One Day Technical Seminar “Recognition with Deep Learning – an AI Approach”

Date : 20-08-2019
Time: 09:50 AM – 12:00 Noon

Resource Person:
Dr. Mohamed Uvaze Ahamed Ayoobkhan
Assistant Professor and Head
Department of Information Technology
Qala University College, Kurdisthan, Iraq
{& Assistant Professor (Part-time) in Cihan University, Kurdisthan, Iraq}

The Department of Computer Science and Engineering of Jansons Institute of Technology organized a One day Technical Seminar on Recognition with Deep Learning – an AI approach on 20th August 2019 at Sivaganga Auditorium.

The speaker showed videos of Boston Dynamics how Machine Learning and Deep Learning were implemented on it and how it evolves in our day to day life. He also stated various real time examples on clustering and classification in the Deep Learning. It ignites our students’ knowledge and they raised questions to the speaker who keenly answered all the questions with patience.

Later the students’ were taught about the computer vision and its applications such as Landmark Identification, Scene Recognition, Robotics and so on. Two approaches that were used in Convolutional Neural Network (CNN) in the concept of Deep Leaning were also shared by the speaker to the students’. He also gave a very good example about SOPHIA, a social humanoid robot developed by Hong Kong based company Hanson Robotics which was activated on 14th February 2016, and made its first public appearance at South by South West Festival (SXSW) in mid-March 2016 in Austin, Texas, United States.He also displayed a few videos about AI and today’s technology which inspired the students’ and made the session much interesting.

He also shared his own research experience titled “Automated Detection of Exudates in colored Retinal Images for diagnosis of diabetic Retinopathy”. This paper presents a method for detection and classification of exudates in colored retinal images, a real time mobile phone-based gaze tracking and eye-blink detection system on Android platform. The eye-blink detection scheme is developed based on the time difference between two open eye states. The students interacted with the speaker by raising questions such as what type of classification algorithm he used, how to load the image dataset values into the system etc., and it were clarified by him.

The Seminar was a very good experience for the students and they also gave positive feedback about the seminar. The feedback received was very commendable in all aspects.