Personal Information Academic Records Research Experience Research Publications Language Proficiency Contact Address
E-mail : [email protected]
Personal Information
Name : RAKIB AHMED
Father's Name : Md. Nazrul Islam
Sex : Male
Citizenship : Bangladeshi
Native Language : Bangla
Spouse : Surovi Sultana
Daughter : Shuhrat Rafia
Son : Rafid Abrar Ahmed



 Academic Records
Name of the Degree Year Institute Result Rank
S.S.C. 1988 Rajshahi Govt. Laboratory
High School, Bangladesh		 First
Division		 4th position in the Board
(Out of approx. 100,000 students)
H.S.C. 1990 Rajshahi College
Bangladesh		 First
Division		 14th position in the Board
(Out of approx. 100,000 students)
B.Sc(Hons) 1993
(Exam. held 1996)		 Applied Physics & Electronics
University of Rajshahi, Bangladesh		 First
Class		 2nd position in the Dept.
(Out of around 50 Students)
M.Sc(Thesis) 1994
(Exam. held 1998)		 Applied Physics & Electronics
University of Rajshahi, Bangladesh		 First
Class		 1st position in the Dept.
(Out of around 45 Students)
PhD 2008
 Faculty of Information Technology
Monash University, Australia		 Requirement fulfilled



Research Experience Leading to Academic Degrees


PATTERN RECOGNITION BY NEURAL NETWORK

Has been successfully completed for the requirement and fullfilment of the degree of Master of Science under the supervision of Dr. Ramesh Chandra Debnath, Professor, Department of Applied Physics & Electronic Engineering, Rajshahi University, Bangladesh.

Abstract of the Research Work
The thesis deals with the recognition of characters using artificial neural network. The English upper-case alphabets and Bangla Swarabarna (vowel) characters have been chosen to test the character recognition capabilities of Back propagation neural network. It also studies the generalization properties of the network using distorted patterns of the characters.


SHAPE AND TEXTURE FOR OBJECT-BASED VIDEO SEGMENTATION

Has been completed successfully for the degree of Doctor of Philosophy in the Faculty of Information Technology, Monash University, Australia under supervisions of Prof. Laurence S Dooley and Dr. Gour Chandra Karmakar.

Abstract of the Research Work
Video object segmentation is of paramount importance and a challenging contemporary research topic in multimedia and computer vision, embracing many disparate application domains, ranging from video coding and retrieval through to security, surveillance and medical imaging. Despite considerable endeavour, fully automatic, computer-based object segmentation still remains a major goal in multimedia technology research, primarily because of the inherent difficulty in semantically defining video objects by low level feature information. No single pixel feature is sufficiently all encompassing to be able to always characterise the many varieties of video objects, and so this demands the integration of high level perceptual attributes to achieve better segmentation quality.
This thesis proposes a number of innovative strategies to seamlessly integrate the perceptual attributes of shape and texture concerning an object into different video object segmentation frameworks, so bridging the perceptual gap between the human visual system and automatic object representation, and consequently furnishing the precise detection of objects-of-interest and improved segmentation. The framework comprises a series of novel algorithms with the major features being: i) the automatic representation of object shape using both contour and region-based strategies for its seamless integration in video segmentation paradigms; ii) defining and representing textural information for its application in object segmentation as a pixel feature; and iii) incorporating both shape and texture based object descriptor for the precise detection and segmentation of objects-of-interest in video sequences. An especially noteworthy characteristic of all these contributions is their generic nature and transferability to any object-based video segmentation model. A comprehensive qualitative and quantitative performance analysis is developed using various standard and medical video test sequences, with the new framework consistently exhibiting superior performance compared with contemporary techniques, in terms of both object detection and quality of segmentation.



Research Publications

Refereed Journal Papers

  1. Rakib Ahmed, G C Karmakar, and L S Dooley, Texture as a Pixel Feature for Video Object Segmentation, pp.1126-1127, Vol 44, No. 19, Electronics Letters, 2008.
  2. R Ahmed, S Sultana, R C Debnath and M M Rahman, Development of Software for the Design of Combinational Logic Circuit, Journal of Applied Science & Technology, Vol 1, No. 2, 2001, Islamic University, Kushtia, BANGLADESH.
  3. P C Barman, R Ahmed, R C Debnath, Increasing the efficiency of Backpropagation Algorithm to train the Neural Network for Bangla scripts, Journal of Applied Science & Technology, Islamic University, Kushtia, BANGLADESH (Accepted).

Full-length Refereed Conference Papers

  Tier -1 Conferences
  1. R Ahmed, G C Karmakar and L S Dooley, Incorporation Of Texture Information For Joint Spatio-Temporal Probabilistic Video Object Segmentation, 14th IEEE International Conference on Image Processing, ICIP 2007, Texas, USA.
  2. R Ahmed, L S Dooley and G C Karmakar, Probabilistic Spatio-Temporal Video Object Segmentation Using A Priori Shape Descriptor, 32nd IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2007, Hawaii, USA.
  3. R Ahmed, G C Karmakar and L S Dooley, Region-Based Shape Incorporation for Probabilistic Spatio-Temporal Video Object Segmentation, 13th IEEE International Conference on Image Processing, ICIP 2006, pp. 2445-2448, Atlanta, USA (Full-text PDF).
  4. R Ahmed, G C Karmakar and L S Dooley, Probabilistic Spatio-Temporal Video Object Segmentation Incorporating Shape Information , 31st IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006, pp. 645-648, Toulouse, FRANCE (Full-text PDF).
  Other Conferences
  1. R Ahmed, G C Karmakar and L S Dooley, Automatic Video Background Replacement Using Shape-Based Probabilistic Spatio-Temporal Object Segmentation, 6th IEEE International Conference on Information, Communications and Signal Processing, ICICS 2007, Singapore.
  2. R Ahmed, G C Karmakar and L S Dooley, Efficient Probabilistic Spatio-Temporal Video Object Segmentation, 6th IEEE International Conference on Computer and Information Science, ICIS 2007, Melbourne, AUSTRALIA.
  3. R Ahmed, Wavelet-based Image Compression Using Support Vector Machine Learning and Encoding Techniques , 8th IASTED International Conference on Computer Graphics and Imaging, CGIM 2005, Honolulu, USA (Full-text PDF).
  4. P C Barman and R Ahmed, Comparison of Genetic Algorithm and Bisection Methods in the Numerical Solution of Transcendental Equations, 6th International Conference on Computing and Information Technology, ICCIT 2003, BANGLADESH.
  5. R Ahmed and R C Debnath, Recognition of Bengali Alphabet by neural Network, 7th RIEC International Symposium on Non-linear Theories and its Applications, NOLTA 2001, JAPAN.
  6. R Ahmed and R C Debnath, Recognition of English Alphabets by Neural Network, 4th International Conference on Computing and Information Technology, ICCIT 2001, BANGLADESH.


Non-Refereed Technical Reports
  1. R Ahmed, G C Karmakar and L S Dooley, Texture as a pixel feature for video object segmentation , TR-2008/2, GSIT , Monash University, Australia. (Full-text PDF)
  2. R Ahmed, G C Karmakar and L S Dooley, Spatio-Temporal Probabilistic Video Object Segmentation Incorporating Texture Information , TR-2007/1, GSIT , Monash University, Australia.
  3. R Ahmed, G C Karmakar and L S Dooley, Probabilistic Spatio-Temporal Video Object Segmentation Using a priori Shape Information , TR-2006/5, GSIT , Monash University, Australia.
  4. R Ahmed, G C Karmakar and L S Dooley, Incorporating Region-Based Shape Information for Probabilistic Spatio-Temporal Video Object Segmentation, TR-2006/1, GSIT , Monash University, Australia.
  5. R Ahmed, G C Karmakar and L S Dooley, Probabilistic Spatio-Temporal Video Object Segmentation Incorporating Generic Shape Information , TR-2005/9, GSIT , Monash University, Australia.





Language Proficiency


Bangla (Bengali) - Mother language

English - Reading/Writing/Speaking

Arabic - Reading/Writing






 Contact Details
Business Address Current Residence Permanent Residence
School of Accounting & Business Information Systems
The Australian National University, Canberra
ACT 0200, Australia.
 		13/18 Zamia Place
Palmerston
ACT 2913
Australia.
Tel : +61 2 6262 2554
210/226
Chotto Bonogram
Professor Para
Rajshahi-6203
Bangladesh.
Tel : +880 721 760168
E-mail : [email protected] [email protected]  
Website : www.geocities.com/rakahmed
Personal Information Academic Records Research Experience Research Publications Language Proficiency Contact Address

Updated on June 2008
Hosted by www.Geocities.ws

1