Papers by Manoj Kumar Singh
ijcsi.org
Page 1. IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010 I... more Page 1. IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 5, September 2010 ISSN (Online): 1694-0814 www.IJCSI.org 272 Modeling and design of evolutionary neural network for heart disease detection ...
In this paper, we have proposed a secure means for fingerprint biometric authentication, which ha... more In this paper, we have proposed a secure means for fingerprint biometric authentication, which has the capability to deliver the user's privacy, their fingerprint template protection, and robustness against the various variations in terms of noise. In this paper, principle based on correlation strength has been presented to defined fingerprint recognition requirement,to achieve the desired objectives and high quality of solution, computational intelligence basedconcept which uses the feed forward architecture of artificial neural network is applied as solution technology. Proposed methods provides numerous advantages like less memory requirement, very high level of security for stored information without any extra means, high speed and simple implementation approach. Proposed method has robustness against various types of noise available with fingerprint image.
International Journal of Engineering & Technology, 2018
In this paper, a complementary approach has applied to obtain the available edges in the image. T... more In this paper, a complementary approach has applied to obtain the available edges in the image. The complementary image has obtained by subtracting the rough mirror mapped image from the input image. The universal approximation capability of feedforward neural network has applied to define the rough mirror mapping. Multilayer perceptron network and radial basis function network have considered obtaining the mapping. Effect of better learning has also explored in both network by applying adaptivenesss in their transform function available in the active nodes. Single image based training has given for few number of iterations in the development of mapping process. It is observed that proposed method has self adjusted content aware oriented edge detection where as many existing methods like Sobel, Prewitt have shown their limitations in observing the edges associated with contents having similar shade in the surroundings.
International Journal of Electrical and Computer Engineering (IJECE), 2018
The problem of noise interference with the image always occurs irrespective of whatever precautio... more The problem of noise interference with the image always occurs irrespective of whatever precaution is taken. Challenging issues with noise reduction are diversity of characteristics involved with source of noise and in result; it is difficult to develop a universal solution. This paper has proposed neural network based generalize solution of noise reduction by mapping the problem as pattern approximation. Considering the statistical relationship among local region pixels in the noise free image as normal patterns, feedforward neural network is applied to acquire the knowledge available within such patterns. Adaptiveness is applied in the slope of transfer function to improve the learning process. Acquired normal patterns knowledge is utilized to reduce the level of different type of noise available within an image by recorrection of noisy patterns through pattern approximation. The proposed restoration method does not need any estimation of noise model characteristics available in t...
Designing low cost and high speed authentication solution for digital images is always an attract... more Designing low cost and high speed authentication solution for digital images is always an attractive area of research in image processing. In past few years because of widespread use of internet and network technology, concept of information distribution has been become habit rather than exception in daily life. In same aspects challenges involved with distribution of authenticate information has been increased manifolds. In this paper a generalize image authentication method has proposed by hybridization of color histogram and associated first four statistical moments to achieve the objectives of low cost and high speed. Proposed method can apply for both gray and color images having any size and any format. Solution generates a very small authentication code with an ease means which is use to analyze the characteristics of received image from tampering perspective.
Journal of Image and Graphics, 2013
In image processing based applications there are very important requirements of noise removal and... more In image processing based applications there are very important requirements of noise removal and edge detection. In this paper universal approximation characteristic of feedforward neural is taken to achieve both of these requirements in a simple but very efficient way. Concept of local pattern generated by small region pixels which define the possibility of relation among pixels is presented. This approach facilitates the solution as universal solution for different types of noise removal compare to conventional solutions which are based on noise characteristics. Same model of neural network with little extension can also be utilized as edge detector has also presented. Another benefit of proposed model is contrast enhancement without any extra computation cost. In effect this solution can be considered as universal solution for noise reduction, edge detection and contrast enhancement. Comparison has made with well established solution like median filter and adaptive Wiener filter for noise reduction where as Canny and Prewitt detectors have taken for edge detection comparison.
International Journal of …, 2011
A neural network may be considered as an adaptive system that progressively self-organizes in ord... more A neural network may be considered as an adaptive system that progressively self-organizes in order to approximate the solution, making the problem solver free from the need to accurately and unambiguously specify the steps towards the solution. Moreover, Evolutionary computation can be integrated with artificial Neural Network to increase the performance at various levels; in result such neural network is called Evolutionary ANN. In this paper very important issue of neural network namely adjustment of connection weights for learning presented by Genetic algorithm over feed forward architecture. To see the performance of developed solution comparison has given with respect to well established method of learning called gradient decent method. A benchmark problem of classification, XOR, has taken to justify the experiment. Presented method is not only having very probability to achieve the global minima but also having very fast convergence.
International Journal of Image Processing
Abstract: The present demands of scientific and social life forced image processing based applica... more Abstract: The present demands of scientific and social life forced image processing based applications tohave a tremendous growth. This growth at the same time has given number of challenges toresearcher to meet the desired objectives of either users or from solution ...
ijser.org
AbstractDesigning of a network which could fulfill most of the requirements is always a challeng... more AbstractDesigning of a network which could fulfill most of the requirements is always a challenging task for a researcher. Often this happens either with manual approach or by applying some kind of conventional methods. In both cases results do not have high level ...
Signal Processing: An …, 2012
The Least Mean square (LMS) algorithm has been extensively used in many applications due to its s... more The Least Mean square (LMS) algorithm has been extensively used in many applications due to its simplicity and robustness. In practical application of the LMS algorithm, a key parameter is the step size. As the step size becomes large/small, the ...
Arxiv preprint arXiv:1002.4004, 2010
Abstract Because of the stochastic nature of traffic requirement matrix, it's very difficul... more Abstract Because of the stochastic nature of traffic requirement matrix, it's very difficult to get the optimal traffic distribution to minimize the delay even with adaptive routing protocol in a fixed connection network where capacity already defined for each link. Hence there is a ...
Arxiv preprint arXiv:0910.1838, 2009
Among the various means of available resource protection including biometrics, password based sys... more Among the various means of available resource protection including biometrics, password based system is most simple, user friendly, cost effective and commonly used. But this method having high sensitivity with attacks. Most of the advanced methods for authentication ...
In the design of frequency-selective filters, the desired filter characteristics are specified in... more In the design of frequency-selective filters, the desired filter characteristics are specified in the frequency domain in terms of desired magnitude and phase response of the filter. In this paper we present a design approach by determining the closely approximated coefficients using powerful Evolutionary Programming to find the solution for the optimization problem in selecting the coefficients. In this paper the design of Causal FIR filter with desired frequency response and phase response is presented. In practice, FIR filters are employed in filtering problems where there is a requirement for linear phase characteristics within the passband of the filter. The Evolutionary Programming is the best search procedure and most powerful than Linear Programming in providing the optimal solution that is desired to minimize the ripple content in both passband and stopband. We presented here how the values of δ 1 and δ 2 are minimized with best optimized approach using Evolutionary Computation. The optimized filter bank structure is implemented in our research work for effective compression of images.
International Journal of Engineering & Technology
In this paper, parameterization of the single diode model for solar cell has presented. The probl... more In this paper, parameterization of the single diode model for solar cell has presented. The problem of obtaining the optimal parameter has transformed as an optimization problem where individual absolute error has minimized by hybrid mutation strategy in the Evolutionary programming. Hybridization has given between Gaussian mutation strategy and Cauchy mutation strategy to obtain the better offspring. To increase the reliability of the solution, two stages based a multiculture architecture has proposed. On the first stage, a multi-population strategy has applied to form a multiculture environment, where each population evolved independently to explore the solution domain.This stage will prevent the solution to trap in the local minima. In the second stage, evolved population from first stage combine and members having high fitness are selected to form a new population of the same size as the individual population in the first stage. This second stage population evolved fur...
Spinger, 2020
In this paper, a stochastic gradient method based
adaptive version of the radial basis function n... more In this paper, a stochastic gradient method based
adaptive version of the radial basis function neural network has proposed to map the pattern features of the control chart patterns in different categories to recognize their belonging class. Adaptiveness has given over the spreadness and centers of Gaussian basis function appeared in the hidden nodes of the radial basis function neural network.Along with normal abnormalities in patterns, the mixture of
different abnormal patterns has also considered capturing the worst possible conditions of abnormalities in real time.The advantages of the proposed method have appeared as very high recognition accuracy, minimum error in learning and generalize performance with small training dataset in control chart pattern recognition. Achieved performance has compared with the state of art results available in the
literature which has applied feature based recognition using
Support vector machine and Genetic algorithm. The proposed method has enhanced the recognition generalization of control chart patterns with simplicity in design and high level of decision confidence. The performances have achieved through the simulation-based experiments over a huge number of patterns containing ten different types of pattern and on average, 99.99% accuracy has achieved.
In this paper, a complementary approach has applied to obtain the available edges in the image. T... more In this paper, a complementary approach has applied to obtain the available edges in the image. The complementary image has obtained by subtracting the rough mirror mapped image from the input image. The universal approximation capability of feedforward neural network has applied to define the rough mirror mapping. Multilayer perceptron network and radial basis function network have considered obtaining the mapping. Effect of better learning has also explored in both network by applying adaptivenesss in their transform function available in the active nodes. Single image based training has given for few number of iterations in the development of mapping process. It is observed that proposed method has self adjusted content aware oriented edge detection where as many existing methods like Sobel, Prewitt have shown their limitations in observing the edges associated with contents having similar shade in the surroundings.
The application of different engineering fields in the discovery and development of new materials... more The application of different engineering fields in the discovery and development of new materials, especially of new catalyst, is changing the conventional research methodology in materials science.For Heterogeneous catalysts, their catalytic activity and selectivity are dependant on chemical composition, micro structure and reaction conditions. Therefore, it is worth to do the research over the composition of the catalyst and the reaction conditions that will boost its performance.This paper proposes a computational intelligence approach based on adaptive social behavior optimization (ASBO) for catalyst composition optimization to enhance the resulting yield or achieving objective maximal.The proposed approach is especially useful in the combinatorial catalysis optimization wherein the fitness function is unknown, in result cost and time can be drastically reduced with intelligent search method instead of applying real time chemical reaction.Challenge of handling higher dimensionality and achieving a global solution can be fulfilled by ASBO which is based on human behavior under social structure which makes human as a most successful species in nature.Two different mathematical models of the catalyst composition problem, which contains the optimal complexity and represents practical scenarios have taken to explore the quality of solution. Particle swarm optimization (PSO) which is considered as a successful heuristic method among others has also been applied to get the comparative performance analysis in detail.
The interactions and influence taking place in the society could be a source of rich inspiration ... more The interactions and influence taking place in the society could be a source of rich inspiration for the development of novel computational methods. In this paper a new optimization method called "Adaptive social behavior optimization (ASBO)" derived from abstract inherent characteristics of competition, influence and self-confidence which are involved behind making a successful social life especially with human society is presented. The characteristics of dynamic leadership and dynamic logical neighbors along with experienced self capability are taken as fundamental social factors to define the growth of individual and in result of whole society. For each entity of a society, characteristics and affect of these three factors are not being constant for whole life span, rather than function of time and present status. To define this dynamic characteristic under a social life, in ASBO, help of self-adaptive mutation strategy is opted. To establish the applicability of proposed method various benchmark optimization problems are taken to obtain the global solutions. Performance comparison between ASBO and various variation of PSO, which is another well established optimization method based on swarm social behavior, is also presented. Proposed method is simple, more generalized and free from parameters setting in working and very efficient from performance perspectives to achieve the global solution.
Self-exploration capability is an important and necessary factor in all social communities where ... more Self-exploration capability is an important and necessary factor in all social communities where individual assumes to have their own intelligence. Macro social influencing factors are responsible for decision nature taken by an individual, whereas self-exploration process can be considered as a refinement of that decision by use of the cognitive capability to explore a number of surrounding possibilities. The mathematical model corresponding to the individual self-exploration process can be expressed with the help of the chaotic search method. In this paper, chaotic search-based self-exploration has integrated with social influenced-based particle swarm optimisation (PSO) to represent better computational model so that the complex optimisation problem could solve more efficiently. Two different levels of self-exploration called intrinsic cascade self-exploration and extrinsic cascade self-exploration have applied in association with PSO. This paper has applied the proposed concept to cluster documents data in the area of information retrieval and to achieve the global solutions for high dimensional numerical optimisation problems. He has R&D background in advanced intelligent computing and solution development in various fields of engineering and technology. His field of technological research includes nano and quantum computing, soft computing, advanced machine learning, etc.
Uploads
Papers by Manoj Kumar Singh
adaptive version of the radial basis function neural network has proposed to map the pattern features of the control chart patterns in different categories to recognize their belonging class. Adaptiveness has given over the spreadness and centers of Gaussian basis function appeared in the hidden nodes of the radial basis function neural network.Along with normal abnormalities in patterns, the mixture of
different abnormal patterns has also considered capturing the worst possible conditions of abnormalities in real time.The advantages of the proposed method have appeared as very high recognition accuracy, minimum error in learning and generalize performance with small training dataset in control chart pattern recognition. Achieved performance has compared with the state of art results available in the
literature which has applied feature based recognition using
Support vector machine and Genetic algorithm. The proposed method has enhanced the recognition generalization of control chart patterns with simplicity in design and high level of decision confidence. The performances have achieved through the simulation-based experiments over a huge number of patterns containing ten different types of pattern and on average, 99.99% accuracy has achieved.
adaptive version of the radial basis function neural network has proposed to map the pattern features of the control chart patterns in different categories to recognize their belonging class. Adaptiveness has given over the spreadness and centers of Gaussian basis function appeared in the hidden nodes of the radial basis function neural network.Along with normal abnormalities in patterns, the mixture of
different abnormal patterns has also considered capturing the worst possible conditions of abnormalities in real time.The advantages of the proposed method have appeared as very high recognition accuracy, minimum error in learning and generalize performance with small training dataset in control chart pattern recognition. Achieved performance has compared with the state of art results available in the
literature which has applied feature based recognition using
Support vector machine and Genetic algorithm. The proposed method has enhanced the recognition generalization of control chart patterns with simplicity in design and high level of decision confidence. The performances have achieved through the simulation-based experiments over a huge number of patterns containing ten different types of pattern and on average, 99.99% accuracy has achieved.