A Smart Stress Identification System

Project Title

A Smart Stress Identification System

Project Area of Specialization Artificial IntelligenceProject Summary

The project is a non-invasive system which diagnoses stress in a person by detecting his iris. The iris is the colored portion of the human eye around the pupil. Whenever a person suffers stress, rings or curves are found in his iris. So by examining the rings in a person's iris, one can easily identify the stress in that person. Our system takes real-time iris images, and based on machine-learning techniques, it identifies the rings in the iris and thus we can get the clue regarding the person's stress status.

Project Objectives

The system will non-invasively diagnose stress and tension in a person accurately. 

Project Implementation Method

The SSSI performs features extraction, features labeling
and classification. The algorithm applies the Canny edge
detector to extract the edges and then Hough line extraction to identify curves or rings inside the iris. Later, Support Vector
Machine is used that takes the line features from specific iris
segments and label the features.

The SSSI algorithm performs training and
testing. During training, the SSSI algorithm takes trained datasets of 50 healthy persons and 50 stress patients, and extracts features from each data-set and label them as trained features.
During testing mode, the algorithm takes real-time iris image
from a subject, extracts features, compares extracted features with the trained features and classifies the subject as healthy or suffering mental stress.

Benefits of the Project

• Non-invasive (No contact between body and system)
• Cheap
• Fast and efficient
• Accurate
• Diagnosis in early stages

Technical Details of Final Deliverable

In the current project, three clusters
based high performance computing system were utilized. Each cluster utilizes a multi-RISC processor and a GPU-accelerated core,which comprises General Purpose (Intel/AMD) multi-core
processor and Graphical Processing Unit GPU. Each cluster
uses an Intel Xeon X5550 general purpose processor and
Nvidia GTX1080 GPU having 2560 Cuda cores.

Final Deliverable of the Project HW/SW integrated systemType of Industry Medical , Health Technologies Artificial Intelligence(AI)Sustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	70000
Iris Camera	Equipment	1	25000	25000
High Performance Single Board Computer	Equipment	1	35000	35000
LCD 5 inch Screen	Equipment	1	10000	10000