IEEE research paper, ppt for college presentation. I've include almost all aspects of this topic. You can see that if you'd like then please share and like this. Thanks you Business : if anybody wants to have ppt related to any topic contact me. My email Id- manojjha491@live.com Contact no : +91-9867299099

1. A
Seminar Presentation
On
Citizen emotion analysis in smart city
By
Manoj jha
Guided By: Mr. M.A. Bhandari
G.H.RAISONI INSTITUTE OF ENGINEERING AND TECHNOLOGY, WAGHOLI, PUNE,
SAVITRIBAI PHULE PUNE UNEVERSITY

2. Outline
• Introduction
• Methodology for emotion recognition
• Emotion classification
• Required H/W and S/W
• Emotion analysis
• Referred paper Result
• Challenges
• Discussion

3. Introduction
• Human-Computer Interaction
– Speech recognition
– Gesture/Action recognition
– Facial expression recognition
– Emotion recognition
• Storage and Analysis
– Data store from different areas of
city
– Data analysis
– Show on mobile device

4. Methodology For Emotion Recognition
• Acquisition of the signals
• Citizen emotion analysis
• Database storage
• Emotion data mapping

5. Signal Acquisition
• Shimmer Module
• Electrocardiography (ECG)
• Galvanic skin response (GSR)
Source: http://biomedikal.in/2011/05/important-physiological-signals-in-
the-body/

6. The Shimmer3 Module
– SHIMMER (Sensing Health with Intelligence, Modularity, Mobility and
Experimental Reusability) Platform
– The goal of SHIMMER is to provide an extremely compact extensible
platform for long-term wearable sensing .
– a highly extensible wireless sensor platform
• SHIMMER firmware is based on TinyOS
• Data transmit via Bluetooth
• Can sense EMG, ECG, GSR, etc.
• Support Matlab, LabView, Android, C#/.Net etc.http://shimmer.sourceforge.net/
http://www.shimmer-research.com/
Adrian Burns, SHIMMER: An Extensible Platform for Physiological Signal Capture, IEEE
EMBS, 2010

8. Electrocardiography
• Electrocardiography (ECG or EKG*) is the process of recording
the electrical activity of the heart over a period of time using
electrodes placed on the skin.
https://en.m.wikipedia.org/wiki/Electrocardiography

9. The GSR Signal
• Galvanic Skin Response (GSR)
– measuring the electrical conductance of the skin
– due to the response of the skin and muscle tissue to external and
internal stimuli, the conductance can vary by several microsiemens
(unit of ohm).
– GSR is highly sensitive to emotions (fear, anger, startle response,
etc.)http://en.wikipedia.org/wiki/Skin_conductance

11. Citizen emotion analysis
• The citizen emotion analysis starts with the processing of digital
signals that comes from acquisition stage.
• Those acquired signals are transmitted to the smart phone
1)Prerequisite of the analysis
2)Emotion analysis

12. Prerequisite of the analysis
• The emotion analysis consider 8 patterns of emotion mentioned in
flowsense project such as : sad, fear, happiness, surprise, disgust,
anger, boredom and neutral.
• Thus acquired signals are processed and compared with flowsense’s
patterns of emotion, to try to identify the citizen’s emotion.

13. Select city area (A1)
Start App (t0)
Start shimmer (t0)
Finish capturing A1
(t1)
Store A2 App out
Store A2 Shimmer out
Stop App
Stop Shimmer
Finish capturing A2
(t2)
Start App (t0)
Start shimmer (t0)
Select city area (A2)
Citizen emotion
analysis
Store A1+A2 App out
Join Shimmer out
(t1+t2)
Join App out
(t1+t2)
Sum time
(t1+t2)
Store A1 App out
Store A1 Shimmer out
Stop App
Stop Shimmer
Fig4- Flow chart of the emotion acquisition in two city’s areas (A1 and A2)

14. Emotion’s classification
• The emotion classification start with a signal or set of emotions(x(t))
captured by Shimmer sensor
• Based in the cross correlation between two emotion and the emotions
patterns from dataset (e(t)).
• Both emotion are cross-related in order to determine the pattern that
better match with the citizen’s emotion.
• Therefore low value of different samples, represent a high percentage of
similarity.
Where T- Total signal sample
- amount of different
samples
-- percentage of similarity of pattern emotion e(t) to citizen
emotion x(t)

15. • The final algorithm procedure is the selection of the biggest percentage of
similarity
• and this selected percentage, is considered as representation of the
emotion felt by the citizen in a determined city’s area.
• After that, this emotion is stored in database to be used on mapping stage.
Store correlation
percentage
Cross-correlation
x(t).e(t)
Select emotion
e(t)
Citizen
emotion
x(t)
Dataset
Emotion
selected
?
Is the biggest
Percentage of
Similarity (P)?
x(t)=e(t)
No

16. Shimmer3
• This is Shimmer repository for shimmer3 application for more
information about shimmer wireless sensor motes see
https://www.shimmersensing.com
A brief description of contents follows
1.Apps/BT streams – all purpose configurable Bluetooth sensing and streaming application
2.Apps/SDLog - sensing application that saves data to microSD card
3.Apps/log and stream – that simultaneously logs to microSD card while streaming over Bluetooth
4.Firmware identifier list.txt – list of identifier used by applicaton to identify themselves

17. Total Hardware
• neuroLynQ sensor is positioned on the subjects wrist
• 2 GSR electrodes positioned on base of fingers
• 1ECG electrode positioned on subject’s chest
• 1 ECG electrode located on subject’s inner wrist.
Software
• Streamline management of all sensors
• Simultaneous live streaming from up to 36 participant
• Visualization of live streamed GSR and heart rate data at 5Hz
• Event annotation capability

18. Emotion analysis
• Considering the ECG signal from citizen (captured by
Shimmer )
1. A baseline correlation algorithm to normalize and produce a common reference to each part
of the signal.
2. A fourth-order Savitzky-Golay FIR smoothing filter to signal noise attenuation
3. A first order Butterworth filter to eliminate most noise.
• These emotions already sampled, are compared with
emotion from Flowsense project dataset.

19. ECG signal from
Shimmer
Correct baseline
signals
Questionnaire
signal
Signal processed
R-R peak distance
Peak detection
Butterwowrth
filter
Savitzky-Golay
filter FIR
Signal processed
Questionnaire
signal
Questionnaire
signal
Steps of signal processing and emotion identification of a ECG signal

20. Database storage
• In the database storage the information about citizens emotion
are organized within tables in a relational structure, with one
table per citizen information.
• This is used the SQLite, that is an open source SQL database that
stores data to a text file on a smart phone.
Fig- Representation of a format of table used with five columns : citizen id, latitude, longitude,
emotion and date-time

21. Mapping
• Mapping is the last stage of the App
• Depend of the data stored in database
• This emotion representation are based on icon that are plotted
with the Google Maps API, and to each considered city’s area, the
more expressive emotion felt by the citizen will represent the
resultant emotion of that area.
• The size of each icon change and can assume three sizes. This size
are according to the size of the city’s area visited and the App uses
the location information already stored in database, to determine
the ideal size. Not depend on emotions quantity.

22. id Location emotion
01 nn,nn Happy
02 nn,nn Surprise
03 nn,nn Neutral
04 nn,nn Sad
Area
icon
Application
view
Mapping
Happy
Surprise
neutral
Disgust
Boredom
anger
Fear
Sad
Red area
Yellow area
Green area
Fig- relationship between database and mapping process
Fig-Representation of the developed application with city areas and emotion felt
by the citizens

23. Referred paper Result
• The developed App is able to capture the citizens’ emotions
associated with the visited city’s areas.
• These emotions were sampled, processed, classified and
matched with emotions patterns from Flowsense project.
• The APP was tested in laboratory using the same experimental
protocol of flowsense project.
• It was tested with 10 citizens (C=10). Each of these emotion
was compared with all eight emotion from dataset.

25. Challenge
• Emotion signal tend to very noisy.
• Emotion signal generally lacks ground truth and emotion is very
subjective.
• Recognition algorithms on Android devices should be light weight
• Dealing with sequential data

26. References
• A. Solanas et aI., Smart health: a context-aware health
paradigm within smart cities.
• c. Patsakis et aI. , Personalized medical services using smart
cities‘ infrastructures.
• B. Desmet and V. Hoste, Emotion detection in suicide notes.
Expert Systems with Applications.
• S. lalitha et aI. , Emotion detection using MFCC and Cepstrum
features.

28. THANK YOU