Research Interests:
Research Interests:
Abstract—This paper addresses the development of an intelligent vision system for human-robot interaction. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual... more
Abstract—This paper addresses the development of an intelligent vision system for human-robot interaction. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual attributes of human skin colors and geometrical analysis of face skeleton. This paper introduces a spatial domain filtering method named ‘Fuzzily skewed filter ’ which incorporates Fuzzy rules for deciding the gray level of pixels in the image in their neighborhoods and takes advantages of both the median and averaging filters. The effectiveness of the method has been justified over implementing the eye tracking commands to an entertainment robot, named “AIBO”. Keywords—Fuzzily skewed filter, human-robot interface, rms contrast, skin color segmentation. I.
Abstract—Gabor-based face representation has achieved enormous success in face recognition. This paper addresses a novel algorithm for face recognition using neural networks trained by Gabor features. The system is commenced on convolving... more
Abstract—Gabor-based face representation has achieved enormous success in face recognition. This paper addresses a novel algorithm for face recognition using neural networks trained by Gabor features. The system is commenced on convolving a face image with a series of Gabor filter coefficients at different scales and orientations. Two novel contributions of this paper are: scaling of rms contrast and introduction of fuzzily skewed filter. The neural network employed for face recognition is based on the multilayer perceptron (MLP) architecture with backpropagation algorithm and incorporates the convolution filter response of Gabor jet. The effectiveness of the algorithm has been justified over a face database with images captured at different illumination conditions. Keywords—Fuzzily skewed filter, Gabor filter, rms contrast, neural network. I.
Research Interests:
This paper presents an approach to recognize human facial expressions for human-robot interaction. For this, the facial features, especially eyes and lip are extracted and approximated using Bézier curves representing the relationship... more
This paper presents an approach to recognize human facial expressions for human-robot interaction. For this, the facial features, especially eyes and lip are extracted and approximated using Bézier curves representing the relationship between the motion of features and changes of expressions. For face detection, color segmentation based on the novel idea of fuzzy classification has been employed that manipulates ambiguity in colors. Experimental results demonstrate that this technique can robustly classify skin region and non-skin region. In order to decide whether the skin region is face or not, largest connectivity analysis has been employed. This method can recognize the facial expression category, as well as the degree of facial expression change. Finally, the system has been implemented by issuing facial expression commands to a manipulator robot. Key words:
Research Interests:
This research is concerned with the study of different improved and faster back-propagation (BP) algorithms of neural networks and the analysis of recognition result in continuous Bangla speech. For speech recognition, a comparison study... more
This research is concerned with the study of different improved and faster back-propagation (BP) algorithms of neural networks and the analysis of recognition result in continuous Bangla speech. For speech recognition, a comparison study on neural networks and speech recognition result analysis with different improved and faster BP algorithms (such as, BP with momentum, variable learning rate BP, resilient BP, conjugate gradient BP and Levenberg-Marquardt BP algorithms) have been done. In this research, the MATLAB Neural Network Toolbox 7.12.0 is used to create, train and simulate the feedforward neural network with the BP learning algorithm. The convergence obtained from standard BP algorithm is very slow; that's why, this research proposes different improved and faster BP algorithms to solve the speech recognition problems. The developed system has been justified by several networks trained with different Bangla speech words. To test the performance of the system, 20 samples of 50 Bangla speech words have been used; from which 10 samples of 50 words are used as training pattern and another 10 samples of 50 words are used as testing pattern in the network. The binary features of speech words have been generated using dynamic thresholding algorithm. The recognition system has been achieved recognition rate of 83% using resilient BP algorithm, 90% using conjugate gradient BP algorithm and 90% using Levenberg-Marquardt BP algorithm, respectively, for recognising 50 speech words.
Research Interests:
Research Interests:
This paper presents a multi-resolution masks based pattern matching method for person identification. The system is commenced with the construction of multi-resolution mask cluster pyramid, where the mask size is chosen depending on the... more
This paper presents a multi-resolution masks based pattern matching method for person identification. The system is commenced with the construction of multi-resolution mask cluster pyramid, where the mask size is chosen depending on the distance between two eyes, computed from the detected face. Experimental results show the effectiveness of the system with significantly higher precision, recall rates and matching probability comparing with conventional single resolution mask based person identification systems. This paper ...
Research Interests:
gestures forhumanmachineinteraction. One is This paperpresntsviionasgloved basedapproach [1] thatrequires wearing of reoitionsyste an-robotbisisthe cumbersome contact devices andgenerally carrying a recognitionsystem isb d... more
gestures forhumanmachineinteraction. One is This paperpresntsviionasgloved basedapproach [1] thatrequires wearing of reoitionsyste an-robotbisisthe cumbersome contact devices andgenerally carrying a recognitionsystem isb d onvsumalnfoorati f t The f loadofcables that connect thedevice toacomputer. gestuemis reognion by connectedcontentac Another approach isvisionbasedtechnique that does gestres ecogitio by onneted ompoent notrequire wearing anyofcontact devices with analysis oftheskin color segmentation ofimages in ntrqiewaigaymfcnatdvcswt HnalysVsoftheskin color mdla PC bsedmenpatten-imathing humanbody part, butusesasetofvideo camerasand HSV color modelandPCAbasedpattemn-matching strategies. Ongesture recognition, therobot isbeingcomputer visiontechniques tointerpret gestures. instructed t p Gesture recognition based on vision technology has insrucedoprfoncrtan tsksbyssung beenemerging withtherapid development of commands. Thesystem hasbeendemonstrated with theimplementation ofthea...
Research Interests:
Summary This paper presents an approach to recognize human facial expressions for human-robot interaction. For this, the facial features, especially eyes and lip are extracted and approximated using Bezier curves representing the... more
Summary This paper presents an approach to recognize human facial expressions for human-robot interaction. For this, the facial features, especially eyes and lip are extracted and approximated using Bezier curves representing the relationship between the motion of features and changes of expressions. For face detection, color segmentation based on the novel idea of fuzzy classification has been employed that manipulates ambiguity in colors. Experimental results demonstrate that this technique can robustly classify skin region and non-skin region. In order to decide whether the skin region is face or not, largest connectivity analysis has been employed. This method can recognize the facial expression category, as well as the degree of facial expression change. Finally, the system has been implemented by issuing facial expression commands to a manipulator robot.
Research Interests:
Research Interests:
This paper provides a robust method for solving transcendental equations. The approach, based on the Genetic Algorithm, is commenced with the evaluation of the mathematical equations by their fitness ratio. As the genetic algorithm is a... more
This paper provides a robust method for solving transcendental equations. The approach, based on the Genetic Algorithm, is commenced with the evaluation of the mathematical equations by their fitness ratio. As the genetic algorithm is a computationally expensive process, the searching space for possible solutions is limited to possible chromosomes for which the function values are closest to zero. These chromosomes are then selected for the next generation using roulette wheel strategy. Thus, the required processing timing is greatly reduced. Experimental results demonstrate that this process works reliably and gives better results than the well-known False Position method and the Bisection method for the numerical solution of transcendental equations.
Twitter is an online social network or a news media where users can post their desirable topical interests in the form tweets. This is a networking model where each user can choose who can follow her and whom she wants to follow. We can... more
Twitter is an online social network or a news media where users can post their desirable topical interests in the form tweets. This is a networking model where each user can choose who can follow her and whom she wants to follow. We can find the users who are very active in the social networks and consider them as influential users. This research addresses on Temporal Twitter Ranking (TTRank) to rank the influential users in Twitter. We apply Twitter-LDA topic modeling method to find the users’ topical interests. The time interval is an important factor as users’ topical interest can change over time i.e. users’ have different degree of topical interests at different timeinterval. So we give more emphasize on users’ most recent tweets. Our proposed approach also considers the impact of “Follower Influence” and “Retweet Influence”. The top influential users have been detected across different time intervals based on all the above mentioned factors and classified as “Highly Influentia...
This paper addresses the development of an intelligent vision system for human-robot interaction. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual... more
This paper addresses the development of an intelligent vision system for human-robot interaction. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual attributes of human skin colors and geometrical analysis of face skeleton. This paper introduces a spatial domain filtering method named ‘Fuzzily skewed filter’ which incorporates Fuzzy rules for deciding the gray level of pixels in the image in their neighborhoods and takes advantages of both the median and averaging filters. The effectiveness of the method has been justified over implementing the eye tracking commands to an entertainment robot, named “AIBO”. Keywords—Fuzzily skewed filter, human-robot interface, rms contrast, skin color segmentation.
Research Interests:
This paper addresses a method to analyze syntactically Bangla sentence using context-sensitive grammar rules which accept almost all types of Bangla sentences including simple, complex and compound sentences and then interpret the input... more
This paper addresses a method to analyze syntactically Bangla sentence using context-sensitive grammar rules which accept almost all types of Bangla sentences including simple, complex and compound sentences and then interpret the input Bangla sentence to English using the NLP conversion unit. The grammar rules employed for this system allow parsing five categories of sentences according to Bangla intonation. The system is based on analyzing an input sentence and converting into a structural representation (SR). Once the SR is created for a particular sentence it is then converted to corresponding English sentence by NLP conversion unit. For conversion, the NLP conversion unit takes help of the Corpus. The effectiveness of this method has been justified over the demonstration of different Bangla sentences with 28 decomposition rules and the success rates in all cases are over 90%.
Research Interests:
Research Interests:
Abstract Protein coding regions are commonly diffused with non-coding regions due to 1/f background noise in such a way that a viable discernment between the two regions becomes cumbersome. Commonly employed digital signal processing... more
Abstract Protein coding regions are commonly diffused with non-coding regions due to 1/f background noise in such a way that a viable discernment between the two regions becomes cumbersome. Commonly employed digital signal processing methodologies lack fundamental genetic code context of nucleotides since these approaches treat DNA signal as normal digital signal that could be processed by traditional DSP tools and techniques. This paper reviews the prevailing approaches for protein coding regions identification that base on common DSP concepts and highlights the importance of genetic code context to be considered for any computational solution for protein coding regions identification. Nucleotides in a DNA signal carry certain natural characteristics i.e. presence in a distinctive triplet format, maintaining distinct structure, owning and further sharing distribution of densities in codons, fuzzy behaviors, semantic similarities, unbalanced nucleotides' distribution producing a relatively high bias for nucleotides' usage in coding regions etc. The computational solutions for protein coding regions identification that exploit these fundamental characteristic of nucleotides can significantly suppress the signal noise and hence can better contribute in identification.
Research Interests:
... This eye tracker system has been employed for the implementation of a human-robot interface. ... Eye tracking A new interface for visual exploration. ... Readership Statistics. 1 Reader on Mendeley. by Discipline. 100% Engineering.... more
... This eye tracker system has been employed for the implementation of a human-robot interface. ... Eye tracking A new interface for visual exploration. ... Readership Statistics. 1 Reader on Mendeley. by Discipline. 100% Engineering. by Academic Status. 100% Ph.D. Student. by ...
Research Interests:
Research Interests:
Image processing employing software routines are time consuming in compared to their counterfeits of hardware strategies. Cellular automata appear as abstract models of systems in image manipulation due to their local nature, inherently... more
Image processing employing software routines are time consuming in compared to their counterfeits of hardware strategies. Cellular automata appear as abstract models of systems in image manipulation due to their local nature, inherently parallel and computationally simple implementation. This paper addresses a hardware framework based on two dimensional cellular automata, for manipulating images in certain specific positions and orientations. The proposed architecture can be easily implemented using VLSI technology.
Research Interests:
Research Interests:
Research Interests:
Research Interests: Psychology, Human Computer Interaction, Robot Vision, Real Time Computing, Facial expression, and 11 moreFace Detection, Face Modeling, Facial Features, Learning Object, Human Robot Interaction, Color Image Segmentation, Fuzzy Classification, Experimental Evaluation, Facial Action Coding System, Skin Color, and Facial Expression Recognition
Abstract This paper presents a multi-resolution masks based pattern matching method for person identification. The system is commenced with the construction of multi-resolution mask cluster pyramid, where the mask size is chosen depending... more
Abstract This paper presents a multi-resolution masks based pattern matching method for person identification. The system is commenced with the construction of multi-resolution mask cluster pyramid, where the mask size is chosen depending on the distance between two eyes, computed from the detected face. Experimental results show the effectiveness of the system with significantly higher precision, recall rates and matching probability comparing with conventional single resolution mask based person identification systems.
Research Interests:
Abstract This paper presents a vision based gesture recognition system for human-robot symbiosis. The system is based on the visual information of the face and is commenced with the recognition of face gesture by connected component... more
Abstract This paper presents a vision based gesture recognition system for human-robot symbiosis. The system is based on the visual information of the face and is commenced with the recognition of face gesture by connected component analysis of the skin color segmentation of images in HSV color model and neural network based pattern-matching strategies. On gesture recognition, robot is being instructed to perform certain tasks by issuing commands. The system is capable of recognizing static gestures comprised of the ...
Research Interests:
Research Interests:
A vision system has been developed for human-robot symbiosis. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual attributes of human skin colors and... more
A vision system has been developed for human-robot symbiosis. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual attributes of human skin colors and geometrical analysis of face skeleton. This paper introduces a spatial domain filtering method named dasiafuzzily skewed filterpsila which incorporates fuzzy rules for deciding the gray level of pixels in the image in their neighborhoods and takes advantages of both the median and averaging filters. The effectiveness of the method has been justified over human-robot interaction system.
Research Interests:
Research Interests:
AbstractThis paper addresses a novel algorithm for face recognition using neural networks trained by Image patterns those are achieved from Gabor features. The system commences on convolving some morphed images of particular face with a... more
AbstractThis paper addresses a novel algorithm for face recognition using neural networks trained by Image patterns those are achieved from Gabor features. The system commences on convolving some morphed images of particular face with a series of Gabor filter ...
Research Interests:
Research Interests:
Image enhancement is a technique to improve the quality of an image. The aim of image enhancement technique is to improve the interpretability or perception of information in images for human viewers, or to provide better input for other... more
Image enhancement is a technique to improve the quality of an image. The aim of image enhancement technique is to improve the interpretability or perception of information in images for human viewers, or to provide better input for other automated image processing techniques. This paper presents a new approach for image enhancement with fuzzy rule-based filtering. Compared to other non-linear
This article presents a new method to determine disparity map useful for three-dimensional (3D) scene reconstruction. The main task behind the computation of disparity map is stereo correspondence matching. In recent years, several stereo... more
This article presents a new method to determine disparity map useful for three-dimensional (3D) scene reconstruction. The main task behind the computation of disparity map is stereo correspondence matching. In recent years, several stereo matching algorithms have been developed to find corresponding pairs in two images: left and right images captured by a stereo camera. But these algorithms exhibit a very high computational cost. With a view to reduce the computation time and produce a smooth and detailed disparity map, a fast and new approach based on average disparity estimation is proposed in this research, which can tackle additive noise. Experimental results confirm that the method achieves a substantial gain in accuracy with less expense of computation time. Key Words: Disparity map, Stereo correspondence, Stereo Vision, 3D Scene Reconstruction. DOI: 10.3329/diujst.v4i1.4348 Daffodil International University Journal of Science and Technology Vol.4(1) 2009 pp.9-13
Research Interests:
Research Interests:
... Al-Amin Bhuiyan¥ * Northern University, Dhaka, Bangladesh, e-mail: rupak_karmadhar@yahoo. com Northern University, Dhaka, Bangladesh, e ... com University of Dhaka, Bangladesh, e-mail: farhan717@univdhaka.edu ¥ Jahangirnagar... more
... Al-Amin Bhuiyan¥ * Northern University, Dhaka, Bangladesh, e-mail: rupak_karmadhar@yahoo. com Northern University, Dhaka, Bangladesh, e ... com University of Dhaka, Bangladesh, e-mail: farhan717@univdhaka.edu ¥ Jahangirnagar University, Savar, Dhaka, Bangladesh ...
Research Interests:
ABSTRACT This paper introduces a new concept for the establishment of human-machine symbiotic relationship through interactive image recognition. The system is based on the implementation of knowledge-based image processing methodologies... more
ABSTRACT This paper introduces a new concept for the establishment of human-machine symbiotic relationship through interactive image recognition. The system is based on the implementation of knowledge-based image processing methodologies for intelligent recognition of image and intelligent responding for an autonomous robot or machine. This paper includes two novel contributions of image application; one is image metamorphosis another is image recognition for intelligent system. Using these two important image applications we have achieved the interactive image recognition for human-machine interfacing. The system is organized with the learning and recognition of images using neural network with back-propagation algorithm. Images have been processed and converted to feature vectors that are compared with the feature vectors of a training set of images. This image recognition system can play a big role over establishing more perfect human-machine interfacing. The proposed system has been tested extensively with a large data set and it performs accuracy of 93.33% on that database.
Research Interests:
Image enhancement is a technique to improve the quality of an image. The aim of image enhancement technique is to improve the interpretability or perception of information in images for human viewers, or to provide better input for other... more
Image enhancement is a technique to improve the quality of an image. The aim of image enhancement technique is to improve the interpretability or perception of information in images for human viewers, or to provide better input for other automated image processing techniques. This paper presents a new approach for image enhancement with fuzzy rule-based filtering. Compared to other non-linear
A vision system has been developed for human-robot symbiosis. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual attributes of human skin colors and... more
A vision system has been developed for human-robot symbiosis. The two novel contributions of this paper are 1) Detection of human faces and 2) Localizing the eye. The method is based on visual attributes of human skin colors and geometrical analysis of face skeleton. This paper introduces a spatial domain filtering method named dasiafuzzily skewed filterpsila which incorporates fuzzy rules for deciding the gray level of pixels in the image in their neighborhoods and takes advantages of both the median and averaging filters. The effectiveness of the method has been justified over human-robot interaction system.
Research Interests:
Research Interests:
... This eye tracker system has been employed for the implementation of a human-robot interface. ... Eye tracking A new interface for visual exploration. ... Readership Statistics. 1 Reader on Mendeley. by Discipline. 100% Engineering.... more
... This eye tracker system has been employed for the implementation of a human-robot interface. ... Eye tracking A new interface for visual exploration. ... Readership Statistics. 1 Reader on Mendeley. by Discipline. 100% Engineering. by Academic Status. 100% Ph.D. Student. by ...
Research Interests:
This paper presents a speech recognition front-end system for segmenting continuously spoken Bangla speech and extracting features from the segmented words/sub-words including windows. The system commenced with recording the original... more
This paper presents a speech recognition front-end system for segmenting continuously spoken Bangla speech and extracting features from the segmented words/sub-words including windows. The system commenced with recording the original speech sentences ...
Research Interests:
AbstractThis paper addresses a novel algorithm for face recognition using neural networks trained by Image patterns those are achieved from Gabor features. The system commences on convolving some morphed images of particular face with a... more
AbstractThis paper addresses a novel algorithm for face recognition using neural networks trained by Image patterns those are achieved from Gabor features. The system commences on convolving some morphed images of particular face with a series of Gabor filter ...
Research Interests:
Research Interests:
... Al-Amin Bhuiyan¥ * Northern University, Dhaka, Bangladesh, e-mail: rupak_karmadhar@yahoo. com Northern University, Dhaka, Bangladesh, e ... com University of Dhaka, Bangladesh, e-mail: farhan717@univdhaka.edu ¥ Jahangirnagar... more
... Al-Amin Bhuiyan¥ * Northern University, Dhaka, Bangladesh, e-mail: rupak_karmadhar@yahoo. com Northern University, Dhaka, Bangladesh, e ... com University of Dhaka, Bangladesh, e-mail: farhan717@univdhaka.edu ¥ Jahangirnagar University, Savar, Dhaka, Bangladesh ...
Research Interests:
This paper presents a robust and novel approach to find the roots of mathematical equations using genetic algorithm. A modified approach of genetic algorithm has been provided where fuzzy logic has been used for chromosomes selection. A... more
This paper presents a robust and novel approach to find the roots of mathematical equations using genetic algorithm. A modified approach of genetic algorithm has been provided where fuzzy logic has been used for chromosomes selection. A GA formula has also been presented to solve equations with multiple iterations in order to justify the error versus iteration graph. Finally, a
Research Interests:
This article presents a new method to determine disparity map useful for three-dimensional (3D) scene reconstruction. The main task behind the computation of disparity map is stereo correspondence matching. In recent years, several stereo... more
This article presents a new method to determine disparity map useful for three-dimensional (3D) scene reconstruction. The main task behind the computation of disparity map is stereo correspondence matching. In recent years, several stereo matching ...
Research Interests:
Summary This paper presents an approach to recognize human facial expressions for human-robot interaction. For this, the facial features, especially eyes and lip are extracted and approximated using Bézier curves representing the... more
Summary This paper presents an approach to recognize human facial expressions for human-robot interaction. For this, the facial features, especially eyes and lip are extracted and approximated using Bézier curves representing the relationship between the motion of features and changes of expressions. For face detection, color segmentation based on the novel idea of fuzzy classification has been employed that manipulates
Research Interests:
This paper presents a vision based gesture recognition system for human-robot symbiosis. The system is based on the visual information of the face and is commenced with the recognition of face gesture by connected component analysis of... more
This paper presents a vision based gesture recognition system for human-robot symbiosis. The system is based on the visual information of the face and is commenced with the recognition of face gesture by connected component analysis of the skin color segmentation of images in HSV color model and neural network based pattern-matching strategies. On gesture recognition, robot is being instructed to perform certain tasks by issuing commands. The system is capable of recognizing static gestures comprised of the ...