Color image quantization using Gaussian Particle Swarm Optimization(CIQ-GPSO)

Barman, Dibyendu; Hasnat, Abul; Sarkar, Suchintya; Murshidanad, Md. Atiqur Rahaman

doi:10.1109/inventive.2016.7823295

Cited by 14 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In fact, (non-stochastic) function approximation has been shown to improve evolutionary algorithms, as the surrogate can be used to evaluate additional candidate solutions within a local neighborhood, while keeping the number of function calls to f unchanged (Regis, 2014a). Examples are surrogate-assisted variants of Gaussian PSO (e. g. Krohling, 2004;Melo & Watada, 2016;Varma et al, 2013;Barman et al, 2016;Liu et al, 2013;Gao et al, 2020), Bayesian PSO (e. g. Zhang et al, 2015;Chen & Yu, 2017;Kang et al, 2018), and modified PSO (e. g. Tian & Shi, 2018;Liu et al, 2015). However, surrogate-assisted algorithms are primarily used to speed up runtime (due to fewer evaluations f ) but with similar convergence characteristics.…”

Section: Surrogate-assisted Psomentioning

confidence: 99%

Directed particle swarm optimization with Gaussian-process-based function forecasting

Jakubik¹,

Binding²,

Feuerriegel³

2021

Preprint

View full text Add to dashboard Cite

Particle swarm optimization (PSO) is an iterative search method that moves a set of candidate solution around a search-space towards the best known global and local solutions with randomized step lengths. PSO frequently accelerates optimization in practical applications, where gradients are not available and function evaluations expensive. Yet the traditional PSO algorithm ignores the potential knowledge that could have been gained of the objective function from the observations by individual particles. Hence, we draw upon concepts from Bayesian optimization and introduce a stochastic surrogate model of the objective function.That is, we fit a Gaussian process to past evaluations of the objective function, forecast its shape and then adapt the particle movements based on it. Our computational experiments demonstrate that baseline implementations of PSO (i. e., SPSO2011) are outperformed. Furthermore, compared to, state-of-art surrogate-assisted evolutionary algorithms, we achieve substantial performance improvements on several popular benchmark functions. Overall, we find that our algorithm attains desirable properties for exploratory and exploitative behavior.

show abstract

Section: Surrogate-assisted Psomentioning

confidence: 99%

Directed particle swarm optimization with Gaussian-process-based function forecasting

Jakubik¹,

Binding²,

Feuerriegel³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In case of lossy image compression huge amount data loss occur as a result blocking artifacts, ring artifacts may appear in the decompressed image hence visual quality of the image degraded drastically but the amount of compression ratio achieved is very high. Vector Quantization (VQ) [7][8][9][10][11][12][13], Color Image Quantization(CIQ) [17][18][19][20][21][22][23], JPEGare the most popular lossy image compression technique [2][3][4].…”

Section: ) Lossy Compressionmentioning

confidence: 99%

Improved LZW Compression Technique using Difference Method

Barman*,

Ahamed

2020

IJITEE

View full text Add to dashboard Cite

This work attempts to give a best approach for selecting one of the popular image compression algorithm. The proposed method is designed to find the best performance approach amongst the several compression algorithms. In this work existing lossless compression technique LZW (Lempel-Ziv-Welch) is redesigned to achieve better compression ratio. LZW compression technique works on the basis of repetition of data. In a situation where all the values are distinct or repetition of data does not present the LZW can’t work properly. To avoid this problem, difference method called difference matrix method which is actually calculate the difference between two consequence data and store it in a resultant matrix is used. In this case the matrix contains repetitive data which is more effective compared to LZW technique. Another problem of LZW is dictionary overflow, because of LZW works on ASCII character there is a limit of 256 dictionary length initially. In this work dynamic dictionary method is used without using the ASCII rather than this static method. As a result, this dictionary can contain the initial value anything in a range of -256 to 255. Here ASCII values are not used because the proposed method is applicable grayscale image, where the pixel values are between in range 0 to 255. Using these two changes the proposed improved LZW method becomes more powerful that can compress a non-repetitive set of data significantly. The proposed method is applied on many standard gray images found in the literature achieved 7% to 18% more compression the normal LZW keeping quality of the image same as existing.

show abstract

“…The second step is pixel-mapping, it is done by changing each color with the color in the palette. Color quantization is an implementation of lossy image compression [7][8][9][10]. Figure 1 shows the representation between a 24-bits original image and 64 colors of quantized image.…”

Section: Introductionmentioning

confidence: 99%

Quality and size assessment of quantized images using K-Means++ clustering

Ongkadinata¹,

Putri²

2020

Bulletin EEI

View full text Add to dashboard Cite

In this paper, an amended K-Means algorithm called K-Means++ is implemented for color quantization. K-Means++ is an improvement to the K-Means algorithm in order to surmount the random selection of the initial centroids. The main advantage of K-Means++ is the centroids chosen are distributed over the data such that it reduces the sum of squared errors (SSE). K-Means++ algorithm is used to analyze the color distribution of an image and create the color palette for transforming to a better quantized image compared to the standard K-Means algorithm. The tests were conducted on several popular true color images with different numbers of K value: 32, 64, 128, and 256. The results show that K-Means++ clustering algorithm yields higher PSNR values and lower file size compared to K-Means algorithm; 2.58% and 1.05%. It is envisaged that this clustering algorithm will benefit in many applications such as document clustering, market segmentation, image compression and image segmentation because it produces accurate and stable results.

show abstract

Color image quantization using Gaussian Particle Swarm Optimization(CIQ-GPSO)

Cited by 14 publications

References 9 publications

Directed particle swarm optimization with Gaussian-process-based function forecasting

Directed particle swarm optimization with Gaussian-process-based function forecasting

Improved LZW Compression Technique using Difference Method

Quality and size assessment of quantized images using K-Means++ clustering

Contact Info

Product

Resources

About