Pronaya P Das | University of Göttingen (original) (raw)

Papers by Pronaya P Das

Research paper thumbnail of Detecting Animal Contacts—A Deep Learning-Based Pig Detection and Tracking Approach for the Quantification of Social Contacts

Sensors, 2021

The identification of social interactions is of fundamental importance for animal behavioral stud... more The identification of social interactions is of fundamental importance for animal behavioral studies, addressing numerous problems like investigating the influence of social hierarchical structures or the drivers of agonistic behavioral disorders. However, the majority of previous studies often rely on manual determination of the number and types of social encounters by direct observation which requires a large amount of personnel and economical efforts. To overcome this limitation and increase research efficiency and, thus, contribute to animal welfare in the long term, we propose in this study a framework for the automated identification of social contacts. In this framework, we apply a convolutional neural network (CNN) to detect the location and orientation of pigs within a video and track their movement trajectories over a period of time using a Kalman filter (KF) algorithm. Based on the tracking information, we automatically identify social contacts in the form of head–head an...

Research paper thumbnail of Image-based approach for the detection of counterfeit banknotes of Bangladesh

2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), 2016

Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is ... more Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is now a common phenomenon due to advanced printing and scanning technology. Bangladesh has been facing serious problem by the increasing rate of fake notes in the market. To get rid of this problem various fake note detection methods are available around the world and most of these are hardware based and costly. In the present paper an automated image-based technique is described for the detection of fake banknotes of Bangladesh. Security features of banknotes such as watermark, micro-printing and hologram etc. are extracted from the banknote images and then detection is performed using Support Vector Machine (SVM). Experimental results confirm the effectiveness of the proposed algorithm.

Research paper thumbnail of Solving Maximum Clique Problem using a novel Quantum-inspired Evolutionary Algorithm

2015 International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), 2015

ABSTRACT

Research paper thumbnail of Detecting Animal Contacts—A Deep Learning-Based Pig Detection and Tracking Approach for the Quantification of Social Contacts

Sensors, 2021

The identification of social interactions is of fundamental importance for animal behavioral stud... more The identification of social interactions is of fundamental importance for animal behavioral studies, addressing numerous problems like investigating the influence of social hierarchical structures or the drivers of agonistic behavioral disorders. However, the majority of previous studies often rely on manual determination of the number and types of social encounters by direct observation which requires a large amount of personnel and economical efforts. To overcome this limitation and increase research efficiency and, thus, contribute to animal welfare in the long term, we propose in this study a framework for the automated identification of social contacts. In this framework, we apply a convolutional neural network (CNN) to detect the location and orientation of pigs within a video and track their movement trajectories over a period of time using a Kalman filter (KF) algorithm. Based on the tracking information, we automatically identify social contacts in the form of head–head an...

Research paper thumbnail of Image-based approach for the detection of counterfeit banknotes of Bangladesh

2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), 2016

Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is ... more Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is now a common phenomenon due to advanced printing and scanning technology. Bangladesh has been facing serious problem by the increasing rate of fake notes in the market. To get rid of this problem various fake note detection methods are available around the world and most of these are hardware based and costly. In the present paper an automated image-based technique is described for the detection of fake banknotes of Bangladesh. Security features of banknotes such as watermark, micro-printing and hologram etc. are extracted from the banknote images and then detection is performed using Support Vector Machine (SVM). Experimental results confirm the effectiveness of the proposed algorithm.

Research paper thumbnail of Solving Maximum Clique Problem using a novel Quantum-inspired Evolutionary Algorithm

2015 International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), 2015

ABSTRACT

Research paper thumbnail of Single Image Face Recognition based on Gabor, Sobel and Local Ternary Pattern

International Journal of Computer Applications, 2015

Research paper thumbnail of Optimal coverage of Wireless Sensor Network using Termite Colony Optimization Algorithm

2015 International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), 2015

Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) ha... more Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) has been become one of the most attractive and intriguing research area in last few years. Large-scale sensor networks has been deployed by interconnecting several hundred to a few thousand sensors nodes opens up diverse technical challenges and tremendous application possibilities. This paper presents a Termite Colony Optimization (TCO) algorithm, an adaptable and balanced way, which optimizes a minimal number of sensors while covering a maximum area. TCO is a population based metaheuristic approach developed from the intelligent behavior of termite that is known to us as Swarm Intelligence. It has the ability to solve the presented problem in efficient way that has shown more effectiveness and robustness compared to other population based algorithms. In this paper, we have justified our simulation results with the results of other papers. After that we have experimented with varying different parameters like grid size, number of sensors and termites. It has also proved the capability of our TCO over the constraints and difficulties as well.

Research paper thumbnail of Quantum-Inspired Evolutionary Algorithm to Solve Graph Coloring Problem

Research paper thumbnail of Identification of regulatory SNPs associated with genetic modifications in lung adenocarcinoma

Research paper thumbnail of Document’s title

Research paper thumbnail of An Effective Quantum-inspired Evolutionary Algorithm for Finding Degree-constrained Minimum Spanning Tree

Research paper thumbnail of Bangladeshi dialect recognition using Mel Frequency Cepstral Coefficient, Delta, Delta-delta and Gaussian Mixture Model

2016 Eighth International Conference on Advanced Computational Intelligence (ICACI), 2016

Automatic recognition systems are generally applied successfully in speech processing to categori... more Automatic recognition systems are generally applied successfully in speech processing to categorize observed utterances by the speaker identity, dialect and linguistic communication. A lot of research has been performed to detect speeches, dialects and languages of different region throughout the world. But the work on dialects of Bangladesh is infrequent to our research. These dialects, in turn, differ quite a bit from each other. In this paper, we present a method to detect Bangladeshi different dialects which utilizes Mel Frequency Cepstral Coefficient (MFCC), its Delta and Delta-delta as main features and Gaussian Mixture Models (GMM) to classify characteristics of a specific dialect. Particularly we extract the MFCCs, Deltas and Delta-deltas from the speech signal. Then they are merged together to form a feature vector for a specific dialect. GMM is trained using the iterative Expectation Maximization (EM) algorithm where feature vectors are served as input. This scheme is tested on 5 databases of 30 speech samples each. Speech samples contain dialects of Borishal, Noakhali, Sylhet, Chittagong and Chapai Nawabganj regions of Bangladesh. Experiments show that GMM adaptation gives comparable good performance.

Research paper thumbnail of Identification of Regulatory SNPs Associated with Vicine and Convicine Content of Vicia faba Based on Genotyping by Sequencing Data Using Deep Learning

Genes, 2020

Faba bean (Vicia faba) is a grain legume, which is globally grown for both human consumption as w... more Faba bean (Vicia faba) is a grain legume, which is globally grown for both human consumption as well as feed for livestock. Despite its agro-ecological importance the usage of Vicia faba is severely hampered by its anti-nutritive seed-compounds vicine and convicine (V+C). The genes responsible for a low V+C content have not yet been identified. In this study, we aim to computationally identify regulatory SNPs (rSNPs), i.e., SNPs in promoter regions of genes that are deemed to govern the V+C content of Vicia faba. For this purpose we first trained a deep learning model with the gene annotations of seven related species of the Leguminosae family. Applying our model, we predicted putative promoters in a partial genome of Vicia faba that we assembled from genotyping-by-sequencing (GBS) data. Exploiting the synteny between Medicago truncatula and Vicia faba, we identified two rSNPs which are statistically significantly associated with V+C content. In particular, the allele substitutions regarding these rSNPs result in dramatic changes of the binding sites of the transcription factors (TFs) MYB4, MYB61, and SQUA. The knowledge about TFs and their rSNPs may enhance our understanding of the regulatory programs controlling V+C content of Vicia faba and could provide new hypotheses for future breeding programs.

Research paper thumbnail of Image-Based Approach for the Detection of Counterfeit Banknotes of Bangladesh

Image-Based Approach for the Detection of Counterfeit Banknotes of Bangladesh

Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is ... more Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is now a common phenomenon due to advanced printing and scanning technology. Bangladesh has been facing serious problem by the increasing rate of fake notes in the market. To get rid of this problem various fake note detection methods are available around the world and most of these are hardware based and costly. In the present paper an automated image-based technique is described for the detection of fake banknotes of Bangladesh. Security features of banknotes such as watermark, micro-printing and hologram etc. are extracted from the banknote images and then detection is performed using Support Vector Machine (SVM). Experimental results confirm the effectiveness of the proposed algorithm.

Research paper thumbnail of Bangladeshi Dialect Recognition using Mel Frequency Cepstral Coefficient, Delta, Delta-delta and Gaussian Mixture Model

Automatic recognition systems are generally applied successfully in speech processing to categori... more Automatic recognition systems are generally applied successfully in speech processing to categorize observed utterances by the speaker identity, dialect and linguistic communication. A lot of research has been performed to detect speeches, dialects and languages of different region throughout the world. But the work on dialects of Bangladesh is infrequent to our research. These dialects, in turn, differ quite a bit from each other. In this paper, we present a method to detect Bangladeshi different dialects which utilizes Mel Frequency Cepstral Coefficient (MFCC), its Delta and Delta-delta as main features and Gaussian Mixture Models (GMM) to classify characteristics of a specific dialect. Particularly we extract the MFCCs, Deltas and Delta-deltas from the speech signal. Then they are merged together to form a feature vector for a specific dialect. GMM is trained using the iterative Expectation Maximization (EM) algorithm where feature vectors are served as input. This scheme is tested on 5 databases of 30 speech samples each. Speech samples contain dialects of Borishal, Noakhali, Sylhet, Chittagong and Chapai Nawabganj regions of Bangladesh. Experiments show that GMM adaptation gives comparable good performance.

Research paper thumbnail of Single Image Face Recognition based on Gabor, Sobel and Local Ternary Pattern

Face recognition has been a fast growing, challenging and fascinating area in real time applicati... more Face recognition has been a fast growing, challenging and fascinating area in real time applications. A large number of face recognition algorithms have been developed in last decades. This paper presents a face recognition approach that utilizes the Sobel operator, Local Ternary Pattern (LTP) descriptor, Gabor features and Principal Component Analysis (PCA) algorithms to attain enhanced recognition accuracy where training set has only one image per person. In particular, the edge information of face image is enhanced using Sobel operator. Then we use LTP on the image to encode the micro-level information of spots, edges and other local characteristics. Finally Gabor-wavelets based features are then extracted and their histograms are concatenated together into a contiguous histogram to be used as a descriptor of the face. As Gabor features cause a very high dimensional histogram vector, therefore PCA is used to reduce the dimension. This approach is compared with the original LBP, LTP and Sobel LTP on gray-level images for face recognition. The experimental results exhibit that our method provides a remarkable performance under various conditions.

Research paper thumbnail of Optimal Coverage of Wireless Sensor Network using Termite Colony Optimization Algorithm

Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) ha... more Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) has been become one of the most attractive and intriguing research area in last few years. Large-scale sensor networks has been deployed by interconnecting several hundred to a few thousand sensors nodes opens up diverse technical challenges and tremendous application possibilities. This paper presents a Termite Colony Optimization (TCO) algorithm, an adaptable and balanced way, which optimizes a minimal number of sensors while covering a maximum area. TCO is a population based metaheuristic approach developed from the intelligent behavior of termite that is known to us as Swarm Intelligence. It has the ability to solve the presented problem in efficient way that has shown more effectiveness and robustness compared to other population based algorithms. In this paper, we have justified our simulation results with the results of other papers. After that we have experimented with varying different parameters like grid size, number of sensors and termites. It has also proved the capability of our TCO over the constraints and difficulties as well.

Research paper thumbnail of Solving Maximum Clique Problem using a Novel Quantum-inspired Evolutionary Algorithm

Maximum Clique Problem (M CP) is one of the most important NP-hard problems in the area of soft c... more Maximum Clique Problem (M CP) is one of the most important NP-hard problems in the area of soft computing and it has many real world applications in numerous fields ranging from coding theory to the determination of the structure of a protein molecule. Different heuristic, Meta heuristic and hybrid solution approaches have been applied to obtain the solution. In this paper, we demonstrate a Quantum-inspired Evolutionary Algorithm (QEA) to solve MCP. We have used one dimensional arrays ofQ­ bits called Q-bit individuals to produce binary individuals. After production of binary individuals, we have repaired and improved them. Here, Q-gate is the main variation operator applied on Q­ bit individuals. Our algorithm was tested on DIMACS benchmark graphs and 40 of them were tested. The results obtained here are extremely encouraging. For almost all of the datasets, we get the optimal results reported on DIMACS benchmark and also compared our results with other related works. For some cases we get better results than other works.

Research paper thumbnail of An Effective Quantum-inspired Evolutionary Algorithm for Finding Degree-constrained Minimum Spanning Tree

The Degree-constrained Minimum Spanning Tree (d-MST) is an extended adaptation of general Minimum... more The Degree-constrained Minimum Spanning Tree (d-MST) is an extended adaptation of general Minimum Spanning Tree (MST) problem. This problem is pertinent in the design of communication networks. It consists of finding a spanning tree whose nodes do not exceed a given maximum degree and whose total edge length is minimum. In this formulation the problem turns into NP-hard, therefore meta heuristic approaches like Ant Colony Optimization, Simulated Annealing, Evolutionary Algorithms etc. are suitable for solving d-MST problem. In this paper, we demonstrate a Quantum-inspired Evolutionary Algorithm (QEA) that solves instances of the problem to optimality. We have used one dimensional array of Q-bits called Q-bit individuals to produce binary individuals. Then binary individuals are repaired according to problem specification. Here, Q-gate is the main variation operator applied on Q-bits. For testing our algorithm, three types of benchmark sets are used. Experimental results show that the algorithm has performed very well and it has also outperformed current best results.

Research paper thumbnail of Detecting Animal Contacts—A Deep Learning-Based Pig Detection and Tracking Approach for the Quantification of Social Contacts

Sensors, 2021

The identification of social interactions is of fundamental importance for animal behavioral stud... more The identification of social interactions is of fundamental importance for animal behavioral studies, addressing numerous problems like investigating the influence of social hierarchical structures or the drivers of agonistic behavioral disorders. However, the majority of previous studies often rely on manual determination of the number and types of social encounters by direct observation which requires a large amount of personnel and economical efforts. To overcome this limitation and increase research efficiency and, thus, contribute to animal welfare in the long term, we propose in this study a framework for the automated identification of social contacts. In this framework, we apply a convolutional neural network (CNN) to detect the location and orientation of pigs within a video and track their movement trajectories over a period of time using a Kalman filter (KF) algorithm. Based on the tracking information, we automatically identify social contacts in the form of head–head an...

Research paper thumbnail of Image-based approach for the detection of counterfeit banknotes of Bangladesh

2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), 2016

Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is ... more Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is now a common phenomenon due to advanced printing and scanning technology. Bangladesh has been facing serious problem by the increasing rate of fake notes in the market. To get rid of this problem various fake note detection methods are available around the world and most of these are hardware based and costly. In the present paper an automated image-based technique is described for the detection of fake banknotes of Bangladesh. Security features of banknotes such as watermark, micro-printing and hologram etc. are extracted from the banknote images and then detection is performed using Support Vector Machine (SVM). Experimental results confirm the effectiveness of the proposed algorithm.

Research paper thumbnail of Solving Maximum Clique Problem using a novel Quantum-inspired Evolutionary Algorithm

2015 International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), 2015

ABSTRACT

Research paper thumbnail of Detecting Animal Contacts—A Deep Learning-Based Pig Detection and Tracking Approach for the Quantification of Social Contacts

Sensors, 2021

The identification of social interactions is of fundamental importance for animal behavioral stud... more The identification of social interactions is of fundamental importance for animal behavioral studies, addressing numerous problems like investigating the influence of social hierarchical structures or the drivers of agonistic behavioral disorders. However, the majority of previous studies often rely on manual determination of the number and types of social encounters by direct observation which requires a large amount of personnel and economical efforts. To overcome this limitation and increase research efficiency and, thus, contribute to animal welfare in the long term, we propose in this study a framework for the automated identification of social contacts. In this framework, we apply a convolutional neural network (CNN) to detect the location and orientation of pigs within a video and track their movement trajectories over a period of time using a Kalman filter (KF) algorithm. Based on the tracking information, we automatically identify social contacts in the form of head–head an...

Research paper thumbnail of Image-based approach for the detection of counterfeit banknotes of Bangladesh

2016 5th International Conference on Informatics, Electronics and Vision (ICIEV), 2016

Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is ... more Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is now a common phenomenon due to advanced printing and scanning technology. Bangladesh has been facing serious problem by the increasing rate of fake notes in the market. To get rid of this problem various fake note detection methods are available around the world and most of these are hardware based and costly. In the present paper an automated image-based technique is described for the detection of fake banknotes of Bangladesh. Security features of banknotes such as watermark, micro-printing and hologram etc. are extracted from the banknote images and then detection is performed using Support Vector Machine (SVM). Experimental results confirm the effectiveness of the proposed algorithm.

Research paper thumbnail of Solving Maximum Clique Problem using a novel Quantum-inspired Evolutionary Algorithm

2015 International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), 2015

ABSTRACT

Research paper thumbnail of Single Image Face Recognition based on Gabor, Sobel and Local Ternary Pattern

International Journal of Computer Applications, 2015

Research paper thumbnail of Optimal coverage of Wireless Sensor Network using Termite Colony Optimization Algorithm

2015 International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), 2015

Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) ha... more Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) has been become one of the most attractive and intriguing research area in last few years. Large-scale sensor networks has been deployed by interconnecting several hundred to a few thousand sensors nodes opens up diverse technical challenges and tremendous application possibilities. This paper presents a Termite Colony Optimization (TCO) algorithm, an adaptable and balanced way, which optimizes a minimal number of sensors while covering a maximum area. TCO is a population based metaheuristic approach developed from the intelligent behavior of termite that is known to us as Swarm Intelligence. It has the ability to solve the presented problem in efficient way that has shown more effectiveness and robustness compared to other population based algorithms. In this paper, we have justified our simulation results with the results of other papers. After that we have experimented with varying different parameters like grid size, number of sensors and termites. It has also proved the capability of our TCO over the constraints and difficulties as well.

Research paper thumbnail of Quantum-Inspired Evolutionary Algorithm to Solve Graph Coloring Problem

Research paper thumbnail of Identification of regulatory SNPs associated with genetic modifications in lung adenocarcinoma

Research paper thumbnail of Document’s title

Research paper thumbnail of An Effective Quantum-inspired Evolutionary Algorithm for Finding Degree-constrained Minimum Spanning Tree

Research paper thumbnail of Bangladeshi dialect recognition using Mel Frequency Cepstral Coefficient, Delta, Delta-delta and Gaussian Mixture Model

2016 Eighth International Conference on Advanced Computational Intelligence (ICACI), 2016

Automatic recognition systems are generally applied successfully in speech processing to categori... more Automatic recognition systems are generally applied successfully in speech processing to categorize observed utterances by the speaker identity, dialect and linguistic communication. A lot of research has been performed to detect speeches, dialects and languages of different region throughout the world. But the work on dialects of Bangladesh is infrequent to our research. These dialects, in turn, differ quite a bit from each other. In this paper, we present a method to detect Bangladeshi different dialects which utilizes Mel Frequency Cepstral Coefficient (MFCC), its Delta and Delta-delta as main features and Gaussian Mixture Models (GMM) to classify characteristics of a specific dialect. Particularly we extract the MFCCs, Deltas and Delta-deltas from the speech signal. Then they are merged together to form a feature vector for a specific dialect. GMM is trained using the iterative Expectation Maximization (EM) algorithm where feature vectors are served as input. This scheme is tested on 5 databases of 30 speech samples each. Speech samples contain dialects of Borishal, Noakhali, Sylhet, Chittagong and Chapai Nawabganj regions of Bangladesh. Experiments show that GMM adaptation gives comparable good performance.

Research paper thumbnail of Identification of Regulatory SNPs Associated with Vicine and Convicine Content of Vicia faba Based on Genotyping by Sequencing Data Using Deep Learning

Genes, 2020

Faba bean (Vicia faba) is a grain legume, which is globally grown for both human consumption as w... more Faba bean (Vicia faba) is a grain legume, which is globally grown for both human consumption as well as feed for livestock. Despite its agro-ecological importance the usage of Vicia faba is severely hampered by its anti-nutritive seed-compounds vicine and convicine (V+C). The genes responsible for a low V+C content have not yet been identified. In this study, we aim to computationally identify regulatory SNPs (rSNPs), i.e., SNPs in promoter regions of genes that are deemed to govern the V+C content of Vicia faba. For this purpose we first trained a deep learning model with the gene annotations of seven related species of the Leguminosae family. Applying our model, we predicted putative promoters in a partial genome of Vicia faba that we assembled from genotyping-by-sequencing (GBS) data. Exploiting the synteny between Medicago truncatula and Vicia faba, we identified two rSNPs which are statistically significantly associated with V+C content. In particular, the allele substitutions regarding these rSNPs result in dramatic changes of the binding sites of the transcription factors (TFs) MYB4, MYB61, and SQUA. The knowledge about TFs and their rSNPs may enhance our understanding of the regulatory programs controlling V+C content of Vicia faba and could provide new hypotheses for future breeding programs.

Research paper thumbnail of Image-Based Approach for the Detection of Counterfeit Banknotes of Bangladesh

Image-Based Approach for the Detection of Counterfeit Banknotes of Bangladesh

Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is ... more Currency duplication also known as counterfeit currency is a vulnerable threat on economy. It is now a common phenomenon due to advanced printing and scanning technology. Bangladesh has been facing serious problem by the increasing rate of fake notes in the market. To get rid of this problem various fake note detection methods are available around the world and most of these are hardware based and costly. In the present paper an automated image-based technique is described for the detection of fake banknotes of Bangladesh. Security features of banknotes such as watermark, micro-printing and hologram etc. are extracted from the banknote images and then detection is performed using Support Vector Machine (SVM). Experimental results confirm the effectiveness of the proposed algorithm.

Research paper thumbnail of Bangladeshi Dialect Recognition using Mel Frequency Cepstral Coefficient, Delta, Delta-delta and Gaussian Mixture Model

Automatic recognition systems are generally applied successfully in speech processing to categori... more Automatic recognition systems are generally applied successfully in speech processing to categorize observed utterances by the speaker identity, dialect and linguistic communication. A lot of research has been performed to detect speeches, dialects and languages of different region throughout the world. But the work on dialects of Bangladesh is infrequent to our research. These dialects, in turn, differ quite a bit from each other. In this paper, we present a method to detect Bangladeshi different dialects which utilizes Mel Frequency Cepstral Coefficient (MFCC), its Delta and Delta-delta as main features and Gaussian Mixture Models (GMM) to classify characteristics of a specific dialect. Particularly we extract the MFCCs, Deltas and Delta-deltas from the speech signal. Then they are merged together to form a feature vector for a specific dialect. GMM is trained using the iterative Expectation Maximization (EM) algorithm where feature vectors are served as input. This scheme is tested on 5 databases of 30 speech samples each. Speech samples contain dialects of Borishal, Noakhali, Sylhet, Chittagong and Chapai Nawabganj regions of Bangladesh. Experiments show that GMM adaptation gives comparable good performance.

Research paper thumbnail of Single Image Face Recognition based on Gabor, Sobel and Local Ternary Pattern

Face recognition has been a fast growing, challenging and fascinating area in real time applicati... more Face recognition has been a fast growing, challenging and fascinating area in real time applications. A large number of face recognition algorithms have been developed in last decades. This paper presents a face recognition approach that utilizes the Sobel operator, Local Ternary Pattern (LTP) descriptor, Gabor features and Principal Component Analysis (PCA) algorithms to attain enhanced recognition accuracy where training set has only one image per person. In particular, the edge information of face image is enhanced using Sobel operator. Then we use LTP on the image to encode the micro-level information of spots, edges and other local characteristics. Finally Gabor-wavelets based features are then extracted and their histograms are concatenated together into a contiguous histogram to be used as a descriptor of the face. As Gabor features cause a very high dimensional histogram vector, therefore PCA is used to reduce the dimension. This approach is compared with the original LBP, LTP and Sobel LTP on gray-level images for face recognition. The experimental results exhibit that our method provides a remarkable performance under various conditions.

Research paper thumbnail of Optimal Coverage of Wireless Sensor Network using Termite Colony Optimization Algorithm

Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) ha... more Due to great advancement in the field of science and technology, Wireless Sensor Network (WSN) has been become one of the most attractive and intriguing research area in last few years. Large-scale sensor networks has been deployed by interconnecting several hundred to a few thousand sensors nodes opens up diverse technical challenges and tremendous application possibilities. This paper presents a Termite Colony Optimization (TCO) algorithm, an adaptable and balanced way, which optimizes a minimal number of sensors while covering a maximum area. TCO is a population based metaheuristic approach developed from the intelligent behavior of termite that is known to us as Swarm Intelligence. It has the ability to solve the presented problem in efficient way that has shown more effectiveness and robustness compared to other population based algorithms. In this paper, we have justified our simulation results with the results of other papers. After that we have experimented with varying different parameters like grid size, number of sensors and termites. It has also proved the capability of our TCO over the constraints and difficulties as well.

Research paper thumbnail of Solving Maximum Clique Problem using a Novel Quantum-inspired Evolutionary Algorithm

Maximum Clique Problem (M CP) is one of the most important NP-hard problems in the area of soft c... more Maximum Clique Problem (M CP) is one of the most important NP-hard problems in the area of soft computing and it has many real world applications in numerous fields ranging from coding theory to the determination of the structure of a protein molecule. Different heuristic, Meta heuristic and hybrid solution approaches have been applied to obtain the solution. In this paper, we demonstrate a Quantum-inspired Evolutionary Algorithm (QEA) to solve MCP. We have used one dimensional arrays ofQ­ bits called Q-bit individuals to produce binary individuals. After production of binary individuals, we have repaired and improved them. Here, Q-gate is the main variation operator applied on Q­ bit individuals. Our algorithm was tested on DIMACS benchmark graphs and 40 of them were tested. The results obtained here are extremely encouraging. For almost all of the datasets, we get the optimal results reported on DIMACS benchmark and also compared our results with other related works. For some cases we get better results than other works.

Research paper thumbnail of An Effective Quantum-inspired Evolutionary Algorithm for Finding Degree-constrained Minimum Spanning Tree

The Degree-constrained Minimum Spanning Tree (d-MST) is an extended adaptation of general Minimum... more The Degree-constrained Minimum Spanning Tree (d-MST) is an extended adaptation of general Minimum Spanning Tree (MST) problem. This problem is pertinent in the design of communication networks. It consists of finding a spanning tree whose nodes do not exceed a given maximum degree and whose total edge length is minimum. In this formulation the problem turns into NP-hard, therefore meta heuristic approaches like Ant Colony Optimization, Simulated Annealing, Evolutionary Algorithms etc. are suitable for solving d-MST problem. In this paper, we demonstrate a Quantum-inspired Evolutionary Algorithm (QEA) that solves instances of the problem to optimality. We have used one dimensional array of Q-bits called Q-bit individuals to produce binary individuals. Then binary individuals are repaired according to problem specification. Here, Q-gate is the main variation operator applied on Q-bits. For testing our algorithm, three types of benchmark sets are used. Experimental results show that the algorithm has performed very well and it has also outperformed current best results.

Research paper thumbnail of Privacy Risks and Solutions in Robotics

The use of robotic devices is increasing in our day to day life and consequently giving rise to t... more The use of robotic devices is increasing in our day to day life and consequently giving rise to the privacy and security issues which are very unique compared to other domains. In spite of the swift development of robotics technology, privacy problems for robots remain unchanged and without functional research. It is not that difficult to find out why robots are raising privacy distress. Robots have the ability to move and sense the world surrounding them. This increase in power to observe may cause privacy issues in the upcoming years. In this paper, we present some statistics related to robotics industries. Then different aspects of robot privacy are examined from the point of views of other works and some prevention techniques are also discussed to alleviate this concerning issue. However, this is a problem that might be hard to distinguish and avoid. In some cases, the problem is inevitable. As consumers and creators, our best way should be to move forward very cautiously.

Research paper thumbnail of Development of an Automatic Pig Detection and Tracking System Using Machine Learning

Georg-August-Universität Göttingen, 2020

Due to the fact that the internal state of the animals is being conveyed through their behaviour,... more Due to the fact that the internal state of the animals is being conveyed through their behaviour, it is possible to identify the early signs of any issues such as biting or fighting by monitoring the changes in the known behaviour. However, it is impractical in a commercial establishment to perform continuous inspection of farm animals by the staff to identify behavioural alterations pertinent to over-early intervention. To address this problem, a system has been designed for the pig detection, as well as, tracking their movement in a pen environment based on video recordings. Furthermore, the animal specific interactions have been estimated from the 2D trajectories produced by the proposed multi-object tracker. The work has been divided into four phases which are annotation, detection, tracking and interaction. Among them the detection phase is crucial for this system, as rest of the phases mainly depend on it. In this work, a recall of 91% and precision of 94% have been achieved during the detection process. The tracking of the pigs has been performed using a Kalman filter based multi-object tracker. The Kalman filter is used to determine the states of a process recursively using a collection of equations in such a manner that mean squared error is minimized. Due to its ability to estimate past, present and future states to some extent, although the characteristics of the system is unfamiliar, the Kalman filter is frequently applied for the object tracking. This work has ended with the fourth phase which is detecting the interaction among the pigs in the pen environment. Currently, head-to-head and head to tail interactions have been calculated from the trajectories and put into a table. This table can be used to do behaviour analysis of the pigs which opens up various possibilities for further research work in terms of improving their health and welfare. As well as, it shows a greater potential to change the livestock monitoring in a commercial farm.