Lab FOr Video and Image Analysis (LFOVIA) Research areas:
Image and Video Quality Assessment:
Top-left: pristine image. All other images are its distorted versions. MSE identical for all distorted images!
The explosive growth of image and video content has put tremendous stress on storage and communication systems. It is imperative to
manage this content to make the best use of available resources and to better plan for next generation systems. Toward this end, we believe that the ability to automatically
sort/order content based on perceptual quality/importance would be extremely useful for storage and communication. It has been conclusively shown that standard lp
type metrics and their functions like PSNR do not correlate well with subjective judgment (see figure for a simple example).
We are actively working on developing objective
quality assessment algorithms for both image and video content. Quality assessment algorithms can broadly be categorized into full-reference, reduced-reference and no-reference
based on the availability of the pristine reference for comparison. From a practical viewpoint, the no-reference case is most relevant and forms the primary
focus of our work.
Relevant publications:
1. V. Neeluri, P. Dendi, M. Chandrashekhar Bh., S. S. Channappayya, S. Medasani, "Blind Image Quality Evaluation Using Perception Based Features," accepted to NCC 2015, IIT Bombay.
2. R. R. Tamboli, K. K. Vupparaboina, J. R. Regatti, S. Jana, S. S. Channappayya, "A Subjective Evaluation of True 3D Images,'' Proc. of IEEE IC3D 2014 Liege, Belgium,
December 2014.
3. K. V. S. N. L. Manasa Priya, S. S. Channappayya, "A Novel Sparsity-inspired Blind Image Quality Assessment Algorithm,'' Proc. of IEEE GlobalSIP 2014, Atlanta, USA, December 2014.
4. K. Manasa, K. V. S. N. L. Manasa Priya, S. S. Channappayya, "A Perceptually Motivated No-reference Video Quality Assessment Algorithm for Packet Loss Artifacts,'' Proc. of QoMEX 2014, Singapore, September 2014.
5. P. Dendi, V. Neeluri, M. Chandrasekar Bh, S. S. Channappayya, S. Medasani, "Blind Distortion Classification Using Content and Perception Based Features,'' Proc. of QoMEX 2014, Singapore, September 2014.
6. K. V. S. N. L. Manasa Priya, K. Manasa, S. S. Channappayya, "A Statistical Evaluation of Sparsity-based Distance Measure (SDM) as an Image Quality Assessment Algorithm,'' Proc. of ICASSP 2014, Florence, Italy, May 2014.
Multimedia Communication:
In continuation with the theme of multimedia content storage and communication, we are exploring problems on the optimal allocation of
network resources where the optimization criterion is the quality of experience (QoE). Specifically, our system considers models of the Long Term Evolution (LTE) network
standard, and video streaming over HTTP.
Relevant publications:
1. S. Sadhana Reddy, K. Manasa, S. S. Channappayya, "Video Packet Priority Assignment Based on Spatio-Temporal Perceptual Importance,'' accepted to NCC 2015, IIT Bombay.
Biomedical Imaging and Image Processing:
Challenges working with typical OCT images. Image courtesy: LVPEI.
We actively collaborate with LV Prasad Eye Institute (LVPEI) on a variety of problems related to the automated analysis of
choroidal images. The primary goal of this research is to assist physicians and experts in speeding up the diagnostic process and to make suggestions on potential disease
conditions.
We are working on photoacoustic imaging (both theory and system design) aimed at phantom imaging. Specifically, we are working on improved reconstruction algorithms and
building a working prototype.
As part of our Grand Challenges Canada award (PI Dr. Siva Vanjari), we are working on building low-cost mobile based solutions for the quantification of thyroid stimulating
hormone (TSH) concentration from the output of fabric-based test kits.
Relevant publications:
1. J. Chhablani, S. S. Channappayya, A. Richhariya, "Can an automated algorithm identify choriocapillaris in 2D-optical coherence tomography images?," Expert Review on Ophthalmology, DOI 10.1586/17469899.2014.922875.
2. K. J. Francis, S. S. Channappayya, P. Rajalakshmi, "Wavelet Domain Frequency interpolation for Photo-acoustic Tomography,'' Proc. IEEE MedCom 2014, Noida, India, November 2014.
3. S. Kakileti, S. S. Channappayya, A. Richhariya, J. Chhablani, "An Automated Algorithm for the Identification of Choriocapillaris in 2D-OCT Images,'' Proc. of SPIE Medical Imaging 2014, San Diego, CA, USA, February 2014.
4. K. Abhishek, M. Haloi, S. S. Channappayya, S. Vanjari, D. Dendukuri, S. Swathy, T. Choudhary, P. Bhandari, "An Enhanced Algorithm for the Quantification of Human Chorionic Gonadotropin (hCG) Level in Commercially Available Home Pregnancy Test Kits,'' Proc. of National Conference on Communications 2014, Kanpur, India, February 2014.
5. K. Manasa, K. V. S. N. L. Manasa Priya, S. Sadhana Reddy, S. S. Channappayya, S. Vanjari, D. Dendukuri, S. Swathy, T. Choudhary, P. Bhandari, "An Automated Algorithm for the Quantification of hCG Level in Novel Fabric-based Home Pregnancy Test Kits," Proc. of IEEE Asilomar Conference on Signals, Systems and Computers 2013, Pacific Grove, CA USA, November 2013.
6. N. R. Mahajan, R. C. Reddy Donapati, S. S. Channappayya, S. Vanjari, A. Richhariya, J. Chhablani, "An Automated Algorithm for Blood Vessel Count and Area
Measurement in 2-D Choroidal Scan Images," Proc. of IEEE EMBC 2013, Osaka, Japan, July 2013.
7. K. T. Javvadi Appanacharya, A. K. Tatinati, H. K. Kunderu, S. K. Mohammad, S. S. Channappayya,
A. Acharyya, S. Tripathi, "A Low-cost Scalable Solution for Digitizing Analog X-rays with
Applications to Rural Healthcare," Proc. of IEEE EMBC 2013, Osaka, Japan, July 2013.
