Krishna Gopal Dhal

Development of firefly algorithm via chaotic sequence and population diversity to enhance the image contrast

Nature-inspired algorithms have been applied in the optimization field including digital image processing like image enhancement or segmentation. Firefly algorithm (FA) is one of the most powerful of them. In this paper two different implementation of FA has been taken into consideration. One of them is FA via lévy flight where step length of lévy flight has been taken from chaotic sequence. Chaotic sequence shows ergodicity property which helps in better searching. But in the second implementation chaotic sequence replaces lévy flight to enhance the capability of FA. Population of individuals has been created in every generation using the information of population diversity. As an affect FA does not converges prematurely. These two modified FA algorithms have been applied to optimize parameters of parameterized contrast stretching function. Entropy, contrast and energy of the image have been used as objective criterion for measuring goodness of image enhancement. Fitness criterion has been maximized in order to get enhanced image with better contrast. From the experimental results it has been shown that FA with chaotic sequence and population diversity information outperforms the Particle swarm optimization and FA via lévy flight.

Performance Enhancement of Differential Evolution by Incorporating Lévy Flight and Chaotic Sequence for the Cases of Satellite Images

Differential Evolution DE is a simple but powerful evolutionary algorithm. Crossover Rate CR and Mutation Factor F are the most important control parameters in DE. Mutation factor controls the diversification. In traditional DE algorithm CR and F are kept constant. In this paper, the values of CR and F are modified to enhance the capability of traditional DE algorithm. In the first modified algorithm chaotic sequence is used to perform this modification. In the next modified algorithm Levy Flight with chaotic step size is used for such enhancement. In the second modified DE, population diversity has been used to build population in every generation. As a result the algorithm does not converge prematurely. Both modified algorithms have been applied to optimize parameters of the parameterized contrast stretching function. The algorithms are tested for satellite image contrast enhancement and the results are compared, which show that DE via chaotic Levy and population diversity information outperforms the traditional and chaotic DE in the image enhancement domain.

A novel hybrid approach to enhance low resolution images using particle swarm optimization

Enhancement of low resolution images is always a priority Enhancement of low resolution images is always a priority field of digital image processing. In this paper, we propose a novel hybrid approach based on discrete wavelet transform (DWT) and particle swarm optimization (PSO). To develop the proposed method we use spatial domain as well as frequency domain. To reduce the low frequencies from the input image we use the frequency domain. DWT is used to decompose the input low resolution image into different sub bands. Each of the interpolated high frequency sub band (LH, HL, HH) is then summed up with the interpolated output image of the frequency domain. In order to achieve high resolution image, the estimated high frequency sub bands of the intermediate stage and the interpolated low resolution input image have been combined by using inverse DWT. To generate a better high resolution image particle swarm optimization (PSO) technique has been used. The quantitative (root mean square error, normalized cross correlation, normalized absolute error) and visual outcome show the strength of this proposed method.

A novel hybrid approach to restore historical degraded documents

Old degraded historical documents carry various important information regarding our culture, economics etc. proper restoration of these documents is very necessary. After digitization of these documents there remain noises and other low resolution components. These affect the overall visual appearance of the documents. In this paper a novel approach is proposed to enhance ancient historical documents. To enhance these digital format documents a two way approach is considered. At first Particle Swarm Optimization (PSO) and bilateral filter is applied. At second level Non-Linear Enhancement with bilateral filter is applied. Both the approaches are then tested visually and quantitively to show the effectiveness of the approach.

Performance Analysis of Chaotic Lévy Bat Algorithm and Chaotic Cuckoo Search Algorithm for Gray Level Image Enhancement

Dark images can be enhanced in a controlled manner with the help of nature inspired metaheuristic algorithm. In this case image enhancement has been taken as a nonlinear optimization problem. Bat algorithm (BA) and Cuckoo Search (CS) algorithm is one of the most powerful metaheuristic algorithms. In this paper these two algorithms have been modified by chaotic sequence and levy flight. In BA levy flight with chaotic step size helps to do intensification. In CS algorithm the random walk has been done via chaotic sequence. Entropy and edge information has been used as objective function. From quantitative and visual analysis it is clear that chaotic levy BA outperforms the chaotic CS algorithm.

Classification of Ancient Coin using Artificial Neural Network

Use of the coins has been started in Asia Minor during 7th century B.C. Dates back between 2500 B.C and 1700 B.C. Coins were used to trade in the Indus valley of Mohenjo-Daro and Harappa. Ancient coins are always tough to identify and recognize. Weathering and other natural causes degrades it overall structure. Classification of such ancient coins using computer vision and machine intelligence is a challenging task. Here in this paper this task has been taken to be addressed. This paper aims to develop a intelligent system which can classify and recognize ancient coins through their images only. The approach involves feature extraction classification and recognition. Standard deviation of the histogram of the image has been considered as a feature which is then classified and recognized by feed forward back propagation artificial neural network. Preprocessing of the image includes filtering of the image for better results.

Parameterless Harmony Search for Image Multi-thresholding

The Harmony Search (HS) Algorithm is one of the efficient nature-inspired optimization algorithms which exhibits interesting search capability within less computational overhead. However, empirical studies showed that the main problem of this kind of algorithms is the proper setting of the associated parameters. HS associated with a few parameters and to find out the proper combination of the parameter values is time consuming. That’s why a parameterless variant has been proposed here, which does not need the tuning over control parameters. The effect of different population size and stopping criterion has been considered in the experiment. The efficiency of the proposed HS is measured in Shannon’s entropy based image multi-thresholding field.

A novel human hand finger gesture recognition using machine learning

Human-Computer Interaction (HCI) using intelligent artificial computing interface is a fast emerging and revolutionary field of study of computer vision. This present study is concerned with making computers responsive to human gestures and postures. In this paper a simple alternative method for hand gesture recognition system has been proposed. The system takes various fingers postures and try to recognize them using machine learning. A pattern of gestures is trained and tested to show the results using linear artificial neural network.

Breast Histopathology Image Clustering using Cuckoo Search Algorithm

Breast histopathological image segmentation is exigent due to the existence of imperceptibly correlated and indistinct multiple regions of concern. Clustering based segmentation is one of the most significant approaches to perform proper segmentation of such complex images. K-means is the well-known clustering techniques but very sensitive to initial cluster centers and easy convergences to local optima. Therefore, researchers are employing Nature-Inspired Optimization Algorithms (NIOA) in this domain. This study develops Cuckoo Search (CS) algorithm based image clustering model for the proper segmentation of breast histopathology images. Experimental results show that CS provides better-quality segmented images compare to classical Kmeans algorithm by considering the computational time, fitness values and the values of quality parameters.