Professor Bhabesh Deka
Senior Member IEEE USA, Fellow IETE India
ResearchGate ResearchGate ResearchGate

Ongoing:

1. Project Title: Visvesvaraya Young Faculty Research Grant

Funding Agency (Scheme): Visvesvaraya Ph.D. Scheme, MeitY-GoI

Research Contingency Fund: 17, 50,000/-

Date of Start: Jan, 2018

Duration: 3 Years 6 Months (2018 - 2021)
2. Project Title: Development of a Real-time Image Super-resolution System using Parallel Processing Hardware for Remote Sensing Applications

Funding Agency (Scheme): ISRO (RESPOND)

Total Fund: 25, 79,000/-

Date of Start: August, 2017

Duration: 3 Years + Extended (2017 - 2021)


Completed:

1. Project Title: Development of Parallel Processing Embedded Hardware for Super-resolution of Diffusion weighted and Spectroscopic Magnetic Resonance Images

Funding Agency (Scheme): AICTE (RPS)

Total Fund: 16, 47,059/-

Date of Start: October, 2017

Duration: 3 Years (2017 - 2020)

Major Findings:
  • Single image super-resolution based on sparse representations over a learned overcomplete dictionary for DW and MRS images.
  • Single image super-resolution based on sparse representations based coupled overcomplete dictionaries are learnt using DW and MRS images and utilized them to extract a priori HR information.
  • MR image super-resolution reconstruction using sparse representation on patch-wise of high-resolution patches from low-resolution feature patches pairs.
  • MR image super-resolution reconstruction not only exploits the sparsity of MR image but also utilize the non-local self-similarity of patches of the input LR image as prior knowledge.
  • Hardware implemented using the GP-GPU based parallel along with sequential for real clinical applications of MR images.
No. of Publications: 03
2. Project Title: Design and Development of Compressed Sensing Based Image Reconstruction Prototype Hardware Architecture for Medical Imaging Applications

Funding Agency: DST-SERB

Total Fund: 15, 08,000/-

Duration: 3 Years (July, 2012 - July, 2015)

Major Findings:
  • A 1-D/2-D FFT hardware implementation using the pipeline CORDIC algorithm on the embedded FPGA development board for obtaining complex k-space data from MR images
  • A pipelined architecture for the efficient implementation of OMP algorithm on FPGA
3. Project Title:Development of Compressed Sensing Based Image Reconstruction Techniques for Medical Imaging Applications
Funding Agency:UGC

Total Fund:12, 25,800 /-

Duration:3 Years (+ 6 months extended) (September, 2012 - December, 2015)

Major Findings:
  • A variable density undersampling pattern for Compressed Sensing MRI
  • A high throughput reconstruction technique for compressed sensing based MRI
  • Multi-slice MR image reconstruction using fast interpolated compressed sensing
  • An efficient MR image reconstruction technique by weighted wavelet tree sparsity
No. of Publications: 05