Multi-Level Thresholding for Image Segmentation With Swarm Optimization Algorithms

Abstract

-Image segmentation is an important problem for image processing. The image processing applications are generally affectedfromthe segmentation success. There is noany image segmentation method which gives good results for all sorts of images. That's why there are many approaches and methods forimage segmentationin the literature. And one of the most used is the thresholding technique. Thresholding techniques can be categorized into two topics: bi-level and multi-level thresholding. Bi-level thresholding technique has one threshold value which separates the image into two groups. However, multi-level thresholding technique uses n threshold values where n greater than one. In this paper, two swarm optimization algorithms (Particle Swarm Optimization, PSO and Cat Swarm Optimization, CSO) are applied on finding the optimum threshold values for the multi-level thresholding. In literature, there are some minimization or maximization functions to find the best threshold values for thresholding problem. Some of these methods are: Tsalli's Entropy, Kapur's Entropy, Renyi's Entropy, Otsu's Method (within class variance/between class variance), the Minimum Cross Entropy Thresholding (MCET) etc.In this work, Otsu's (within class variance) method, which is one of these popular functions,is used as the fitness function of algorithms.In the experiments, five real images are segmented by usingParticle Swarm Algorithm and Cat Swarm Optimization Algorithms. The performances of the swarm algorithms on multi-level thresholding problem arecompared with Peak Signal-to-Noise Ratio (PSNR) and fitness function (FS) values. As a result, the PSO yields better performance than CSO.

Cat Swarm Optimization Algorithm: A Survey and Performance Evaluation

“…CSO successfully optimized the switching parameters of CSI and hence minimized the total harmonic distortion [52] Applied both CSO, PCSO, PSO-CFA, and ACO-ABC on distributed generation units on distribution networks IEEE 33-bus and IEEE 69-bus distribution systems were used in the simulation experiments and CSO outperformed the other algorithms [53] Applied MCSO on MPPT to achieve global maximum power point (GMPP) tracking MCSO outperformed PSO, MPSO, DE, GA, and HC algorithms [54] Applied BCSO to optimize the location of phasor measurement units and reduce the required number of PMUs IEEE 14-bus and IEEE 30-bus test systems were used in the simulation. BCSO outperformed BPSO, generalized integer linear programming, and effective data structure-based algorithm [55] Used CSO algorithm to identify the parameters of single and double diode models in solar cell system CSO outperformed PSO, GA, SA, PS, Newton, HS, GGHS, IGHS, ABSO, DE, and LMSA [56] Applied CSO and SVM to classify students' facial expression e results show 100% classification accuracy for the selected 9 face expressions [39] Applied CSO and SVM to classify students' facial expression e system achieved satisfactory results [40] Applied CSO-GA-PSOSVM to classify students' facial expression e system achieved 99% classification accuracy [23] Applied CSO, HCSO and ICSO in block matching for efficient motion estimation e system reduced computational complexity and provided faster convergence [16,17,57] Used CSO algorithm to retrieve watermarks similar to the original copy CSO outperformed PSO and PSO time-varying inertia weight factor algorithms [58,59] Sabah used EHCSO in an object-tracking system to obtain further efficiency and accuracy e system yielded desirable results in terms of efficiency and accuracy [60] Used BCSO as a band selection method for hyperspectral images BCSO outperformed PSO [61] Used CSO and multilevel thresholding for image segmentation CSO outperformed PSO [62] Used CSO and multilevel thresholding for image segmentation PSO outperformed CSO [63] Used CSO, ANN and wavelet entropy to build an AUD identification system. CSO outperformed GA, IGA, PSO, and CSPSO [64] Used CSO and FLANN to remove the unwanted Gaussian noises from CT images e proposed system outperformed mean filter and adaptive Wiener filter.…”

“…In computer vision, image segmentation refers to the process of dividing an image into multiple parts. Ansar and Bhattacharya and Karakoyun et al [62,63] proposed using CSO algorithm incorporation with the concept of multilevel thresholding for image segmentation purposes. Zhang et al combined CSO and K-median provides better modularity than similar models based on PSO and BAT algorithm [92] Applied MOCSO, fitness sharing, and fuzzy mechanism on CR design MOCSO outperformed MOPSO, NSGA-II and MOBFO [93,94] Applied CSO and five other metaheuristic algorithms to design a CR engine CSO outperformed the GA, PSO, DE, BFO and ABC algorithms [95] Applied EPCSO on WSN to be used as a routing algorithm EPCSO outperformed AODV, a ladder diffusion using ACO and a ladder diffusion using CSO.…”

Cat Swarm Optimization Algorithm - A Survey and Performance Evaluation

“…In computer vision, image segmentation refers to the process of dividing an image into multiple parts. Reference [63,64] 5.3 Signal processing: IIR filter stands for Infinite impulse response. It is a discrete-time filter, which has applications in signal processing and communication.…”

Cat Swarm Optimization Algorithm - A Survey and Performance Evaluation

“…In computer vision, image segmentation refers to the process of dividing an image into multiple parts. Reference [63,64]…”