Engr. Dr. Muhammad Hanif

muhammad.hanif@giki.edu.pk
Ext: 2504

Engr. Dr. Muhammad Hanif

Assistant Professor

Qualifications: Ph. D. Australian National University, Canberra (2015)
Research Interests: Image Deconvolution, Sparse Image and Signal Representation, Image and Video Compression, Image Restoration, Registration and Segmentation, Object Detection and Tracking

Background

Dr. Hanif obtained his B.Sc degree (March 2002 – Jan 2006) in Computer Engineering from Electrical and Computer Engineering Department, COMSATS Institute of Information Technology, Abbottabad, Pakistan. During his bachelor’s studies, he start taking interests in digital image processing which is evident from his first research paper on image fusion published in IEEE Information Fusion 2006 during his undergrade study. Soon after his graduation, he joined COMSATS Institute of Information Technology, Islamabad as Lecturer.

He completed his M.Sc degree (Aug 2007 – Aug 2009) in Information Technology with specialization in Signal Processing from Department of Signal Processing, Faculty of Engineering Sciences at Tampere University of Technology, Tampere, Finland. His master thesis, which was part of a joint funded project from European Union (EU), focused on the digital image processing applications in paper making industry. Specifically, in his thesis he outlined the three dimensional (3D) reconstruction of fiber flocs from two orthogonal views. The reconstructed 3D flocs model is then used to get the statistical analysis like volume, dimensions, shape, and flow direction of flocs during the paper making process. The findings of his thesis were published in Journal of Engineered Fibers and Fabrics (JEFF).  After completing his master studies, he served as Researcher in Faculty of Engineering, Tampere University of Technology.

In 2010, Dr. Hanif received TUBITAK fellowship from the Scientific and Technological Research Council of Turkey to work as visiting researcher with Dr. Gozde Unal on brain image registration at Engineering and Natural Sciences, Sabanci University, Istanbul (Turkey).

Dr. Hanif pursued his PhD degree (Jun. 2011 – Jun 2015) in Image Processing from College of Engineering and Computer Science, Australian National University, Canberra, Australia. He was associated with Computer Vision and Robotics research group, National ICT Australia (NICTA), now known as DTAT61. During his PhD studies, his research was mainly focused on blind image deconvolution and sparse image processing.  Specifically, his thesis outlined some information geometric based approaches for blind image deconvolution and optimized dictionary learning methods for sparse image modeling.

Dr. Hanif received the European Research Consortium for Informatics and Mathematics (ERCIM) fellowship (Jan 2017 – Nov 2019) for his postdoc at Italian National Research Council (CNR), Pisa, Italy. He was associated with Istituto di Scienza e Tecnologie dell’Informazione “A. Faedo”, CNR. During his postdoc, he mainly work on sparse signal representation based approaches to address ancient manuscripts restoration, retrieval and classification.

Research Interests
  • Sparse Image Processing
  • Dictionary learning
  • Image deconvolution
  • Surveillance systems
  • Text and data mining
  • Deep learning
Experience

Prior to joining the GIK institute, Dr. Hanif worked as ERCIM fellow at Italian National Research Council (CNR), Pisa (Italy). He also served as Visiting Staff and course tutor at School of Engineering, University of Melbourne, Melbourne (Australia). In 2013 during his PhD studies, he was part of The Florey Institute of Neuroscience and Mental Health, Melbourne (Australia) as visiting researcher. In 2010, Dr. Hanif received TUBITAK fellowship from the Scientific and Technological Research Council of Turkey to work as visiting researcher with Dr. Gozde Unal on image registration at Engineering and Natural Sciences, Sabanci University, Istanbul (Turkey). He also served as Researcher at Faculty of Engineering Sciences, Tampere University of Technology, Tampere (Finland). Dr. Hanif also served as Lecturer at COMSATS Institute of Information Technology, Islamabad (2007 – 2009).

Professional Affiliations
  • Registered Member, Australian Computer Society. (Jan, 2014 – to-date)
  • Reviewer, IEEE Transaction on Signal Processing (May, 2018 – to-date)
  • Reviewer, Journal of Imaging (Feb2018 – to-date)
  • Reviewer, IEEE Transactions on Image Processing (Feb, 2015 – Nov, 2016)
  • Review Committee, IEEE Intl. Workshop on Statistical Signal Processing 2014
  • Student Member, IEEE Signal Processing Society (2012 – to-date)
  • Registered Member, Pakistan Engineering Council (2006 – to-date)
  • Organizing Committee, IEEE Intl. Multitopic Conference (INMIC)
Research Publications
  1. Muhammad Hanif, Anna Tonazzini, Pasquale Savino, and Emanuele Salerno Blind Bleedthrough Removal in Color Ancient Manuscripts Using GMM Based Segmentation. Int. Journal on Document Analysis and Recognition (IJDAR), -2019
  2. Muhammad Hanif, Anna Tonazzini, Pasquale Savino, and Emanuele Salerno – Non-local Sparse Image Inpainting for Document Bleed-Through Removal. Journal of Imaging, 2018.
  3. Muhammad Hanif, Anna Tonazzini, Pasquale Savino, Emanuele Salerno, Greg Tsagkatakis – Document Bleedthrough Removal using Sparse Image Inpainting. IAPR Document Analysis Systems (DAS), 2018
  4. Franca Debole, Muhammad Hanif, Anna Tonazzini- A First Step towards NLP from Digitized Manuscripts: Virtual Restoration. IEEE Machine Learning and Natural Language Processing, 2018. (Best Paper Award)
  5. Muhammad Hanif, Anna Tonazzini- Sparse Representation Based Inpainting for Bleed-Through Restoration. European Signal Processing Conference 2017 (EUSIPCO).
  6. A. K. Seghouane and Muhammad Hanif– A Sequential Dictionary Learning Algorithm with Enforced Sparsity. IEEE Int. Conference on Acousics Speech and Signal Processing (ICASSP) 2016, pp. 3876-3880.
  7. Muhammad Hanif- Non-local Noise Estimation for Adaptive Image Denoising. Int. Conference on Digital Image Computing Techniques and Applications, (DICTA- 2015)
  8. Muhammad Hanif and A. K. Seghouane –Blind Image Deblurring Using Non-negative Sparse Approximation.IEEE Int. Conference on Image Processing (ICIP-2014)
  9. Muhammad Hanif and A. K. Seghouane- An E?ective Image Restoration using Kullback- Leibler Divergence. IEEE Int. Conference on Image Processing (ICIP-2014)
  10. Muhammad Hanif and A. K. Seghouane- Maximum Likelihood Orthogonal Dictionary Learning. IEEE Int. Workshop on Statistical Signal Processing (SSP-2014)
  11. Muhammad Hanif and A. K. Seghouane- An EM Based Hybrid Fourier Wavelet Image Deconvolution Algorithm. IEEE Int. Conference on Image Processing (ICIP-2013)
  12. A. K. Seghouane and Muhammad Hanif– A Kullback-Leibler Divergence Approach For Wavelet-Base Blind Image Deconvolution. IEEE MLSP-2012
  13. Muhammad Hanif and A. K. Seghouane- An EM Based Hybrid Fourier Wavelet Image Deconvolution Algorithm. IEEE Int. Conference on Image Processing (ICIP-2013) pp. 591-595. doi:10.1109/ICIP.2013.6738122.
  14.  A. K. Seghouane and Muhammad Hanif– A Kullback-Leibler Divergence Approach For Wavelet-Base Blind Image Deconvolution. IEEE Int. Workshop on Machine Learning for Signal Processing, (MLSP-2012)pp. 1-5.
  15. Muhammad Hanif and A. K. Seghouane- Blurred Image Restoration using Gaussian Scale Mixtures Model in Wavelet Domain. International Conference on Digital Image Computing Techniques and Applications,(DICTA-2012) pp. 1-6.
  16. J.Larkomaa, M.Honkanen, Muhammad Hanif, J.Niinimaki, P.Saarenrinne- E?ect of Fibre Properties on Flocculation and Fractionation of Cellulosic Fibres in Dry State. Journal of Engineered Fibers and Fabrics (JEFF-2009) Vol.4 Issue 4. Pages 1-10.
  17. Muhammad Hanif and Usman Ali- Optimized Visual and Thermal Image Fusion for E?icient Face Recognition. IEEE Int. Conference on Information Fusion (FUSION 2006) pp. 1-6

Under Review

  1. Muhammad Hanif, Anna Tonazzini, Pasquale Savino, Emanuele Salerno Blind Degradation Removal in Color Ancient Manuscripts Using Sparse Dictionary based Segmentation. International Journal on Document Analysis and Recognition- IJDAR (Submitted)
  2. Muhammad Hanif and A. K. Seghouane– Double Sparsity Revisited: Regularized Sequential Dictionary Learning. IEEE Transaction on Signal Processing (Under Review)
  3. Muhammad Hanif and A. K. Seghouane– Regularized Sequential Dictionary Learning Algorithms for Image Restoration. Signal, Image and Video Processing (Under Review).

 

Awards
  • Research Fellow: European Research Consortium for Informatics and Mathematics (ERCIM) Fellowship (Jan 2017 – Oct 2018)
  • PhD Scholar:  Austarlian National University PhD Scholarship (Jun 2011- Jun 2015).
  • Research Scholar: National ICT Australia (NICTA)  research award (Jun 2011- Jun2015).
  • Visiting Researcher: The Scientific and Technological Research Council of Turkey (TUBITAK) research grant (2010).
  • M. Sc Scholarship: Higher Education Commission (HEC) merit scholarship (Aug 2007- Aug 2009),
  • Merit Scholarship:  Pakistan Ministry of Science merit scholarship (March 2002- March 2006).