Novel Image Segmentation Using Particle Swarm Optimization
23 Apr 2018-pp 46-50
TL;DR: Image segmentation which is complex optimization problem can be solved by this simple nature inspired PSO (Particle swarm optimization) model which is formulated in this paper.
Abstract: Data clustering and classification technique algorithms often need to possess enough and prominent number of features in the data. Repeating and dominant features are useful in clustering or segmenting the image. The image segmentation method based on k-mean clustering, hierarchical clustering, and expectation maximization derives the optimum cluster centers based on the number of features such as similar intensity region. Deriving such number optimum number of clusters and its centers is an optimization problem. The aim of this paper is to improve the image segmentation using nature inspired techniques. Image segmentation which is complex optimization problem can be solved by this simple nature inspired PSO (Particle swarm optimization) model which is formulated in this paper. PSO model is generic model which is used to solve number of scientific problems. This paper formulates simple PSO model to solve the image segmentation problem. The proposed algorithm randomly assigns the centers to swarm and best value of objective function is initialized best on the color histogram of an image. This is discussed in section 2 and 3 of paper. Section 4 and 5 discusses and results and concluding remarks on results.
Citations
More filters
TL;DR: The improved segmentation method based on the particle swarm algorithm makes it possible to segment complex structured images acquired from space surveillance systems and reduces segmentation errors of the first kind by an average of 12 % and that of the second kind by 8 %.
Abstract: This paper considers the improved method for segmenting complex structured images acquired from space observation systems based on the particle swarm algorithm. Unlike known ones, the method for segmenting complex structured images based on the particle swarm algorithm involves the following:
– highlighting brightness channels in the Red-Green-Blue color space;
– using a particle swarm method in the image in each channel of brightness of the RGB color space;
– image segmentation is reduced to calculating the objective function, moving speed, and a new location for each swarm particle in the image in each RGB color space brightness channel.
Experimental studies have been conducted on the segmentation of a complex structured image by a method based on the particle swarm algorithm. It was established that the improved segmentation method based on the particle swarm algorithm makes it possible to segment complex structured images acquired from space surveillance systems.
A comparison of the quality of segmenting a complex structured image was carried out. The comparative visual analysis of well-known and improved segmentation methods indicates the following:
– the improved segmentation method based on the particle swarm algorithm highlights more objects of interest (objects of military equipment);
– the well-known k-means method assigns some objects of interest (especially those partially covered with snow) to the snow cover (marked in blue);
– the improved segmentation method also associates some objects of interest that are almost completely covered with snow with the snow cover (marked in blue).
It has been established that the improved segmentation method based on the particle swarm algorithm reduces segmentation errors of the first kind by an average of 12 % and reduces segmentation errors of the second kind by an average of 8 %
12 citations
TL;DR: The improved segmentation method using a genetic algorithm can be implemented in software and hardware imaging systems from space surveillance systems and makes it possible to segment images from space Surveillance systems.
Abstract: The object of this research is the process of segmentation of camouflaged military equipment in images from space surveillance systems.
The method of segmentation of camouflaged military equipment in images from space surveillance systems has been improved using a genetic algorithm. Unlike known methods, the method of segmentation of camouflaged military equipment using a genetic algorithm involves the following:
– highlighting brightness channels in the Red-Green-Blue color space;
– the use of a genetic algorithm in the image in each channel of brightness of the RGB color space;
– image segmentation is reduced to the formation of generations and populations of chromosomes, the calculation of the objective function, selection, crossing, mutation, and decoding of chromosomes in each brightness channel of the Red-Green-Blue color space.
Experimental studies were conducted on the segmentation of camouflaged military equipment using a genetic algorithm. It is established that the improved method of segmentation using a genetic algorithm makes it possible to segment images from space surveillance systems.
A comparison of the quality of segmentation was carried out. It is established that the improved method of segmentation using a genetic algorithm reduces segmentation errors in the following way:
– compared to the known k-means method, by an average of 15 % of errors of the first kind and an average of 7 % of errors of the second kind;
– compared to the method of segmentation based on the algorithm of swarm of particles, by an average of 3.8 % of errors of the first kind and an average of 2.9 % of errors of the second kind.
The improved segmentation method using a genetic algorithm can be implemented in software and hardware imaging systems from space surveillance systems
9 citations
02 Jul 2020
TL;DR: Various segmentation techniques like Otsu Thresholding, K-means clustering, Watershed Technique, Active Contours and Particle Swarm Optimization have been reviewed under this study.
Abstract: Segmentation is the technique where an image is partitioned into various regions comprising of similar attributes like intensity, texture, graysclae, etc represented by pixel patterns. Image segmentation has been generating research opportunities in this field for a long period. Various segmentation techniques like Otsu Thresholding, K-means clustering, Watershed Technique, Active Contours and Particle Swarm Optimization have been reviewed under this study. Survey of the mathematical equations along with the carrying-out of above segmentation techniques practically is the desired purpose of this study. The performance is analysed by estimation of the performance measurements i.e, TP, TN, FP and FN in numerical values.
3 citations
Cites methods from "Novel Image Segmentation Using Part..."
...In medical images, PSO combined with HMRF (Hidden Markov Random Fields) produced superior segmentation results than those by threshold and k-means techniques indicating its rigour and resistance to noise [37]....
[...]
01 Jan 2023
TL;DR: In this article , a web-mobile AI-powered automatic skin lesion recognition system optimized by a new hybrid Whale-Shark optimization algorithm (WOA-SSO-ANN) is presented.
Abstract: Neglected TROPICAL diseases related to the skin with similar manifestations in their early phase persist in remote areas and are characterized by the prevailing poverty of populations. If not detected early, they very often lead to severe ulcerations and permanent disabilities. We present an approach to optimize the early detection of neglected tropical skin diseases (NTDs) by automatic identification of skin lesions. As contributions, we propose a web-mobile AI-powered automatic skin lesion recognition system optimized by a new hybrid Whale-Shark optimization algorithm (WOA-SSO-ANN) that can help frontline health workers without state-of-the-art equipment to detect NTDs in their beginning stage. We extract the relevant regions features of the lesions. The dataset resulting from this preprocessing is classified by artificial neural networks optimized by a new hybrid Whale-Shark optimization algorithm to develop an improved artificial neural network in terms of processing time and/or accuracy. The best result was obtained with an overall classification accuracy of 93% and a processing time reduced by almost half compared to other optimizers. The proposed application is able to recognize cases of Buruli ulcer, leprosy, and leishmaniasis in our database (nodule and plaque) and classify new patients, thus reducing the cost of management of these diseases when they are detected late. The AI models implemented in this work have satisfactory accuracy and could be a complementary diagnostic tool, especially in remote areas where medical specialists are scarce.
References
More filters
TL;DR: The experimental results show that the two PSO-based multi-objective algorithms can automatically evolve a set of nondominated solutions and the first algorithm outperforms the two conventional methods, the single objective method, and the two-stage algorithm.
Abstract: Classification problems often have a large number of features in the data sets, but not all of them are useful for classification. Irrelevant and redundant features may even reduce the performance. Feature selection aims to choose a small number of relevant features to achieve similar or even better classification performance than using all features. It has two main conflicting objectives of maximizing the classification performance and minimizing the number of features. However, most existing feature selection algorithms treat the task as a single objective problem. This paper presents the first study on multi-objective particle swarm optimization (PSO) for feature selection. The task is to generate a Pareto front of nondominated solutions (feature subsets). We investigate two PSO-based multi-objective feature selection algorithms. The first algorithm introduces the idea of nondominated sorting into PSO to address feature selection problems. The second algorithm applies the ideas of crowding, mutation, and dominance to PSO to search for the Pareto front solutions. The two multi-objective algorithms are compared with two conventional feature selection methods, a single objective feature selection method, a two-stage feature selection algorithm, and three well-known evolutionary multi-objective algorithms on 12 benchmark data sets. The experimental results show that the two PSO-based multi-objective algorithms can automatically evolve a set of nondominated solutions. The first algorithm outperforms the two conventional methods, the single objective method, and the two-stage algorithm. It achieves comparable results with the existing three well-known multi-objective algorithms in most cases. The second algorithm achieves better results than the first algorithm and all other methods mentioned previously.
855 citations
TL;DR: This study presents, discusses, and extracts the major trends from an exhaustive overview of various nuclei detection, segmentation, feature computation, and classification techniques used in histopathology imagery, specifically in hematoxylin-eosin and immunohistochemical staining protocols.
Abstract: Digital pathology represents one of the major evolutions in modern medicine. Pathological examinations constitute the gold standard in many medical protocols, and also play a critical and legal role in the diagnosis process. In the conventional cancer diagnosis, pathologists analyze biopsies to make diagnostic and prognostic assessments, mainly based on the cell morphology and architecture distribution. Recently, computerized methods have been rapidly evolving in the area of digital pathology, with growing applications related to nuclei detection, segmentation, and classification. In cancer research, these approaches have played, and will continue to play a key (often bottleneck) role in minimizing human intervention, consolidating pertinent second opinions, and providing traceable clinical information. Pathological studies have been conducted for numerous cancer detection and grading applications, including brain, breast, cervix, lung, and prostate cancer grading. Our study presents, discusses, and extracts the major trends from an exhaustive overview of various nuclei detection, segmentation, feature computation, and classification techniques used in histopathology imagery, specifically in hematoxylin-eosin and immunohistochemical staining protocols. This study also enables us to measure the challenges that remain, in order to reach robust analysis of whole slide images, essential high content imaging with diagnostic biomarkers and prognosis support in digital pathology.
567 citations
Additional excerpts
...Apply velocity update using equation (7)...
[...]
07 Jul 2012
TL;DR: Experimental results show that both proposed algorithms can automatically evolve a smaller number of features and achieve better classification performance than using all features and feature subsets obtained from the two single objective methods and the conventional method.
Abstract: Feature selection (FS) is an important data preprocessing technique, which has two goals of minimising the classification error and minimising the number of features selected. Based on particle swarm optimisation (PSO), this paper proposes two multi-objective algorithms for selecting the Pareto front of non-dominated solutions (feature subsets) for classification. The first algorithm introduces the idea of non-dominated sorting based multi-objective genetic algorithm II into PSO for FS. In the second algorithm, multi-objective PSO uses the ideas of crowding, mutation and dominance to search for the Pareto front solutions. The two algorithms are compared with two single objective FS methods and a conventional FS method on nine datasets. Experimental results show that both proposed algorithms can automatically evolve a smaller number of features and achieve better classification performance than using all features and feature subsets obtained from the two single objective methods and the conventional method. Both the continuous and the binary versions of PSO are investigated in the two proposed algorithms and the results show that continuous version generally achieves better performance than the binary version. The second new algorithm outperforms the first algorithm in both continuous and binary versions.
89 citations
Journal Article•
TL;DR: A new segmentation method for images based on particle swarm optimization (PSO) is proposed through combining PSO algorithm with one of region-based image segmentation methods, which is named Seeded Region Growing (SRG).
Abstract: I n this paper, a new segmentation method for images based on particle swarm optimization (PSO) is proposed. The new method is produced through combining PSO algorithm with one of region-based image segmentation methods, which is named Seeded Region Growing (SRG).The algorithm of SRG method performs a segmentation of an image with respect to a set of points known as seeds. Two problems are related with SRG method, the first one is the choice of the similarity criteria of pixels in regions and the second problem is how to select the seeds. In the proposed method, PSO algorithm tries to solve the two problems of SRG method. The similarity criteria that will be solved is the best similarity difference between the pixel intensity and the region mean value. The proposed algorithm randomly initialise each particle in the swarm to contain K seed points (each seed point contains its location and similarity difference value) and then SRG algorithm is applied to each particle. PSO technique is then applied to refine the locations and similarity difference values of the K seed points. Finally, region merging is applied to remove small regions from the segmented image.
53 citations