Siddhartha Chatterjee
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Siddhartha Chatterjee has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
Ahona Ghosh
Brainware University
1 shared publication- Ankita Mandal
Institute of Engineering and Management, Kolkata University of Engineering and Management
1 shared publication - Babli Kumari
Computer Science and Engineering, Dr B R Ambedkar National Institute of Technology, Jalandhar, Punjab - 144011, India
1 shared publication - Bijoy Kumar Mandal
NSHM Knowledge Campus
1 shared publication Biswaroop Nath
KIIT Deemed to be University
1 shared publicationChandan Mukherjee
The Assam Kaziranga University
1 shared publication
Latest Publications
- Open Access
- 4 January 2025
- Published by Association for Computing Machinery (ACM)
- p. 408-413
Automated Identification of Traffic Accidents in Images and Videos Employing Advanced Deep Learning Methods
An accident detection is essential since it ensures emergency response and raises the bar for road safety in general. In this paper, A CNN (Convolutional Neural Network) based collision recognition technique using the Tensorflow framework has been proposed. Deep learning has been used to automatically pick up information and identify important characteristics from pictures or videos taken by in-car cameras or roadside cameras. By utilizing a CNN architecture, model can effectively identify various accident-related patterns and distinguish them from normal traffic scenes. To train the CNN model, a large-scale dataset consisting of both accident and non-accident samples. The dataset is carefully labelled to ensure accurate ground truth annotations. During the training phase, data augmentation techniques has been employed to enhance the generalization ability of the model and mitigate overfitting. This technique has been used such as random crop, platooning and rotation to artificially enhance training diversity samples. For cropping, flipping and rotation of the images we have used image augmentation which is a function of ImageDataGenerator. After training, we evaluated the efficiency of our model on a separate test set, consisting of previously unseen accident and non-accident instances. Our model achieved 98% accuracy and robustness, demonstrating its effectiveness in accident detection. An OTP verification feature has been integrated to further enhance the system's functionality. This OTP verification confirms that emergency responders reach the accident location, adding a layer of security and ensuring follow-through in emergency responses. Overall, this study showcases the potential of deep learning techniques, specifically Convolutional Neural Networks, in addressing the crucial task of accident detection. This proposed method can integrate with the Computing Classification System (CCS), ensuring efficient categorization and real-time processing of data for accurate accident detection.
- 4 January 2025
- Published by Association for Computing Machinery (ACM)
- p. 408-413
Open AccessAutomated Identification of Traffic Accidents in Images and Videos Employing Advanced Deep Learning Methods
An accident detection is essential since it ensures emergency response and raises the bar for road safety in general. In this paper, A CNN (Convolutional Neural Network) based collision recognition technique using the Tensorflow framework has been proposed. Deep learning has been used to automatically pick up information and identify important characteristics from pictures or videos taken by in-car cameras or roadside cameras. By utilizing a CNN architecture, model can effectively identify various accident-related patterns and distinguish them from normal traffic scenes. To train the CNN model, a large-scale dataset consisting of both accident and non-accident samples. The dataset is carefully labelled to ensure accurate ground truth annotations. During the training phase, data augmentation techniques has been employed to enhance the generalization ability of the model and mitigate overfitting. This technique has been used such as random crop, platooning and rotation to artificially enhance training diversity samples. For cropping, flipping and rotation of the images we have used image augmentation which is a function of ImageDataGenerator. After training, we evaluated the efficiency of our model on a separate test set, consisting of previously unseen accident and non-accident instances. Our model achieved 98% accuracy and robustness, demonstrating its effectiveness in accident detection. An OTP verification feature has been integrated to further enhance the system's functionality. This OTP verification confirms that emergency responders reach the accident location, adding a layer of security and ensuring follow-through in emergency responses. Overall, this study showcases the potential of deep learning techniques, specifically Convolutional Neural Networks, in addressing the crucial task of accident detection. This proposed method can integrate with the Computing Classification System (CCS), ensuring efficient categorization and real-time processing of data for accurate accident detection.
- 10 September 2024
- Published by Springer Nature
- p. 71-83
Latent Dirichlet Allocation for Topic Modeling and Intelligent Document Classification
- 10 September 2024
- Published by Springer Nature
- p. 71-83
Latent Dirichlet Allocation for Topic Modeling and Intelligent Document Classification
- 25 July 2023
- Published by Springer Nature
- p. 539-549
An Analysis of Duckworth-Lewis-Stern Method in the Context of Interrupted Limited over Cricket Matches
- 25 July 2023
- Published by Springer Nature
- p. 539-549
An Analysis of Duckworth-Lewis-Stern Method in the Context of Interrupted Limited over Cricket Matches
- 30 March 2023
- Published by Springer Nature
- p. 179-190
Social Media Sentiment Analysis on Third Booster Dosage for COVID-19 Vaccination: A Holistic Machine Learning Approach
- 30 March 2023
- Published by Springer Nature
- p. 179-190
Social Media Sentiment Analysis on Third Booster Dosage for COVID-19 Vaccination: A Holistic Machine Learning Approach
- Open Access
- 10 April 2020
- Published by Elsevier in SSRN Electronic Journal
Comparison Study and Operation Collapsing Issues for Serial Implementation of Square Matrix Multiplication Approach Suitable in High Performance Computing Environment
Multiplication between two square matrices is one of the fundamental computational approach in the domain of mathematics and computer science where it is fully recognized as a fore most technique for several interdisciplinary domain and sub domains like linear algebra, graph theory, multidimensional graphics, cryptographic computation, convolution neural network, deep learning, digital signal processing, medical image processing, steganography, relativity theory, quantum computing and many others. In a high-performance computing environment computational complexity analysis of matrix multiplication algorithm ensures a powerful paradox that takes a massive data processing approach to the world where problem solution feasibility comes down in respect of operation and time. In our paper we have analyze different methods to multiply two square matrices (like 2×2 matrices) and their arithmetic complexity analysis in a asymptotic flavor along with operation collapsing issues through serial processing technique. We have analytically and experimentally explained and shown using MATLAB 9.3 simulator that fast matrix multiplication approach like Strassen and Winograd perform much better than conventional matrix multiplication algorithm especially for large amount of data.
- 10 April 2020
- Published by Elsevier in SSRN Electronic Journal
Open AccessComparison Study and Operation Collapsing Issues for Serial Implementation of Square Matrix Multiplication Approach Suitable in High Performance Computing Environment
Multiplication between two square matrices is one of the fundamental computational approach in the domain of mathematics and computer science where it is fully recognized as a fore most technique for several interdisciplinary domain and sub domains like linear algebra, graph theory, multidimensional graphics, cryptographic computation, convolution neural network, deep learning, digital signal processing, medical image processing, steganography, relativity theory, quantum computing and many others. In a high-performance computing environment computational complexity analysis of matrix multiplication algorithm ensures a powerful paradox that takes a massive data processing approach to the world where problem solution feasibility comes down in respect of operation and time. In our paper we have analyze different methods to multiply two square matrices (like 2×2 matrices) and their arithmetic complexity analysis in a asymptotic flavor along with operation collapsing issues through serial processing technique. We have analytically and experimentally explained and shown using MATLAB 9.3 simulator that fast matrix multiplication approach like Strassen and Winograd perform much better than conventional matrix multiplication algorithm especially for large amount of data.
- 1 February 1995
- Published by Elsevier in Journal of Parallel and Distributed Computing
- Vol. 25 (1) , 91-97
Solving Linear Recurrences with Loop Raking
- 1 February 1995
- Published by Elsevier in Journal of Parallel and Distributed Computing
- Vol. 25 (1) , 91-97
Solving Linear Recurrences with Loop Raking