This project is designed and developed in CVPR Lab, IIT Ropar, India.

Guide : Dr. Subrahmanyam Murala

Sponsored By : Yamaha Labs, India.

In automatomibe industry, engine assembly is considered as one of the crucial part. During engine assenbly various parts assembled together to complate desired two stroke engine. Piston is a part which responsible for the production of stroke which prodce mechanical energy for acceleration. Piston contains various rings depending on different types of engines. During engine assembly, piston without rings can leads to the faulty engine. Due to this it is mandetory to take care of piston rings during engine assembly. Our project aims the real time detection of the rings in piston. Our objective is to develope a small device which real time detect piston rings. We consider the fact that the cost of our device should be less than 20,000 INR.

Design and Developed fully functional dynamic website for CVPR Lab, IIT Ropar.

Effectively use of Django functionality for creating relational database in backend.

Robust and Responsive design which leads to faster and efficient behaviour.

Object Segmentation is done with Background estimation with GAN having two end to end generators.

Video-wise results obtained with proposed architecture of GAN outperforms state-of-the-art methods for moving object segmentation in videos.

Reinforcement Learning (RL) is a subfield of Machine Learning where an agent learns by interacting with its environment, observing the results of these interactions and receiving a reward (positive or negative) accordingly. This way of learning mimics the fundamental way in which we humans (and animals alike) learn. Currently, Machines are not as intelligent as humans in terms of knowledge extraction and new discoveries with using previous learned knowledge. Reinforcement Training is the solution for training of Robots in real world like we humans does right from our childhood. In this project I used Reinforcement approach (Q Learning) to train the Robot so that it should follow line precisely without being explicitely programmed for it.

People with diabetes can have an eye disease known as diabetic retinopathy. This disease is one of the leading cause of blindness among the peoples in united states. Detection of this disease is possible with the neural network over the ratinal image using several algorithms in deep learning. When the retinal image is given to fully trained convolutional neural network it identifies the anomalies in the retinal image which are the leading cause of the diabetic retinopathy. There are several anomalies in the retinal image if the person is sufferers such as macula, exudates, hemorrhage etc. Sometimes an ophthalmologist also not able to detect them. So using Convolutional neural network we can extract the deep feature of an image which correctly identifies infected macula and exudates. To Identified the anomalies in the retinal image I use Faster RNN NAS which segment outs the infected objects in the retinal image.

Forecasting motion of chaotic system such as double pendulum can be significantly distorted because of interpolative nature of various regression algorithm in machine learning. It isn't conceivable to foresee future position of a chaotic system with specifically preparing the relapse calculations. So to defeat the addition of relapse prediction in developing the movement of double pendulum we used the recurrent neural network with the goal that it will extrapolate the outcome and furthermore utilized forecasting using recurrent neural networks.

Predicting or Forecasting path of the particle in Brownian motion/Random motion is one of the most fundamental problems of kinematical physics. In the year 1905, Albert Einstein proposed a theory for Brownian motion, in which he derived an equation for random movement of particles which was purely based on kinematics. This theory required higher level mathematics and physics to deduce factors governing the motion of any random particle. This problem can be redefined by modern methods like machine learning or ANN by considering it as a problem of Regression or Forecasting domain. Our main task is to redefine the problem of Brownian motion to the domain of Forecasting and predicting the path of a random particle moving in a confined space. In Deep Learning domain Recurrent Neural Networks are best suitable for prediction of time series data, So I use RNN network to predict the trajectory of particles in Brownian motion in confined space.

Django is a high-level Python Web framework that encourages rapid development and clean pragmatic design. A Web framework is a set of components that provide a standard way to develop websites fast and easily. Django’s primary goal is to ease the creation of complex database-driven websites. Photographyadda is a social networking website for photographers and photo lovers created in Django. It provides the platform for emerging photographers to boost their skills and share their photos and albums with people.

JavaScript (JS) is a lightweight interpreted or JIT-compiled programming language with first-class functions. While it is most well-known as the scripting language for Web pages, many non-browser environments also use it, such as Node.js, Apache CouchDB and Adobe Acrobat. JavaScript is a prototype-based, multi-paradigm, dynamic language, supporting object-oriented, imperative, and declarative (e.g. functional programming) styles. This game is purely created in JavaScript which uses JavaScript Objects and other building blocks in JavaScript. This is the lightweight game and runs on client side and provide good user interface to the user for controlling the player using a keyboard.