Roumen Kountchev

Inverse pyramidal decomposition with multiple DCT

Abstract This paper presents a method for pyramidal image decomposition called “inverse” because of the order followed to obtain the pyramid levels: from top to bottom, in correspondence with the requirement for “progressive” image transmission. The pyramid top (level zero) consists in selecting the low-frequency coefficients of the discrete cosine image transform. The following pyramid levels are made up of low-frequency discrete cosine transform (DCT) coefficients of the subimages obtained from quadtree division at each level. The quadtree root coincides with the pyramid top. The first level is the difference between the image and its approximation obtained by inverse DCT. The following (second) level is a difference too, between the previous (first) level and its approximation obtained with inverse DCT for every subimage in the first level, etc. The paper describes the principle of image decomposition, the possibilities for recursive calculation, its basic characteristics and modifications. The block diagram and the generalised scheme of the decomposition are given and some results of its modelling show the application capacities in image coding systems.

Image Representation with Reduced Spectrum Pyramid

The paper presents one new approach for image decomposition based on spectrum pyramid with reduced number of coefficients (lower or equal to the number of the image pixels). The decomposition permits multi-layer image transfer with high compression ratio and good visual quality. The computational complexity of the new decomposition is relatively low. Some results, obtained with the simulation of the presented algorithm and the most important application areas are presented in the paper as well.

Documents Image Compression with IDP and Adaptive RLE

In the paper is presented a new method for efficient lossless documents image compression, based on the inverse difference pyramid (IDP) decomposition. The method is aimed at the processing of graphic and text grayscale and color images. The IDP decomposition is presented in brief and the method ability to process efficiently different kinds of images are presented. In the paper are included the results of the compression of graphics and texts and they are compared with those obtained with JPEG2000 and other widely used lossless compression methods

Multimedia watermarking with complex Hadamard transform in the inverse pyramid decomposition

A new method for digital watermarking of multimedia signals (audio signals and images) is offered, based on decomposition with inverse difference pyramid (IDP), whose coefficients are obtained with complex Hadamard transform (CHT). Advantages of the method, compared with those based on DFT or DWT transform, are that there is no quantization of the values of the transform coefficients; it has lower computational complexity and permits insertion of different watermark in every consecutive pyramid level. The method ensures practical invisibility (insensibility) of the inserted watermark, and high resistance against tampering, compression, affine transforms, filtration, dithering and other kinds of processing. Together with this, the method permits sure detection of the inserted watermark and low probability for mistakes in the water mark extraction.

Image Color Space Transform with Enhanced KLT

The use of the Karhunen-Loeve Transform (KLT) for the processing of the image primary color components gives as a result their decorrelation, which ensures the enhancement of such operations as: compression, color-based segmentation, etc. The basic problem is the high computational complexity of the KLT. In this paper is offered a simplified algorithm for the calculation of the KL color transform matrix. The presented approach is based on non-recursive approach for the color covariance matrix eigenvectors detection. The new algorithm surpasses the existing similar algorithms in its lower computational complexity, which is a prerequisite for fast color segmentation or for adaptive coding of color images aimed at real time applications.

Adaptive Compression of Compound Images

In the paper is presented new method for efficient compression of compound still images, containing pictures and texts/graphics. The method is based on the inverse difference pyramid (IDP) image decomposition and lossless coding of the obtained data. The method permits the recognition of texts and graphics in compound images, the setting of corresponding regions of interest (ROI) and their coding with the most efficient tools. The method ensures easy access and transfer of visual information via Internet aimed at distance learning applications.

Resistant image watermarking in the phases of the Complex Hadamard Transform coefficients

In the paper a new method is presented for digital content protection based on watermark data insertion in the image transform domain. For this, the still digital image is transformed using Complex Hadamard Transform (CHT), and the watermark data is then inserted in the imaginary part of the transform coefficients. The selection of the suitable for watermarking transform coefficients is done in accordance with pre-defined rules. The so inserted watermark is perceptually invisible. The method permits the insertion of relatively large amount of data, retaining the high quality of the protected image. The main advantages of the algorithm for digital watermarking, based on the CHT are that it is resistant against attacks, based on high-frequency filtration (JPEG compression); it permits the insertion of significant amount of data, and the watermark detection could be done without using the original image.

Contrast Enhancement with Histogram-Adaptive Image Segmentation

In the paper is presented a specific approach aimed at the improvement of the visual quality of underexposed or low-contrast images. For this is developed a new contrast-enhancement algorithm based on the segmentation of the image area with relatively high density of dark elements. The problem is solved changing the brightness intervals of the selected segments followed by equalization (in particular - linear stretch and skew) of the corresponding parts of the histogram. The method implementation is relatively simple and permits easy adaptation of the contrasting algorithm in accordance with the image contents, requiring the setting of small number of parameters only. The obtained results prove the efficiency of the new method on the processed image quality

Compression of CT Images with Branched Inverse Pyramidal Decomposition

In this chapter a new approach is suggested for compression of CT images with branched inverse pyramidal decomposition. A packet of CT images is analyzed and the correlation between each couple inside it is found. Then the packet is split into groups of images with almost even correlation, typically into six or more. One is chosen as a referent being mostly correlated with all of the others. From the rest difference images with the referent are found. After the pyramidal decomposition a packet of spectral coefficients is formed and difference levels which are coded by entropy coder. Scalable high compression is achieved at higher image quality in comparison to that of the JPEG2000 coder. The proposed approach is considered perspective also for compression of MRI images.

Compression of multispectral images with inverse pyramid decomposition

In the paper is presented a new method for compression of multispectral images, based on the Inverse Difference Pyramid decomposition. The method is applicable for any number of multispectral images of same object. The processing is performed as follows. First, the histograms of the multispectral images are calculated and compared. The image, whose histogram is most similar with these of the remaining ones, is chosen to be a reference one. The image decomposition starts with the reference image, which is processed with some kind of orthogonal transform, using a limited number of transform coefficients only. With the so obtained coefficients values is calculated the coarse approximation of the processed image. The IDP decomposition then branches out into several directions, corresponding to the number of multispectral images. The first approximation for all multispectral images is that of the reference image. Each branch is developed individually, using the same approximation. In result of this processing is obtained high compression and very good visual quality of the restored images. This approach gives better results than these, obtained with methods, based on the JPEG and JPEG 2000 standards.