Biomedical application of image processing

1. DETECTION OF
STRESS USING IMAGE
PROCESSING

2. INDEX
Introduction
Block diagram
Conclusion
Reference
2

3. INTRODUCTION
Stress is a part of life it is an unpleasant state of emotional arousal
that people experience in situations like working for long hours in
front of computer.
Monitoring the emotional status of a person who is working in front
of a computer for longer duration is crucial for the safety of a person
 A camera is used to capture the near frontal view of the person while
he is working in front of the computer. The camera is mounted facing
a person.

4. Video captured is divided into three sections of equal length and set
of equal number of image frames are extracted from each section
correspondingly and are analysed.
The image analysis includes the calculation of the variation in the
position of the eyebrow from its mean position
we employ the technique of deep learning which gives the computer
an ability to learn without being explicitly programmed.

5. BLOCK DIAGRAM

6. Emotional stress detection:
1.Task Definition:
a. Different people behave different under stress.
b. So eyebrow movement which shows rigid transformation
is used as area of intrest in stress recognition.
c. It does not have universal pattern but variation
corresponding to person is used for detection.
2.Methodology:
I.Image preprocessing:
a.Video sequence captured by camera.
b. It includes two transformation of extracted frame.
1.Pixel transformation(Brightness and contrast)
2.Binary Transformation.(Multiplication and addition
with constant)

7. c.RGB to gray conversion.
d.Find the threshold and set that like
if pixel value > threshold =1 and 0 otherwise;

8. 3. Eyebrow detection:
a. Binary image scan from top left through each row
b. 0 –black pixel, recorded with its co ordinates, which is
considered as eyebrow tip coordinate of normalised image.

9. 4. Stress detection:
a. Offline displacement calculation-
Calculate eyebrow position co ordinate obtained in previous state
with the current mean position.
a. Calculation of variance-
Eyebrow transformation in every subsequent image over a period
of time is estimated by calculating its variability.
5. Deep learning:
Submodules are
a.Training dataset ,learning linear regression algorithm and testing
dataset prediction.

11. CONCLUSION
We developed a monitoring system for emotional stress
detection of person working in front of computer. So it is
possible to give pre alarm to the person so that he will remain
healthy.

12. Thank you!