Salvador Hinojosa

Cross entropy based thresholding for magnetic resonance brain images using Crow Search Algorithm

We use an evolutionary mechanism to improve the image segmentation process.We optimize the minimum cross entropy with an evolutionary method for image segmentation.We test the approach in multidimensional spaces.An alternative method for MR brain image segmentation is proposed.Comparisons and non-parametric test support the experimental results. Segmentation is considered the central part of an image processing system due to its high influence on the posterior image analysis. In recent years, the segmentation of magnetic resonance (MR) images has attracted the attention of the scientific community with the objective of assisting the diagnosis in different brain diseases. From several techniques, thresholding represents one of the most popular methods for image segmentation. Currently, an extensive amount of contributions has been proposed in the literature, where thresholding values are obtained by optimizing relevant criteria such as the cross entropy. However, most of such approaches are computationally expensive, since they conduct an exhaustive search strategy for obtaining the optimal thresholding values. This paper presents a general method for image segmentation. To estimate the thresholding values, the proposed approach uses the recently published evolutionary method called the Crow Search Algorithm (CSA) which is based on the behavior in flocks of crows. Different to other optimization techniques used for segmentation proposes, CSA presents a better performance, avoiding critical flaws such as the premature convergence to sub-optimal solutions and the limited exploration-exploitation balance in the search strategy. Although the proposed method can be used as a generic segmentation algorithm, its characteristics allow obtaining excellent results in the automatic segmentation of complex MR images. Under such circumstances, our approach has been evaluated using two sets of benchmark images; the first set is composed of general images commonly used in the image processing literature, while the second set corresponds to MR brain images. Experimental results, statistically validated, demonstrate that the proposed technique obtains better results in terms of quality and consistency.

A Comparison of Evolutionary Computation Techniques for IIR Model Identification

System identification is a complex optimization problem which has recently attracted the attention in the field of science and engineering. In particular, the use of infinite impulse response (IIR) models for identification is preferred over their equivalent FIR (finite impulse response) models since the former yield more accurate models of physical plants for real world applications. However, IIR structures tend to produce multimodal error surfaces whose cost functions are significantly difficult to minimize. Evolutionary computation techniques (ECT) are used to estimate the solution to complex optimization problems. They are often designed to meet the requirements of particular problems because no single optimization algorithm can solve all problems competitively. Therefore, when new algorithms are proposed, their relative efficacies must be appropriately evaluated. Several comparisons among ECT have been reported in the literature. Nevertheless, they suffer from one limitation: their conclusions are based on the performance of popular evolutionary approaches over a set of synthetic functions with exact solutions and well-known behaviors, without considering the application context or including recent developments. This study presents the comparison of various evolutionary computation optimization techniques applied to IIR model identification. Results over several models are presented and statistically validated.

Image segmentation by minimum cross entropy using evolutionary methods

The segmentation of digital images is one of the most important steps in an image processing or computer vision system. It helps to classify the pixels in different regions according to their intensity level. Several segmentation techniques have been proposed, and some of them use complex operators. The techniques based on thresholding are the easiest to implement; the problem is to select correctly the best threshold that divides the pixels. An interesting method to choose the best thresholds is the minimum cross entropy (MCET), which provides excellent results for bi-level thresholding. Nevertheless, the extension of the segmentation problem into multiple thresholds increases significantly the computational effort required to find optimal threshold values. Each new threshold adds complexity to the formulation of the problem. Classic methods for image thresholding perform extensive searches, while new approaches take advantage of heuristics to reduce the search. Evolutionary algorithms use heuristics to optimize criteria over a finite number of iterations. The correct selection of an evolutionary algorithm to minimize the MCET directly impacts the performance of the method. Current approaches take a large number of iterations to converge and a high rate of MCET function evaluations. The electromagnetism-like optimization (EMO) algorithm is an evolutionary technique which emulates the attraction–repulsion mechanism among charges for evolving the individuals of a population. Such technique requires only a small number of evaluations to find the optimum. This paper proposes the use of EMO to search for optimal threshold values by minimizing the cross entropy function while reducing the amount of iterations and function evaluations. The approach is tested on a set of benchmark images to demonstrate that is able to improve the convergence and velocity; additionally, it is compared with similar state-of-the-art optimization approaches.

Improving multi-criterion optimization with chaos: a novel Multi-Objective Chaotic Crow Search Algorithm

This paper presents two multi-criteria optimization techniques: the Multi-Objective Crow Search Algorithm (MOCSA) and an improved chaotic version called Multi-Objective Chaotic Crow Search Algorithm (MOCCSA). Both methods MOCSA and MOCCSA are based on an enhanced version of the recently published Crow Search Algorithm. Crows are intelligent animals with interesting strategies for protecting their food hatches. This compelling behavior is extended into a Multi-Objective approach. MOCCSA uses chaotic-based criteria on the optimization process to improve the diversity of solutions. To determinate if the performance of the algorithm is significantly enhanced, the incorporation of a chaotic operator is further validated by a statistical comparison between the proposed MOCCSA and its chaotic-free counterpart (MOCSA) indicating that the results of the two algorithms are significantly different from each other. The performance of MOCCSA is evaluated by a set of standard benchmark functions, and the results are contrasted with two well-known algorithms: Multi-Objective Dragonfly Algorithm and Multi-Objective Particle Swarm Optimization. Both quantitative and qualitative results show competitive results for the proposed approach.

Engineering applications of metaheuristics: an introduction

Metaheuristic algorithms are important tools that in recent years have been used extensively in several fields. In engineering, there is a big amount of problems that can be solved from an optimization point of view. This paper is an introduction of how metaheuristics can be used to solve complex problems of engineering. Their use produces accurate results in problems that are computationally expensive. Experimental results support the performance obtained by the selected algorithms in such specific problems as digital filter design, image processing and solar cells design.

Context based image segmentation using antlion optimization and sine cosine algorithm

Multilevel thresholding (MTH) is one of the most commonly used approaches to perform segmentation on images. However, as most methods are based on the histogram of the image to be segmented, MTH methods only consider the occurrence frequency of certain intensity level disregarding all spatial information. Contextual information can help to enhance the quality of the segmented image as it considers not only the value of the pixel but also its vicinity. The energy curve was designed to bring spatial information into a curve with the same properties as the histogram. In this paper, two recently proposed Evolutionary Computational Algorithms (ECAs) are coupled with two classical thresholding criteria to perform MTH over the energy curve. The selected ECAs are the Antlion Optimizer (ALO) and the Sine Cosine Algorithm (SCA). The proposed methods are evaluated intensively regarding quality, and a statistical analysis is presented to compare the results of the algorithms against similar approaches. Experimental evidence encourages the use ALO for MTH while it concludes that SCA does not outperform other ECAs form the state-of-the-art.

Unassisted thresholding based on multi-objective evolutionary algorithms

Abstract Multi-Objective Evolutionary Algorithms (MOEAs) are known to solve problems where two or more conflicting goals are involved. To accomplish it, MOEAs incorporate strategies to determinate optimal trade-offs between each of the objective functions. In this paper, an Unassisted image Thresholding (UTH) methodology is proposed based on MOEAs. UTH takes advantage of the trade-off mechanisms present on MOEAs to perform the image thresholding while simultaneously determinating the number thresholds required to segment each image and the best placement of each threshold along the histogram of the image. The image thresholding problem is commonly addressed as the search for the best possible thresholds able to partition a given image into a finite number of homogeneous classes. Such approach requires the assistance of a designer to determinate the number of threshold values that will properly segment the image. However, as images can vary significantly, the definition of an optimal number of thresholds should be performed for each image. Thus, a methodology able to determinate both the number of thresholds and the best placement of each value contributes to a general segmentation scheme. In the proposed approach, UTH redefines the thresholding problem as a multi-objective task with two conflicting goals. The first goal is the quality of the segmented image, and it is computed as a non-parametric criteria to evaluate candidate threshold points. The second goal is the normalized number of threshold points. Since the number of thresholds is not fixed, a particle encoding the thresholds with variable length is used. The strategy of UTH is coupled with three MOEAs namely NSGA-III, PESA-II and MOPSO using as the non-parametric criteria the Cross Entropy. According to the results, the UTH NSGA-III formulation outperforms UTH-PESA-II and UTH-MOPSO regarding convergence and quality of the resulting image.

Thermal Image Segmentation Using Evolutionary Computation Techniques

This chapter analyzes the performance of selected evolutionary computation techniques (ECT) applied to the segmentation of forward looking infrared (FLIR) images. FLIR images arise challenges for classical image processing techniques since the capture devices usually generate low-resolution images prone to noise and blurry outlines. Traditional ECTs such as artificial bee colony (ABC), differential evolution (DE), harmony search (HS), and the recently published flower pollination algorithm (FPA) are implemented and evaluated using as objective function the between-class variance (Otsu’s method) and the Kapur’s entropy. The comparison pays particular attention to the quality of the segmented image by evaluating three specific metrics named peak-to-signal noise ratio (PSNR), structural similarity index (SSIM), and feature similarity index (FSIM).

Entropy-based imagery segmentation for breast histology using the Stochastic Fractal Search

Abstract Breast cancer is one of the leading causes of death for women around the world. Its early diagnosis can significantly enhance the survival rate of the patient. Image processing techniques are used to help to diagnose this disease. The analysis of breast tissue samples on histological images is a challenging task where computer vision techniques can contribute. In this paper, the Stochastic Fractal Search (SFS) algorithm is applied to the image thresholding problem on breast histology imagery using as objective function three different entropies. The SFS is a new Evolutionary Algorithm (EA) which emulates the growth mechanism of fractals. SFS has been successfully applied to other applications, but its performance on image thresholding is unknown. The implementation of SFS is conducted using as objective function three of the most representative entropies being Kapur, Minimum Cross Entropy, and Tsallis. To provide a comparison point, two EAs commonly used for image thresholding are selected; the Artificial Bee Colony (ABC) and the Differential Evolution (DE). In this context, the resulting nine combinations are evaluated concerning the quality of the segmented images. Since the nine evaluated methods share either EA or entropy, the nonparametric test of Kruskal–Wallis is conducted to analyze the similarity of the results among methods. Results indicate that the combination of SFS and Minimum Cross Entropy yields the best results for the segmentation of histological imagery.

Electromagnetism-like mechanism with collective animal behavior for multimodal optimization

Evolutionary Computation Algorithms (ECA) are conceived as alternative methods for solving complex optimization problems through the search for the global optimum. Therefore, from a practical point of view, the acquisition of multiple promissory solutions is especially useful in engineering, since the global solution may not always be realizable due to several realistic constraints. Although ECAs perform well on the detection of the global solution, they are not suitable for finding multiple optima in a single execution due to their exploration-exploitation operators. This paper proposes a new algorithm called Collective Electromagnetism-like Optimization (CEMO). Under CEMO, a collective animal behavior is implemented as a memory mechanism simulating natural animal dominance over the population to extend the original Electromagnetism-like Optimization algorithm (EMO) operators to efficiently register and maintain all possible Optima in an optimization problem. The performance of the proposed CEMO is compared against several multimodal schemes over a set of benchmark functions considering the evaluation of multimodal performance indexes typically found in the literature. Experimental results are statistically validated to eliminate the random effect in the obtained solutions. The proposed method exhibits higher and more consistent performance against the rest of the tested multimodal techniques.