Color Image Compression using PCA (original) (raw)

New PCA-based compression method for natural color images

International Workshop on Computer …, 2004

The color information in a natural image can be considered as a highly correlated vector space. This high correlation is the first motivation towards using linear dimensionality reduction methods like principal component analysis for the sake of data compression. In this paper new color image decomposition methods are proposed and compared experimentally. Using a newly proposed gray-scale image colorizing method, a new compression method is proposed for natural color images, that while reducing the spectral redundancy of natural color images, it leaves the spatial redundancy unchanged, to be handled with a specialized spatial-compression method independently, and is proved to be highly efficient.

Color Image Compression Using 2Dimensional Principal Component Analysis (2DPCA

Two dimensional principal component analyses (2DPCA) is recently proposed technique for face representation and recognition. The standard PCA works on 1-dimensional vectors which has inherent problem of dealing with high dimensional vector space data such as images, whereas 2DPCA directly works on matrices i.e. in 2DPCA, PCA technique is applied directly on original image without transforming into 1 dimensional vector. This feature of 2DPCA has advantage over standard PCA in terms of dealing with high dimensional vector space data. In this paper a working principle is proposed for color image compression using 2DPCA. Several other variants of 2DPCA are also applied and the proposed method effectively combines several 2DPCA based techniques. Method is tested on several standard test images and found that the quality of reconstructed image is better than standard PCA based image compression. The other performance measures, such as computational time, compression ratio are ameliorated. A comparative study is made for color image compression using 2DPCA.

Wavelet-PCA-Based Compression Method for Color Images

From the birth of multi-spectral imaging techniques, there has been a tendency to consider and process this new type of data as a set of parallel gray-scale images, (instead of an ensemble of an n-D realization). Although, even now, some researchers make the same assumption, it is proved that using vector geometries leads to more realistic results. In this paper, using a proposed PCA-based eigenimage extraction method, incorporated with a wavelet-based grayscale image compression algorithm, a new compression method is proposed. The paper includes comprehensive performance analysis of the proposed eigenimage extraction and compression methods.

Comparative Study on Image Compression Using Various Principal Component Analysis Algorithms

Principal Component analysis (PCA) is one of the statistical methods employed in image compression. Presented paper deals with four different types of PCA algorithms those are 2D-PCA, 3D-PCA, 2D -Kernel PCA (2D-KPCA) and 3DKPCA. A comparative study is made for all four types of PCA based on their PSNR values. These algorithms are also tested on several standard test images. It has been found that the quality of reconstructed image of 3DKPCA is better than other types of PCA based image compression.

Compression of Color Images Using Clustering Techniques

2013

This paper deals mainly with the image compression algorithms and presents a new color space normalization (CSN) technique for enhancing the discriminating power of color space along with the principal component analysis (PCA) which enables compression of colour images. Context-based modeling is an important step in image compression. We used optimized clustering algorithms to compress images effectively by making use of a large number of image contexts by separating a finite unlabeled data set into a finite and discrete set of natural, hidden data structures, rather than provide an accurate characterization of unobserved samples generated from the same probability distribution. Since images contain large number of varying density regions, we used an optimized density based algorithm from a pool. PCA is used to express the large 1-D vector of pixels constructed from 2-D color image into the compact principal components of the feature space. Each image may be represented as a weighte...

Hybrid Techniques On Color And Multispectral Image For Compression

Image Compression is a technique to reduce the number of bits required to represent and store an image. This technique is also used to compress two dimensional color shapes without loss of data as well as quality of the Image. Even though Simple Principal Component Analysis can apply to make enough compression on multispectral image, it needs to extend another version called Enhanced PCA(E-PCA). The given multispectral image is converted into component image and transformed as Column Vector with help of E-PCA. Covariance matrix and eigen values are derived from vector. Multispectral images are reconstructed using only few principal component images with the largest variance of eigen value. Then the component image is divided into block. After finding block sum value, mean value, the number of bits required to represent an image can be reduced by E-BTC model. The features are extracted and constructed in Table form. The proposed algorithm is repeated for all multispectral images as well as color image in the database. Finally, compression ratio table is generated. This proposed algorithm is tested and implemented on various parameters such as MSE, PSNR. These experiments are initially carried out on the standard color image and are to be followed by multispectral imager using MATLAB.

Image Compression Based on Data Folding and Principal Component Analysis

Image Compression Based on Data Folding and Principal Component Analysis, 2016

image compression assumes a fundamental part in image handling field particularly when we need to send the image through a system. While imaging methods produce restrictive measures of information and preparing expansive information is computationally costly, information compression is crucial instrument for capacity and correspondence purposes. Numerous present compression strategies give a high compression rates however with impressive loss of image quality. This paper displays a methodology for image compression in spatial space utilizing an idea of data folding. data folding procedure has been connected on shading images with various size. A row folding is connected on the gray image grid took after by a column folding iteratively till the image size diminishes to predefined esteem as indicated by the levels of folding and unfolding iteration) reconstruction the original image). While Data unfolding process connected in adores mode. Then using principal component analysis as a statistical technique concerned with elucidating the covariance structure of a set of variables and uses orthogonal transformation to convert that set of observations of possibly correlated variables into a set of values of linearly uncorrelated and ordered variables called principal components. Method is tested on several standard test images and found that the quality of reconstructed image and compression ratio are ameliorated. The proposed Method is tried on a few standard test images and found that the nature of reproduced image and compression proportion are improved.

A Novel Color Image Compression Method Using Eigenimages

Journal of Iranian Association of Electrical and Electronics Engineers, 2008

From the birth of multi–spectral imaging techniques, there has been a tendency to consider and process this new type of data as a set of parallel gray–scale images, (instead of an ensemble of an n–D realization). Although, even now, some researchers make the same assumption, it is proved that using vector geometries leads to more realistic results. In this paper, based on the proposed method for extracting the eigenimages of a color image, a new color image compression method is proposed and analyzed which performs in the vectorial domain. Experimental results show that the proposed

Principal Component Analysis using Singular Value Decomposition for Image Compression

International Journal of Computer Applications, 2014

Principal components analysis (PCA) is one of a family of techniques for taking high-dimensional data, and using the dependencies between the variables to represent it in a more tractable, lower-dimensional form, without losing too much information. PCA is one of the simplest and most robust ways of doing such dimensionality reduction. It is also one of the best, and has been rediscovered many times in many fields, so it is also known as the Karhunen-Lo_eve transformation, the Hotelling transformation, the method of empirical orthogonal functions, and singular value decomposition.

Color PCA eigenimages and their application to compression and watermarking

Image and Vision Computing, 2008

From the birth of multi-spectral imaging techniques, there has been a tendency to consider and process this new type of data as a set of parallel gray-scale images, instead of an ensemble of an n-D realization. However, it has been proved that using vector-based tools leads to a more appropriate understanding of color images and thus more efficient algorithms for processing them. Such tools are able to take into consideration the high correlation of the color components and thus to successfully carry out energy compaction. In this paper, a novel method is proposed to utilize the principal component analysis in the neighborhoods of an image in order to extract the corresponding eigenimages. These eigenimages exhibit high levels of energy compaction and thus are appropriate for such operations as compression and watermarking. Subsequently, two such methods are proposed in this paper and their comparison with available approaches is presented.