial
expressions. The application is performed with the help of machine learning concept.
1. Experimental results of face Recognition
For Face Recognition, we have used webcam or from database to capturing the faces. This algorithm is
capable for recognizing different persons in a different window. The recognition should done under
lighting then the recognition rate will be high.
2. Experimental results of Emotion Detection
For Facial expression, we have used webcam to capturing the faces. Gabor is used to extract facial
expression, the experimental result show the detection of different emotions. The implementation is done
by OpenCV and python. Our approach makes no assumptions about the specific emotions used for
training or classification and works for arbitrary, user defined emotion categories.
Priyam Kumar Singh, Sumit Kumar Mehta, Samarpit Gupta, Prof. Rohini Khalkar 60
International Journal of Computer Engineering and Applications,
Volume XII, Issue V, May 18, [Link] ISSN 2321-3469
Table-1: Accuracy of Emotion
Emotion Accuracy
Anger 84.1%
Disgust 83.9%
Fear 76.2%
Happiness 95.3%
Sadness 89.4%
Surprise 98.8%
Neutral 81%
[5] CONCLUSION
Through this paper we performed and built the application for face recognition and emotion detection.
We come to know how to train system and face recognition is implemented for recognize human faces
with their name. Face emotion detection is implemented using the concept of support vector machine
and help to detect the facial expression of human faces. This kind of project is helpful in future in various
field like robotics, artificial intelligence etc.
REFERENCES
[1] Rajesh K M, Naveenkumar M, “An Adaptive-Profile Modified Active Shape Model for Automatic Landmark
Annotation Using Open CV”, International Journal of Engineering Research in Electronic and Communication
Engineering (IJERECE), Vol.3, Issue.5, pp:18-21, May 2016.
[2] Ambika Ramchandra , Ravindra Kumar, “Overview Of Face Recognition System Challenges”, International
journal of scientific technology research , Vol.2,pp: 234-236, August 2013.
Priyam Kumar Singh, Sumit Kumar Mehta, Samarpit Gupta, Prof. Rohini Khalkar 61
View publication stats