CENTRE FOR ARTIFICIAL INTELLIGENCE (CAI)


About Centre


The CENTRE FOR ARTIFICIAL INTELLIGENCE is initiated to acclimatize our budding Engineers and Faculty to the future Technologies of the world, the Centre provides a platform to learn, apply, design, develop and realize the realities of AI in inter-disciplinary domains. This centre provides a platform to explore more research areas and nurture the learning skills among faculty and students of our college.


Objectives of the Centre


• To nurture the development of application platforms based on the industry needs and relevance
• To understand the advanced fields of computer science which involves use of Mathematics, Statistics, Information Technology and Information Sciences in discovering new information and knowledge from large databases and optimize Human effort overall.
• To have a basic skill in a traditional AI language including an ability to write simple to intermediate programs and an ability to understand code written in that language.
• To have a basic understanding of some of the more advanced topics of AI such as learning, natural language processing, agents and robotics, expert systems, and planning.
• To facilitate the students and faculty to learn the Techniques through training programs, workshops, seminars provided by industrial experts, professionals and domain experts in reputed institutes.

Facilities Available

Server Configuration

• Server Board: S2600WFT
• Processor: Intel Xeon Silver4210 CPU @2.20GHZ 2.19GHZ
• RAM: 64 GB HDD :1863 GB
(Change Photos – give front view photos)

System Configuration
• 11th Gen Intel(R) Core(TM) i5-11400 @ 2.60GHz 2.59 GHz
• RAM: 16 GB HDD: 878 GB


Details of Completed/Ongoing Projects


1. SPINOVER (Guest Visit Photos): Spin-over is an autonomous rover designed to follow humans based on their movement. Basic idea of the rover gets the input from the camera of the rover and transport image data through USB bus to raspberry pi. A python program gathers image data (Real time video), Using pretrained model called YOLO V3 Where it detects human movements and gives the commands (left, right, forward, backward) to the motor drivers through raspberry pi. It tracks and follows a human through the Computer Vision (CV) technology. This rover is an Artificial Intelligence (AI) powered autonomous vehicle in which the CV technology itself is supported by the Convolutional Neural Network (CNN) which is a familiar neural structure used in Deep Learning technology. This rover is just a TIER-1 prototype and it will be upgraded further for remote area analysis.






2. Heavy Vehicles Detection using RCNN :We present a classification and detection system of heavy vehicles based on computer vision techniques. The system is designed to automatically gather important statistics for policy makers and regulators in an automated fashion. These statistics include vehicle counting, vehicle type classification. The core of such system is the detection and classification of heavy vehicles in traffic videos. We implement two models for this purpose, first is a MoG & SVM system and the second is based on Faster RCNN, a recently popular deep learning architecture for detection of objects in images. We show in our experiments that Faster RCNN outperforms MoG in detection of vehicles that are static, overlapping or in night time conditions. Faster RCNN also outperforms SVM for the task of classifying vehicle types based on appearances. In the proposed heavy vehicle detection and counting system, the road surface in the image is first extracted and divided into a remote area and a proximal area by a newly proposed segmentation method; the method is crucial for improving vehicle detection. Then, the above two areas are placed into the YOLOv3 network to detect the type of the vehicle. Finally, the vehicle trajectories are obtained by the ORB algorithm, which can be used to judge the driving direction of the vehicle and obtain the number of different vehicles. The experimental results verify that using the proposed segmentation method can provide higher detection accuracy, especially for the detection of Heavy vehicle objects. Moreover, the novel strategy described in this article performs notably well in judging driving direction and counting vehicles.






3. Dystopian:Inspired by the science of learning, the player learns concepts, clarifies their understanding and also develops speed in applying the concepts in this gameplay. Game learns the player’s current level of understanding based on choices made and game complexity is varied to balance the engagement and difficulty level of the game. Simple learning games can be built using any interactive activity framework. Description: Concepts in the Games are expected to be tagged to Nipun Bharat (Foundational Literacy and Numeracy) and NCERT Learning outcome registry(as applicable), DIKSHA telemetry specifications and need to be tagged to relevant NCERT textbooks using DIKSHA dial framework. Games can choose to use DIKSHA’s ECML content framework for interactive content and could leverage robotics, digital toys to add Interactivity, engagement and sophistication. Team can develop relevant plugins for NDEAR DIKSHA content development studio to enhance the interactive experience of game play and contribute to opensource community. More enhanced games could explore multiplayer environments, and leveraging AI/ML to induce engagement and challenge. By carrying out real time analysis, proficiency of users can be made available on dashboard leveraging the NDEAR data framework and tools such as cQube leveraging the learning outcome Tags and proficiency demonstrated by gamers. Games could provide either a PC app, or phone or web-ui based experience






4. Design of Sign Language Visual Chat bot for Challenged Passengers using the ticket vending machine Deaf and dumb people could not interact with people with their sign language,this could cause difficulty in railway station to convey their destination and booking tickets. So, these people have to rely on hand assistive device or they have to depend on their caretaker.Hand assistive devices were available in the market,but still hand assistive device was the only way to communicate with others and this was not cost efficient too.Moreover this device could not interact with automated ticket vending machines in railway station and other necessary places.In order to overcome this difficulties, an automated ticket booking chatbot was designed for speech and hearing impaired persons to convey their destination through hand sign and make their ticket booking much easier way.









Events conducted by the Centre during 2021-2022







Click here to know more...



Enquire Now