Dr. Samiran Das

Dr. Samiron Das

Assistant Professor, Centre for Data Science

PhD: Indian Institute of Technology Kharagpur, West Bengal, India

Email: samiran@jisiasr.org, itzsamirandas@gmail.com

Linked-In: https://www.linkedin.com/in/samiran-das/

Google Scholar: https://scholar.google.co.in/citations?user=1p3HJ-cAAAAJ&hl=en

Publons: https://publons.com/researcher/3302198/samiran-das/

Webpage: https://sites.google.com/view/samiran-das/home?authuser=0

Ph.D., Advanced Technology Development Center, Indian Institute of Technology Kharagpur, India

MTech. Instrumentation & Control Engineering,Rajabazar Science College, University of Calcutta

Research Interests:

Signal, and Image Processing: Object recognition from hyperspectral image, band selection and compression of hyperspectral image

Machine learning: Face recognition, unsupervised learning

Deep Learning: Industrial computer vision

Professional Experience

  • Assistant Professor JIS Institute for Advanced Studies and Research, JIS University

Jan 2022 to present

  • Assistant Professor School of Computer Science, University of Petroleum and Energy Studies (NIRF Ranking-91), Sept 2020- Dec 2021
  • Research Engineer Micelio Labs Pvt. Ltd., Machine Learning Research and Development, Dec 2019-Aug 2020
  • Project Fellow Central Mechanical Engineering Research Institute, Aug 2013-Dec 2013

Topic- Electrostatic Characterization of Cell

Research Focus:
The fields of signal processing, image processing, machine learning, and deep learning are fascinating as they form the basics of data analytics. As far as I understand, the subjects lead the door to many interesting theoretical and applied research avenues. These concepts can be realized and implemented in dedicated software, tools, or applications, which can significantly help non-experts perform their analysis. The tremendous scope motivated me to undertake research assignments based on the foundations provided by these subjects.

To be more precise, I would like to explore machine learning applications in hyperspectral image processing, face recognition between different modalities, and industrial computer vision. Hyperspectral imaging is an advanced imaging technique that captured image information from ground scenes at hundreds of spectral bands across different electromagnetic wavelengths spanning visible and near-infrared regions. These images are very useful in accurate classification, object identification, and target detection. Deep learning and advanced pattern recognition methods will be helpful in the efficient visualization and compression of hyperspectral images. Recently recognition of visual, sketch and/or thermal facial images have emerged as an imperative task. Since, these images come from different modalities, traditional machine learning-based approaches are ineffective. Hence, I am exploring domain adaptation, generative adversarial network, and deep learning approaches for the recognition task. I also wish to explore visual recognition problems in industrial applications. This includes characterization of insulators in power plant, categorization and grading of solar photovoltaic cells, etc.

Awards & Achievements:

UGC Junior Research Fellowship, Electronic Science, 2013

GATE Fellowship, Instrumentation Engineering, 2013

Professional Activities:

Serving as a reviewer in

IEEE Transactions on Geoscience and Remote Sensing

IEEE Geoscience and Remote Sensing Letters

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing

IET Image Processing

IET Signal Processing

SPIE Journal of Applied Remote Sensing

IET Electronic Letters

International Journal of Remote Sensing

Geocarto International

Publication List

International Journals

1. Samiran Das, Aurobinda Routray, and Alok Kanti Deb, “Fast Semi-supervised Unmixing of Hyperspectral Image by Mutual Coherence Reduction and Recursive PCA”, Remote Sensing, vol. 10, no. 7, pp-1106, 2018, DOI- 10.3390/rs10071106.

2. Samiran Das, and Aurobinda Routray, “Covariance Similarity Approach for Semi-blind Unmixing of Hyperspectral Image” IEEE Geoscience and Remote Sensing Letters, vol. 16, no. 6, pp. 937-941, 2019.

3. Samiran Das, Aurobinda Routray and Alok Kanti Deb, “Sparsity Measures for Library Aided Unmixing of Hyperspectral Image”, IET Image Processing, vol. 13, no. 12,pp. 2077-2085, 2019.

4. Samiran Das​, Shubhobrata Bhattacharya, Aurobinda Routray and Alok Kanti Deb,, “Band Selection of Hyperspectral Image by Sparse Manifold Clustering”, IET Image Processing, vol. 13, no. 10, pp. 1625-1635, 2019.

5. Samiran Das, Aurobinda Routray and Alok Kanti Deb, “Efficient Tensor Decomposition Approach for Estimation of the Number of Endmembers in a Hyperspectral Image”, Journal of Applied Remote Sening, vol 14, Issue 1, 2020.

6. Samiran Das, Sohom Chakraborty, Aurobinda Routray and Alok Kanti Deb, “Library Aided Bilinear Unmixing of Hyperspectral Image using Subspace Clustering and Multi Step Pruning”, Journal of Applied Remote Sensing, Vol 13, Issue 4, 2019.

7. Samiran Das, Shubhobrata Bhattacharya, and Pushkar Kumar Khatri, “Feature Extraction Approach for Quality Assessment of Remotely Sensed Hyperspectral Images”, Journal of Applied Remote Sensing , vol 14, no 4, 2020.

8. Samiran Das, “Hyperspectral Image, Video Compression by Sparse Tucker Tensor Decomposition”, IET Image Processing, Vol 15, no.4, pp.- 964-973, 2021.

9. Samiran Das, Shubhobrata Bhattacharya, and Sawon Pratiher, “Efficient multi-way sparsity estimation for hyperspectral image processing”, Journal of Applied Remote Sensing , vol 15, no 2, 2021.

10. Samiran Das, and Chirag Kyal, “Efficient multichannel EEG compression by optimal tensor truncation”, Biomedical Signal Processing and Control, vol 68, Pages 102749, 2021.

International Conferences

1. Shubhobrata Bhattacharya, ​Samiran Das, and Aurobinda Routray, “Graph Manifold Clustering based Band Selection for Hyperspectral Face Recognition” 26th European Signal Processing Conference, EUSIPCO 2018, pp- 1990-1994, 2018.

2. Shubhobrata Bhattacharya, ​Samiran Das​, Sohom Chakraborty and Aurobinda Routray, “Combining Pixel Selection with Covariance Similarity Approach in Hyperspectral Face Recognition”, IEEE Industrial Electronics Conference, IECON, 2018, pp- 2695–2699, 2018.

3. ​Samiran Das​, Sohom Chakraborty, Aurobinda Routray, and Alok Kanti Deb, “Fast Linear Unmixing of Hyperspectral Image by Slow Feature Analysis and Simplex Volume Ratio Approach”, 2019 IEEE International Geoscience and Remote Sensing Symposium, IGARSS, pp. 560-563, 2019. 

4. Samiran Das, Jogendra Nath Kundu, and Aurobinda Routray, “Estimation of number of endmembers in a hyperspectral image using eigen thresholding”, 2015 Annual IEEE India Conference (INDICON), pp. 1-5, 2015.

5. Samiran Das, Aurobinda Routray and Alok Kanti Deb, “Noise robust estimation of number of endmembers in a hyperspectral image by eigenvalue based gap index”, IEEE 8th Workshop on. Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), pp. 1-5, 2016.

6. Samiran Das, Aurobinda Routray and Alok Kanti Deb, “Convex set based abundance constrained unmixing of hyperspectral image”, 2017 Annual IEEE India Conference (INDICON), pp. 1-5, 2017.

7. Samiran Das, Aurobinda Routray and Alok Kanti Deb, “Hyperspectral Unmixing by Nuclear Norm Difference Minimization based Dictionary”, 2017 Annual IEEE India Conference (INDICON), 2017, pp. 1-5. 8. Samiran Das, Chirag Kyal, and Sawon Pratiher, “On Sparsity Measures In Deep Subspace Clustering” Second International Conference on Power, Control and Computing Technologies, (ICPCT 2022) (presented)