EMOTION BASED MUSIC

RECOMMENDATION SYSTEM

Presented By:
Guided by:

KULAKARNI SHRAVYA
206B1A0556
KEERTHANA GALDA 206B1A0525
KOGANTI BHUMIKA 206B1A0551
 DISCUSSION OUTLINE
 Abstract
 Introduction
 Literature Survey
 System Analysis
 Feasibility Study
 System Requirements
 System Design
 DISCUSSION OUTLINE
 Implementation
 System Testing
 Input Design and Output Design
 Screenshots
 Future works
 Conclusion
 Biblography
 ABSTRACT
 The human face plays an important role in knowing an individual's mood.
 Emotion Based Music Recommendation framework that learns the emotion of a users face
directly using a camera.
 The input - human.
 This data - list of songs that comply with the “mood”.
 Time-consuming
 Manually Segregating or grouping songs
 Facial Expression Based Music Player aims at scanning and interpreting the data and accordingly
creating a playlist based the parameters provided.
 INTRODUCTION

 A face detection includes classifying image into two classes:

 One with faces (targets).
 Other containing the background (clutter) which needs to be removed.
 The face detection analysis can be broken into two tasks.
 The first task is a classification task
 Takes some random regions of image as input and outputs a binary value of yes or no,
 The other task is the face localization task
 Tan image as input and output the location of any face or faces within that image as some bounding box/boxes with
(x, y, width, height)
 There are six universal expressions according to Ekman they are fear, disgust surprise, anger, sadness and
happiness.
 We can say a person is happy which can be identified as a gesture of smile by tightened
eyelids and raised lips corners.
 There are basically three steps in expression recognition systems
 Feature learning
 Feature selection
 Classifier construction
 Not only maximizing inter class variation but they also should try to minimize the intra class
variations of expressions. Because same expressions of different individuals in image are far
from each other in pixel’s space so minimizing the intra class variation of expressions is a
problem. Techniques that can be used for facial detection are YOLO, SDD, RCNN, Faster
RCNN.
 LITERATURE SURVEY
 Facial Expression Recognition with Histogram of Oriented Gradients using CNN
Authors: Fayyaz Ali, Sahar Zafar, Irfan Ali, Subhash Guriro, Asif KhanAdnan Zaidi

 Active Clustering with Ensembles for Social structure extraction

Authors: J. R. Barr, L. A. Cament, K. W. Bowyer, and P. J. Flynn

 Authors: Q. Zhu, M.-C. Yeh, K.-T. Cheng, and S. Avidan

 Local directional ternary pattern for facial expression recognition

Authors: Ryu, A. R. Rivera, J. Kim, and O. Chae
SOFTWARE ENVIRONMENT
 Python is a high-level, interpreted scripting language developed in the late 1980s by Guido van
Rossum at the National Research Institute for Mathematics and Computer Science in the
Netherlands.

WHY CHOOSE PYTHON

 Python is Popular
 Python is interpreted
 Python is Simple
 Python use [change / change source]
 SYSTEM ANALYSIS
Existing System
 Using traditional music players,
 a user had to manually browse through his playlist and select songs that would soothe his mood and
emotional experience .
 DISADVANTAGE : Time Consuming.

Propose system
 This is Automated system using AI concept .
 This focuses on detecting human emotions for developing emotion based music player.
 A brief idea is about our systems working, playlist generation and emotion classification.
 FEASIBILITY STUDY
 The feasibility of the project is analyzed in this phase and business proposal is put forth with very
general plan for the project and some cost estimates.
 During system analysis the feasibility study of the proposed system is to Additionally, a very
extensive library of classes and functions is available that provides capability well beyond what is
t into the language, such as database manipulation or GUI programming.
 Three key considerations involved in the feasibility analysis are

 ECONOMICAL FEASIBILITY
 TECHNICAL FEASIBILITY

 SOCIAL FEASIBILITY
 SYSTEM REQUIREMENTS

Hardware requirements
• System : Pentium Dual Core Software Requirements

• Hard Disk : 120 GB. • Operating system : Windows 10

• Monitor : 15’’ LED • Coding Language : python

• Input Devices : Keyboard,• Tool : PyCharm

Mouse
• Database : MYSQL
• Ram : 1 GB
 SYSTEM DESIGN
 System Architecture
 Data Flow Diagram
 USE CASE Diagram
 Sequence Diagram
 IMPLEMENTATION

 Modules
 FACE DETECTION
 The main objective of face detection technique is to identify
the face in the frame by reducing the external noises and other factors.
 The use of this technique is simply to extract features
while reducing the noise and the other factors.
 the process needs to be repeated a few times in order to obtain
a perfect result with a decreased resolution and an increased smoothing level.
 The steps involved in the FACE DETECTION PROCESS are
o 1. Image pyramid
o 2. Histogram of Oriented Gradients
o 3. Linear Classifier
 EMOTION CLASSIFICATION

 When the face is successfully detected, a bounding box will be applied as an

overlay on the image to extract the ROI (face) for further analysis.

 The extracted ROI will next be processed using the “Predictor” function which
is also a called script to extract the 68 facial landmark points and save them in an array.

 A multiclass SVM with a linear kernel is employed to compare

the inputted data with stored one to see in what class (emotion) it belongs.

 If one of the three emotions anger, fear, or surprise is detected

a speed decreasing command will be executed to reduce the speed
of the wheelchair to prevent the user from endangerment.
 MUSIC RECOMMENDATION

o The input is acquired in real-time so the camera is used to capture the video and then the
framing are done.

o The hidden markov model classification are used for processing the framed images.

o The frames that are obtained are considered in all frames and all pixel formats for the purpose
of emotion classification .

o The efficiency of classifier is about 90-95%. so that even when there is any changes in the face
due to environmental conditions the system can still dentify the face and the emotion being
expressed
 SYSTEM TESTING

 The purpose of testing is to discover errors.

 Testing is the process of trying to discover every conceivable fault or weakness in a work product.
 There are various types of testing.
 Unit testing
 Integration testing
 Functional test
 System Test
 White Box Testing
 Black Box Testing
 INPUT DESIGN
 The input design is the link between the information system and the user.

 The input is designed in such a way so that it provides security and ease of use with
retaining the privacy. Input Design considered the following things:
 What data should be given as input?

 How the data should be arranged or coded?

 The dialog to guide the operating personnel in providing input.

 Methods for preparing input validations and steps to follow when error occur.
 OUTPUT DESIGN
 A quality output is one, which meets the requirements of the end user and presents the information
clearly.

 The output form of an information system should accomplish one or more of the following objectives.
 Convey information about past activities, current status or projections of the

 Future.

 Signal important events, opportunities, problems, or warnings.

 Trigger an action.

 Confirm an action.
SCREENSHOTS
UPLODING IMAGE
PROCESSING THE IMAGE
DETECTING EMOTION
PLAY MUSIC
 FUTURE WORKS

 In future, this work shall be extended, and order to extract the important features from images
by combining Local Binary Pattern (LBP) and HOG operator using Deep Learning models.

 The music player based on facial recognition system is highly essential for all the person in
modern day life ecology. This system is further enhanced with benefit able features for
upgrading in future. The methodology of enhancement in the automatic play of songs are done
by detection of the facial expression. The facial expression is detected by programming
interface with the RPI camera. An alternative method, based on additional emotions which is
excluded in our system as disgust and fear. On this emotion included to support the playing of
music automatically.
 CONCLUSION

 In this project, we presented a model to recommend a music based om the emotion based detected
from the facial expression.
 This project proposed designed & developed an emotion based music recommendation system using
face recognition System. Music are the one that has the power to heal any stress or any kind of
emotions. Recent development promises a wide scope in developing emotion based music
recommendation system. Thus the proposed system presents Face based emotion recognition system
to detect the emotions and play music from the emotion detected..
THANK YOU !