Distracted Driver Detection using Deep Learning

Tags	: Deep Learning, CNN, Regression, Classification, Safety
Technologies	: Python3, Keras, Tensorflow, Jupyter Notebook, OpenCV, CUDA

One of the critical problems prevailing in India is the deaths caused by road accidents. Almost 80% of the accidents are caused by the inattentiveness of the driver. Usage of mobile phones, talking to passengers, reaching behind to grab something and drinking while driving are some of the reasons due to which driver may lose attention. Distractions are of numerous types, out of which we focus on the manual distraction which is based on the posture of the driver. In this project, we propose a system where we make use of Convolutional Neural Networks and data augmentation techniques. Data augmentation techniques are used to increase the variability of the dataset and decrease overfitting. We have used the first publicly available dataset form Kaggle as input for our model. Our aim was to categorize a test image into one of the nine distinct distracted states of the driver that we have considered. Conclusively, the experimental analysis has shown that applying data augmentation techniques, the proposed model gives better results.

Weather Induced Airline Delays Prediction

Tags	: Data mining, ML, Delay Prediction, Imbalanced data, Sampling, classification
Technologies	: Python3, SciKit Learn, MatplotLib, Jupyter Notebook

The reason behind choosing this topic was to identify potential causes behind delays and then attain significant operational cost savings, better flight scheduling and find ways to alleviate the impact. The primary goal of the model proposed in this Project is to predict airline delays caused by inclement of weather conditions using data mining and supervised machine learning algorithm (Random Forest). 2008 US domestic flight data and weather data were used for training and prediction. After Data analysis, we learnt about various regions and duration in which flight delays occurred the most. After the analysis, Random Forest was trained on available data. But more samples of non-delayed flight data than delayed flights resulted in "Accuracy Paradox". hence to oversample minor data we used SMOTE ( Synthetic Minority Oversampling TechniquE ). Training results after oversampling were much better. Hence we concluded that oversampling the dataset improves the performance of the model.

Bounce - Desktop

Tags	: Games, Bounce, Nokia, unity3D
Technologies	: Unity, C#, JS

I have always been fascinated by games. I always wanted to make my own game and Computer Graphics Course in the 2nd year of Engineering gave me the opportunity. we were asked to develop a small game using any technology in 1 week. I came across unity3D and started developing "Bounce". It was not only the most intuitive but at the same time one of the most simple games I have ever played. Due to time constraints, I was able to make only 2 levels. Teachers were impressed by work and asked me to put the game in the college exhibition. The response received from many viewers was very positive.

Twitter Bot

Tags	: Twitter, Bot, Railway, PNR, Train
Technologies	: Python, Urllib, RailwayAPI

When I was coming home from Bhopal. I wanted to check my PNR status. I had internet on my phone but it was very slow and my phone was almost out of battery. Plus using the internet, Opening the website, entering PNR number, waiting for results was way too much work. I wondered we use social media almost every day. what if there was a way to check the status of reservation and use the social media? that is how I came up with the idea of making a Bot to check PNR status. I wanted to make a Whatsapp bot but it is not allowed. So I made a twitter bot instead. All you had to do was to tweet @getMyPNRStatus < your PNR number > and the status was retweeted as long as my bot was online. Later I added many more functionalities like Train status, timings, etc. the results were obtained from RailwayAPI. But it a paid service and my free trial period have now expired so in order to use it. Log on to RailwayAPI and enter your secret keys at the appropriate places in the code.

Journey Planner

Tags	: Tourism, Shortest Distance, Journey, Google Maps, itinery
Technologies	: HTML, CSS, JS, Google Maps API

When You add multiple destinations to Google Maps. It appends the new destination to the previous path. It does not provide the shortest distance between all the destinations provided. So I used Google Maps API and applied "Travelling Salesperson Problem" to it. Which gives the smallest distance as well as time to visit all the destinations provided. I actually used it to plan our Mahabaleshwar Trip. Image Above is the screenshot of the same. I used my application to plan how to visit various locations at Mahabaleshwar And it was quite helpful.

Wall Street

Tags	: Stock Market, Buy, Sell, Stocks
Technologies	: HTML, CSS, JS, AJAX, JQuery, Php, MySQL

We developed a Virtual stock market event "WallStreet" under CREDENZ' 16 using Web Technologies. The web app consisted of all basic stock market operations. People could buy stocks, sell them, they would also get a notification for news and when their stocks sold. More than 800 people registered for this event and throughout the event, there were no bugs in the transactions people made. We Learnt so many things like huge lines of Complex SQL queries, Transactions, and real-time updations. Many players appreciated the work we had done for the web app.

Tags

: Deep Learning, CNN, Regression, Classification, Safety

Technologies

: Python3, Keras, Tensorflow, Jupyter Notebook, OpenCV, CUDA

Tags

: Data mining, ML, Delay Prediction, Imbalanced data, Sampling, classification

Technologies

: Python3, SciKit Learn, MatplotLib, Jupyter Notebook

Tags

: Games, Bounce, Nokia, unity3D

Technologies

: Unity, C#, JS

Tags

: Twitter, Bot, Railway, PNR, Train

Technologies

: Python, Urllib, RailwayAPI

Tags

: Tourism, Shortest Distance, Journey, Google Maps, itinery

Technologies

: HTML, CSS, JS, Google Maps API

Tags

: Stock Market, Buy, Sell, Stocks

Technologies

: HTML, CSS, JS, AJAX, JQuery, Php, MySQL

+91 7743872712

ani.humane@gmail.com

ani.humane@outlook.com