Project School started this year, as one of the basic pillars of Learning
The ultimate objective of Project School is to provide students with a well-rounded skill set that includes hands-on experience with end-to-end projects, as well as an understanding of the integration of multiple technologies into a single frame.
Eligibility of Project School is open to all students, with selection based on their performance in a machine test on programming languages. Seats are limited and will be allotted based on test scores and availability. Once selected, students are free to form their own teams of 5 members.
Resources allocated for the project:
To ensure that the projects are completed to the highest standard, students will be given time and space to discuss and review their work.
This review process will be designed through tasks and reading assignments that guide students through different methods of problem-solving and help them to finish their projects successfully.
The venue for the project school will be the Virtusa Excellence Centre, where students will have access to the necessary resources and support to complete their projects.
It is our aim that every student in project school will complete 4 different projects, gaining a
strong understanding of the latest technologies and industry practices. We are confident that this initiative will equip our students with the necessary skills and knowledge to secure high-paying jobs with confidence
Details about Project School:
1. Students selection starts from the first semester of the second year.
2. Students have to complete one project per semester per student.
3. Every 10 students per project (2 teams of 5 each).
4. Every project has a mentor.
5. Students have to select their project domain in AI / ML, AWS, Cyber Security, Blockchain, IoT,
Unreal, YOLO v4, Etc.,
6. The project review will be done weekly basis only.
7. Project School Process Flow
Project School Project Titles:
- AutoSurgery: AI-driven auto instrument and operation procedure detection for surgeries
- Earth Observatory Data Cube for Nizamabad district
- Command and Control movements of a drone using Alexa
- FabricRealEstate: HyperledgerFabric driven RealEstate Procurement System
- SwachCampus – Litter Detection and Reporting System
- False Data Injection Attacks in the Internet of Things and Deep Learning enabled Predictive Analytics
- Aves (Bird species) Detecting Mobile application
- SwachCampus – Litter Detection and Reporting System
- Remote Monitoring and Management of Drip Irrigation using Digital Twin Technology:
Details of a few present semester projects
- Swatch campus – litter detection and reporting system
- Remote Monitoring and Management of Drip Irrigation using Digital Twin Technologies
Aves Bird Species Detection
● To Design a responsive mobile application for Bird species identification.
● To help user to detect the species of birds.
● It also helps the user listen to the sound made by the bird.
● Bird species are identified by using the inceptionV3 Deep Learning Model.
● The frontend provides users with information and predicted results which is developed in
Flutter using Dart.
● Firebase is used as a database.
● The deep learning model is trained using the Tensor-Flow library.
● A web viewer option is used to fetch information and an audio picker is used to pick the audio.
Name: – BIRDS 450 SPECIES
Source: – Kaggle
Link: – https://www.kaggle.com/datasets/gpiosenka/100-bird-species
● This Data set is of 450 bird species. 70,626 training images, 22500 test images (5 images per
species), and 2250 validation images (5 images per species). The data set also includes a file
birds.csv. This CSV file contains 5 columns. The file paths column contains the relative file
path to an image file. The labels column contains the bird species class name associated with
the image file. The scientific label column contains the Latin scientific name for the image.
The data set column denotes which dataset (train, test, or valid) the file path resides in. The
class_id column contains the class index value associated with the image file’s class.
Model name: – InceptionV3
● Android Studio for Front-end development and Back-end development.
● Kaggle for Dataset
● Google Colab and Jupyter Notebook for model building
● Name: – DESKTOP-SF7E57H
● Processor: – Intel(R) Core (TM) i7-10750H CPU @ 2.60GHz 2.59 GHz
● RAM: – 16.0 GB (15.8 GB usable)
This bird species identification application is a project that uses machine learning
techniques to identify different bird species along with audio from images. The overall goal
of this project is to create a system that can accurately classify other bird species based on
their visual characteristics.
ePashudhan-an online Portal for Dairy Farmers:
To Design a responsive Mobile application for the online registration of dairy farmers
producing milk and other livestock products organically. This portal shall help dairy farmers to gain more information about the livestock
and their produce, dairy market trends, and many more.
● Full stack project was done with MERN stack.
● The user interface provides farmers with information and predicted results which are developed in React Native.
● Backend development is done in Express.js and Node.js runtime environments.
● MongoDB is used as a database.
● Database:- Any credentials used to Register to the project will be saved in MongoDB.
● Login: Credentials are verified by checking the credentials present in the database.
● Visual Studio Code for Front-end and Back-end development.
● MongoDB Atlas Cluster used for database storage
● Name:- Lenovo Ideapad 330
● Processor :- Intel Core i3 7020U
● RAM:- 4GB Unified Memor
The mobile application for online processing and registration of dairy farmers producing milk
and other livestock products organically. So that this portal shall help dairy farmers to gain
more information about livestock and their produce, dairy market trends, and many more.
False Data Injection attacks:
IoT is the latest industrial revolution primarily merging automation with advanced
manufacturing. But IoT is vulnerable to cyber attacks like false data injection. In this project, I
created a MERN stack web app that helps us to analyze various predictive models using
LTSM, GRU, and CNN. The false data is injected using Python/MATLAB.
This application deals with the problem of developing a Predictive Maintenance system
with the use of Machine Learning (SVM, kNN, Random Forests, and Logistic Regression using Auto Encoders) to classify the life cycle of a sensor. The proposed system consists of mainly three phases:
> The first phase (i.e., pre-processing), the next phase (i.e., model training), and the final phase (i.e., classification). The first phase includes the calculation of the EOL (End of Life) of sensors used and using that I am calculating LR (Life Ratio).
>Then, using LR I am creating labels. These are then classified into three categories:
Good: Good Condition represents 0.
Moderate: Moderate Condition represents 1.
Warning: Warning Condition represents 2.
The technologies used in this application are:
· Python 3.10.0
The modules used in this application are:
The algorithms used in this application are:
· kNN (k-Nearest Neighbours)
· Logistic Regression
· Random Forests