Rupesh Sushir - Academia.edu (original) (raw)

Papers by Rupesh Sushir

Research paper thumbnail of Exploring Novel Self-Supervised Learning Techniques for Image Reconstruction Tasks

International Journal of Advanced Research in Science, Communication and Technology, Mar 2, 2024

Image reconstruction tasks, such as super-resolution, inpainting, and denoising, play a crucial r... more Image reconstruction tasks, such as super-resolution, inpainting, and denoising, play a crucial role in various computer vision applications. Traditional methods often rely heavily on large labeled datasets for training, which can be costly and time-consuming to acquire. Self-supervised learning has emerged as a promising alternative, aiming to reduce this dependency by leveraging the inherent structures within the data itself. In this paper, we explore novel self-supervised learning techniques tailored specifically for image reconstruction tasks. We propose approaches that exploit the inherent relationships between low and high-resolution images, utilize context-aware information for inpainting, and incorporate generative adversarial networks for denoising. Through extensive experimentation, we demonstrate the efficacy of our methods in achieving competitive performance compared to supervised approaches while significantly reducing the need for labeled data. Our findings pave the way for more efficient and scalable solutions in image reconstruction, offering practical benefits across a wide range of applications.

Research paper thumbnail of Image Super-Resolution using Convolutional Neural Networks

International Journal of Advanced Research in Science, Communication and Technology, Mar 2, 2024

Image super-resolution is the process of enhancing the resolution of an image, typically from a l... more Image super-resolution is the process of enhancing the resolution of an image, typically from a lower resolution input to a higher resolution output. This research aims to explore the application of convolutional neural networks (CNNs) for image super-resolution. Specifically, the study will focus on developing a deep learning model capable of generating high-resolution images from low-resolution inputs. Various CNN architectures, such as SRCNN (Super-Resolution Convolutional Neural Network) or SRGAN (Super-Resolution Generative Adversarial Network), will be investigated and compared for their effectiveness in producing visually pleasing and perceptually accurate high-resolution images. Additionally, techniques such as residual learning, attention mechanisms, and adversarial training may be incorporated to further improve the quality of super-resolved images. The performance of the proposed models will be evaluated using standard image quality metrics and subjective assessments. This research has practical applications in enhancing the visual quality of low-resolution images in fields such as medical imaging, surveillance, and entertainment.

Research paper thumbnail of Enhanced blind image forgery detection using an accurate deep learning based hybrid DCCAE and ADFC

Multimedia Tools and Applications

Research paper thumbnail of Customized haptic control for VRML object

Nonproprietary language like VRML can be used for scientific simulation and visualization. VRML l... more Nonproprietary language like VRML can be used for scientific simulation and visualization. VRML language has the capacity to hold the dynamic numerical values in the matrix form for its interactive objects. These matrices hold the values for manipulation like rotation, scaling, transparency, color, translation etc. Using Simulink’s MATLAB interface, these numeric values can be varied in real time which in turn will allow the user to exercise control over VRML object. The Simulink Graphic User Interface (GUI) demonstrates real time simulation of interactive Virtual Table/Desk Lamp using MATLAB’s Simulink Interface along with Electronic Hardware circuitry. The article explains how interactive VRML object can be controlled using MATLAB’s Simulink interface. The MATLAB’s Simulink capabilities can be utilized for real time simulation of interactive Table/Desk Lamp in virtual Environment. The concept presented in this paper can be extended for full fledged complex simulator design.

Research paper thumbnail of An improved detection of blind image forgery using hybrid deep belief network and adaptive fuzzy clustering

Multimedia Tools and Applications

Research paper thumbnail of Digital Image Forensic: Comparative Scrutiny of Foregoing Techniques

2021 International Conference on Computing, Communication and Green Engineering (CCGE)

Research paper thumbnail of Detection and Estimation of Adulteration in Oil Sample Using Digital Image Processing

International Journal of Scientific Research in Science and Technology, 2018

Now-a-days adulteration can cause several health and safety problem. Many techniques such as chro... more Now-a-days adulteration can cause several health and safety problem. Many techniques such as chromatographic and spectroscopic method have recently been employed to check the purity of oil. For most vegetable oil adulteration detection research methods, it remains difficult to popularize due to the fact that the application of experimental facility needs professional to operate; and it is usually expensive. Hence to solve this problem method is proposed. This project describes the development of an image processing algorithm, which can estimate the amount of adulteration oil sample from a captured photo. The algorithm is implemented into an application for modern smart phone where the user can measure the quality of a sample of oil only by taking photo of the sample. Then any other mixture of oil can be identified using the derived model and the methodology, which is based on color model based segmentation.

Research paper thumbnail of Exploring Novel Self-Supervised Learning Techniques for Image Reconstruction Tasks

International Journal of Advanced Research in Science, Communication and Technology, Mar 2, 2024

Image reconstruction tasks, such as super-resolution, inpainting, and denoising, play a crucial r... more Image reconstruction tasks, such as super-resolution, inpainting, and denoising, play a crucial role in various computer vision applications. Traditional methods often rely heavily on large labeled datasets for training, which can be costly and time-consuming to acquire. Self-supervised learning has emerged as a promising alternative, aiming to reduce this dependency by leveraging the inherent structures within the data itself. In this paper, we explore novel self-supervised learning techniques tailored specifically for image reconstruction tasks. We propose approaches that exploit the inherent relationships between low and high-resolution images, utilize context-aware information for inpainting, and incorporate generative adversarial networks for denoising. Through extensive experimentation, we demonstrate the efficacy of our methods in achieving competitive performance compared to supervised approaches while significantly reducing the need for labeled data. Our findings pave the way for more efficient and scalable solutions in image reconstruction, offering practical benefits across a wide range of applications.

Research paper thumbnail of Image Super-Resolution using Convolutional Neural Networks

International Journal of Advanced Research in Science, Communication and Technology, Mar 2, 2024

Image super-resolution is the process of enhancing the resolution of an image, typically from a l... more Image super-resolution is the process of enhancing the resolution of an image, typically from a lower resolution input to a higher resolution output. This research aims to explore the application of convolutional neural networks (CNNs) for image super-resolution. Specifically, the study will focus on developing a deep learning model capable of generating high-resolution images from low-resolution inputs. Various CNN architectures, such as SRCNN (Super-Resolution Convolutional Neural Network) or SRGAN (Super-Resolution Generative Adversarial Network), will be investigated and compared for their effectiveness in producing visually pleasing and perceptually accurate high-resolution images. Additionally, techniques such as residual learning, attention mechanisms, and adversarial training may be incorporated to further improve the quality of super-resolved images. The performance of the proposed models will be evaluated using standard image quality metrics and subjective assessments. This research has practical applications in enhancing the visual quality of low-resolution images in fields such as medical imaging, surveillance, and entertainment.

Research paper thumbnail of Enhanced blind image forgery detection using an accurate deep learning based hybrid DCCAE and ADFC

Multimedia Tools and Applications

Research paper thumbnail of Customized haptic control for VRML object

Nonproprietary language like VRML can be used for scientific simulation and visualization. VRML l... more Nonproprietary language like VRML can be used for scientific simulation and visualization. VRML language has the capacity to hold the dynamic numerical values in the matrix form for its interactive objects. These matrices hold the values for manipulation like rotation, scaling, transparency, color, translation etc. Using Simulink’s MATLAB interface, these numeric values can be varied in real time which in turn will allow the user to exercise control over VRML object. The Simulink Graphic User Interface (GUI) demonstrates real time simulation of interactive Virtual Table/Desk Lamp using MATLAB’s Simulink Interface along with Electronic Hardware circuitry. The article explains how interactive VRML object can be controlled using MATLAB’s Simulink interface. The MATLAB’s Simulink capabilities can be utilized for real time simulation of interactive Table/Desk Lamp in virtual Environment. The concept presented in this paper can be extended for full fledged complex simulator design.

Research paper thumbnail of An improved detection of blind image forgery using hybrid deep belief network and adaptive fuzzy clustering

Multimedia Tools and Applications

Research paper thumbnail of Digital Image Forensic: Comparative Scrutiny of Foregoing Techniques

2021 International Conference on Computing, Communication and Green Engineering (CCGE)

Research paper thumbnail of Detection and Estimation of Adulteration in Oil Sample Using Digital Image Processing

International Journal of Scientific Research in Science and Technology, 2018

Now-a-days adulteration can cause several health and safety problem. Many techniques such as chro... more Now-a-days adulteration can cause several health and safety problem. Many techniques such as chromatographic and spectroscopic method have recently been employed to check the purity of oil. For most vegetable oil adulteration detection research methods, it remains difficult to popularize due to the fact that the application of experimental facility needs professional to operate; and it is usually expensive. Hence to solve this problem method is proposed. This project describes the development of an image processing algorithm, which can estimate the amount of adulteration oil sample from a captured photo. The algorithm is implemented into an application for modern smart phone where the user can measure the quality of a sample of oil only by taking photo of the sample. Then any other mixture of oil can be identified using the derived model and the methodology, which is based on color model based segmentation.