Publications

2023

1. 

   Towards Real-Time Analysis of Broadcast Badminton Videos

   Nitin Nilesh, Tushar Sharma, Anurag Ghosh, and C.V. Jawahar

   Arxiv Preprint, Oct 2023

   An end-to-end framework for real-time badminton analysis, tested live in Premier Badminton League (PBL). Find the live broadcast by Hotstar showing the analysis here.

   Abs arXiv HTML Blog Code

   Analysis of player movements is a crucial subset of sports analysis. Existing player movement analysis methods use recorded videos after the match is over. In this work, we propose an end-to-end framework for player movement analysis for badminton matches on live broadcast match videos. We only use the visual inputs from the match and, unlike other approaches which use multi-modal sensor data, our approach uses only visual cues. We propose a method to calculate the on-court distance covered by both the players from the video feed of a live broadcast badminton match. To perform this analysis, we focus on the gameplay by removing replays and other redundant parts of the broadcast match. We then perform player tracking to identify and track the movements of both players in each frame. Finally, we calculate the distance covered by each player and the average speed with which they move on the court. We further show a heatmap of the areas covered by the player on the court which is useful for analyzing the gameplay of the player. Our proposed framework was successfully used to analyze live broadcast matches in real-time during the Premier Badminton League 2019 (PBL 2019), with commentators and broadcasters appreciating the utility.

2. 

   DoA Estimation using Cascaded Neural Networks and Angle Classification for Coherent Signals

   Jigyasu Khandelwal, M. Madhuri Latha, Nitin Nilesh, and Sachin Chaudhari

   34th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC), Sep 2023

   Abs PDF

   This paper focuses on the direction of arrival (DoA) estimation for two coherent sources using a uniform linear array (ULA). It is demonstrated that the DoA estimation error increases as the angle of the incoming signal moves away from the center in the range (-90°, 90°) for existing schemes. In addition, this paper also focuses on improving the DoA estimation performance in low signal to noise ratio (SNR). Therefore a cascaded neural network (CaNN) is proposed to improve the DoA estimation consisting of two stages of neural networks (NNs). The first NN is the enhanced SNR (ESNR) classifier, which is used to improve performance for different SNRs. The second NN is the angle estimator, which improves the performance for different angle ranges. For overcoming the issue of coherent signals, a spatially smoothened auto covariance matrix is fed to the SNR classifier and angle estimator blocks. A performance comparison with an existing scheme such as spatial smoothing- multiple signal classification (SS-MUSIC) and ESNR CaNN shows that the proposed CaNN for different angles and SNR ranges performs better than the existing schemes.

3. 

   Using Miniature Setups and Partial Streams for Scalable Remote Labs

   Animesh Das, K. S. Viswanadh, Rishabh Agrawal, Akshit Gureja, Nitin Nilesh, and Sachin Chaudhari

   10th International Conference on Future Internet of Things and Cloud (FiCloud), Aug 2023

   Abs PDF

   Remote labs allow students from anywhere in the world to access and conduct experiments without the need to physically be present in a lab at anytime. This is extremely important for students with little to no access to proper science labs because of a lack of infrastructure or a pandemic. There is a potential for scaling up the process so that queue delay can be removed and people are able to access dedicated experiment setups. This paper proposes a way for cost-effective scaling of Remote Labs by miniaturization of setups, image processing techniques, and partial streaming (using single camera to stream multiple experiments). For this, a use-case of the Vanishing Glass Rods experiment is considered. An end-to-end remote lab is created, including hardware setups and a dashboard to demonstrate the efficacy of the proposed approach in comparison to the existing approach of using lab-scale setups and one camera per experiment.

2022

1. 

   IoT-based AQI Estimation using Image Processing and Learning Methods

   Nitin Nilesh, Ishan Patwardhan, Jayati Narang, and Sachin Chaudhari

   IEEE 8th World Forum on Internet of Things (WF-IoT), Oct 2022

   Winner of the Environmental Sensing Project Competition 2022.

   Abs PDF Blog Code Website

   Air pollution is a concern to the health of all living beings. It is essential to check on the quality of air in the surroundings. This article presents an IoT-based real-time air quality index (AQI) estimation technique using images and weather sensors on Indian roads. A mixture of image features, i.e., traffic density, visibility, and sensor features, i.e., temperature and humidity, were used to predict the AQI. Object detection and localization-based Deep Learning (DL) method along with image processing techniques were used to extract image features while an Machine Learning (ML) model was trained on those features to estimate the AQI. In order to conduct this experiment, a dataset containing 5048 images along with co-located AQI values across different seasons was collected by driving on the roads of Hyderabad city in India. The experimental results report an overall accuracy of 82% for AQI prediction.

2. 

   IoT and ML-based AQI Estimation using Real-time Traffic Data

   Nitin Nilesh, Jayati Narang, Ayu Parmar, and Sachin Chaudhari

   IEEE 8th World Forum on Internet of Things (WF-IoT), Oct 2022

   Abs PDF Website

   This paper proposes an IoT and machine learning (ML)-based novel method to estimate the air quality index (AQI) using traffic data in real-time. With the help of particulate matter (PM) monitoring nodes deployed in fifteen locations with diverse traffic scenarios of Indian roads, and using digital map service providers, a rich traffic dataset with approximately 210,000 samples has been collected. Three different ML models, namely random forest (RF), support vector machine (SVM), and multi-layer perceptron (MLP), are trained on this dataset to predict the AQI category into five levels. The experimental results show an accuracy of 82.60% with the F1-score of 83.67% on the complete dataset. Apart from this, ML models were also trained on individual node datasets, and the behavior of AQI levels was observed.

3. 

   Improving IoT-based Smart Retrofit Model for Analog Water Meters using DL based Algorithm

   Ayush Kumar Lall, Ansh Khandelwal, Nitin Nilesh, and Sachin Chaudhari

   9th International Conference on Future Internet of Things and Cloud (FiCloud), Aug 2022

   Abs PDF

   This paper proposes a deep learning (DL)-based algorithm which is used for improving the performance of digit detection from internet-of-things (IoT)-based analog water meters. The DL algorithm is trained on a rich dataset of over 160,000 images collected from six water nodes deployed at locations with different environmental conditions. A detailed comparison between the proposed DL and machine learning (ML) algorithm is made based on detection accuracy, feature analysis, error analysis, and computational complexity analysis. It is observed that compared to the ML model, the proposed DL model maintained a higher detection accuracy and is more generalized in terms of feature extraction, which makes the algorithm robust.

4. 

   CV and IoT-based Remote Triggered Labs: Use Case of Conservation of Mechanical Energy

   K. S. Viswanadh, Om Kathalkar, Piyusha Vinzey, Nitin Nilesh, Sachin Chaudhari, and Venkatesh Choppella

   9th International Conference on Future Internet of Things and Cloud (FiCloud), Aug 2022

   Abs PDF Blog Website

   Remote Triggered Labs (RTL) are helpful for students to work on laboratory experiments virtually anytime, anywhere. Such setups can facilitate distance learning and are helpful during pandemics. In this paper, the use of Computer Vision (CV) is demonstrated for RTL experiments. For this, a use-case of the Conservation of Mechanical Energy experiment is considered. A CV-based approach is used to estimate an object’s velocity whose setup primarily consists of a microprocessor, a camera and infrared (IR) sensors. The experiment is recorded, and various CV techniques are employed to estimate the object’s velocity. This paper also compares a CV-based and an IR sensor-based approach to estimate the object’s velocity. Linear regression applied to the CV-based implementation resulted in an optimal mean-squared error (MSE), nearly 10 times better than IR-based implementation.

2021

1. 

   Making Analog Water Meter Smart using ML and IoT-based Low-Cost Retrofitting

   Ayush Kumar Lall, Ansh Khandelwal, Rishikesh Bose, Nilesh Bawankar, Nitin Nilesh, Ayush Dwivedi, and Sachin Chaudhari

   8th International Conference on Future Internet of Things and Cloud (FiCloud), Aug 2021

   Abs PDF

   This paper introduces an internet-of-things (IoT) based economic retrofitting setup for digitising the analog water meters to make them smart. The setup contains a Raspberry-Pi microcontroller and a Pi-camera mounted on top of the analog water meter to take its images. The captured images are then preprocessed to estimate readings using a machine learning (ML) model. The employed ML algorithm is trained on a rich dataset that includes digits from the images of water meters captured by the hardware setup for ten days. The readings are posted on a cloud server in real-time using Raspberry-Pi. High temporal resolution plots of flow rate and volume are generated to derive inferences. The collected data can be used for deriving water consumption patterns and fault detection for efficient water management.

2024

1. 

   System and Method for Implementing an Experiment Remotely and Determining an Output Using a Computer Vision Model

   Sachin Chaudhari, Venkatesh Choppella, Om Rajendra Kathalkar, Viswanadh Savita Kandala, and Nitin Nilesh

   United States Patent and Trademark Office (USPTO), Mar 2024

   Abs PDF Website

   A system and method for implementing an experiment remotely and determining an output using a computer-vision model is provided. The system includes an image capturing device, an experiment setup, a microcontroller, a user device, and a relay unit. The microcontroller (i) receives the input of the experiment from the image capturing device, (ii) extracts one or more frames from the input data, (iii) pre-process the one or more frames to obtain a binary image, (iv) obtain a closed curve around the binary image to locate the experiment, (v) determine the coordinates of the experiment to track the experiment in each frame, (vi) determine an output of the experiment from every two consecutive frames of the one or more frames, and (vii) optimize the determined output of the experiment using a linear regression model.

2023

1. 

   Learning Methods for IoT: Use Cases of Air Pollution Monitoring

   Nitin Nilesh

   International Institute of Information Technology, Hyderabad, Jul 2023

   Master’s Thesis

   Abs PDF

   Air pollution monitoring is crucial for assessing the health risks posed by pollutants, identifying pollution sources, and developing effective strategies for reducing pollution and protecting public health. The Indian government has taken initiatives, which provides real-time air quality data to the public and raises awareness of air pollution levels. The initiative measures several pollutants, including Particulate Matter (PM), and categorizes Air Quality Index (AQI) level into six categories ranging from “Good” to “Severe”. The air pollution sensors employed for calculating the AQI are associated with several limitations. Consequently, the central objective of this thesis is to estimate the AQI without relying on any pollution sensor. To achieve this goal, the proposed methodology employs real-time traffic data and images to estimate the AQI in real-time. Firstly, this thesis propose an image processing based technique to estimate the AQI levels using traffic images and weather parameters, which can be used in rural and sub-urban areas where sensors are hard to deploy and maintain. This approach allows for real-time estimation of AQI through smartphones and can be used portably. The proposed method achieves up to 90% accuracy for the AQI classification. Furthermore, a feature-rich dataset is made publicly available to encourage further research. After that, a novel method based on the Internet-of-Things (IoT) and Machine Learning (ML) is proposed to estimate the AQI using real-time traffic data. To build a rich traffic dataset, PM monitoring nodes were deployed in 15 diverse traffic scenarios across Indian roads, and digital map service providers were utilized. Three ML models, namely Random Forest (RF), Support Vector machine (SVM), and Multi Layer Perceptron (MLP), were trained on this dataset to predict AQI categories into five levels. Experimental results demonstrate an accuracy of 82.60% and an F1-score of 83.67% on the complete dataset. In addition, individual node datasets were used to train ML models, and the behavior of AQI levels was observed.