Shiv Ram Dubey
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shiv Ram Dubey has a total of 140 co-authors. Here are the top co-authors by number of shared publications:
- Satish Kumar Singh
Indian Institute of Information Technology Allahabad
39 shared publications - Swalpa Kumar Roy
Department of Computer Science and Engineering, Tezpur University, Assam, India
14 shared publications Rajat Kumar Singh
11 shared publications- Bidyarthi Baran Chaudhuri
Techno India University Kolkata India
9 shared publications Snehasis Mukherjee
Shiv Nadar University India
8 shared publications- Nand Kumar Yadav
AI Research Lab, Department of Computer Science, University of South Dakota, Vermillion, SD 57069, USA
6 shared publications
Latest Publications
- 24 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7  (6) , 3171-3180
Investigating the Robustness of Fuzzy Deep Learning on Noisy Medical Images
Biomedical imaging often encounters noise-related challenges, compromising the accuracy of automated systems. The utilization of fuzzy systems present a compelling avenue to enhance the robustness of image processing in the biomedical domain. This research explores the robustness of a deep neurofuzzy system (DNFS) for medical image classification to study the challenges posed by noisy medical images, utilizing the robust characteristics of fuzzy systems. Six types of noise – Gaussian, Salt and Pepper, Speckle, Blur, Contrast, and Inverse, were used to investigate the DNFS performance under diverse conditions. Adversarial attacks were implemented to evaluate the model’s susceptibility to perturbations. The DNFS was trained and tested on seven distinct biomedical image classification datasets: colon pathology, Dermatoscope, retinal OCT, chest Xray, breast ultrasound, blood cell microscope, and abdominal CT. Comparative analyses were conducted with state-of-the-art models. Results indicated that the DNFS achieved similar accuracy to these benchmark models on original images, and exhibited superior average accuracy on noisy data, depicting the model’s resilience to noise compared to other models. This finding underscores the efficacy of deep neuro-fuzzy models in handling diverse noise types commonly encountered in medical imaging. The robust performance was evident across multiple datasets, reinforcing the generalizability of the DNFS approach. The study also delved into the vulnerability of DNFS and the benchmark models to adversarial attacks, showcasing the weakness of deep learning.
- 24 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7  (6) , 3171-3180
Investigating the Robustness of Fuzzy Deep Learning on Noisy Medical Images
Biomedical imaging often encounters noise-related challenges, compromising the accuracy of automated systems. The utilization of fuzzy systems present a compelling avenue to enhance the robustness of image processing in the biomedical domain. This research explores the robustness of a deep neurofuzzy system (DNFS) for medical image classification to study the challenges posed by noisy medical images, utilizing the robust characteristics of fuzzy systems. Six types of noise – Gaussian, Salt and Pepper, Speckle, Blur, Contrast, and Inverse, were used to investigate the DNFS performance under diverse conditions. Adversarial attacks were implemented to evaluate the model’s susceptibility to perturbations. The DNFS was trained and tested on seven distinct biomedical image classification datasets: colon pathology, Dermatoscope, retinal OCT, chest Xray, breast ultrasound, blood cell microscope, and abdominal CT. Comparative analyses were conducted with state-of-the-art models. Results indicated that the DNFS achieved similar accuracy to these benchmark models on original images, and exhibited superior average accuracy on noisy data, depicting the model’s resilience to noise compared to other models. This finding underscores the efficacy of deep neuro-fuzzy models in handling diverse noise types commonly encountered in medical imaging. The robust performance was evident across multiple datasets, reinforcing the generalizability of the DNFS approach. The study also delved into the vulnerability of DNFS and the benchmark models to adversarial attacks, showcasing the weakness of deep learning.
- 7 November 2025
- Published by Elsevier in Journal of Thermal Biology
- Vol. 134, 104325
Diagnosis of superficial ailments using infrared thermal imaging and CapsNet
- 7 November 2025
- Published by Elsevier in Journal of Thermal Biology
- Vol. 134, 104325
Diagnosis of superficial ailments using infrared thermal imaging and CapsNet
- 13 September 2025
- Published by Elsevier in Image and Vision Computing
- Vol. 163
UpAttTrans: Upscaled attention based transformer for facial image super-resolution
- 13 September 2025
- Published by Elsevier in Image and Vision Computing
- Vol. 163
UpAttTrans: Upscaled attention based transformer for facial image super-resolution
- 29 July 2025
- Published by Springer Nature
- p. 308-322
SRTransGAN: Image Super-Resolution Using Transformer Based Generative Adversarial Network
- 29 July 2025
- Published by Springer Nature
- p. 308-322
SRTransGAN: Image Super-Resolution Using Transformer Based Generative Adversarial Network
- 29 July 2025
- Published by Springer Nature
- p. 353-365
Face Recognition in Selfie Images Using Vision Transformer Variants
- 29 July 2025
- Published by Springer Nature
- p. 353-365
Face Recognition in Selfie Images Using Vision Transformer Variants
- 29 July 2025
- Published by Springer Nature
- p. 16-31
Impact of Color Channel Perturbation Attack on Vision Transformers
- 29 July 2025
- Published by Springer Nature
- p. 16-31
Impact of Color Channel Perturbation Attack on Vision Transformers
- 26 May 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 195, 73-79
PTSR: A patch-based translator model for image super-resolution
- 26 May 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 195, 73-79
PTSR: A patch-based translator model for image super-resolution
- 22 April 2025
- Published by Springer Nature in Machine Vision and Applications
- Vol. 36  (3) , 1-14
Adaptive adam-based optimizers using second-order weight decoupling and gradient-aware weight decay for vision transformer
- 22 April 2025
- Published by Springer Nature in Machine Vision and Applications
- Vol. 36  (3) , 1-14
Adaptive adam-based optimizers using second-order weight decoupling and gradient-aware weight decay for vision transformer
- 28 March 2025
- Published by Springer Nature
- p. 337-353
3D-CmT: 3D-CNN Meets Transformer for Hyperspectral Image Classification
- 28 March 2025
- Published by Springer Nature
- p. 337-353
3D-CmT: 3D-CNN Meets Transformer for Hyperspectral Image Classification