Adaptive vector quantization based video compression scheme

Combined Source Adaptive and Channel Optimized Matrix Quantization algorithm

“…The CSACOMQ algorithm can be utilised by modern audio and speech codec standards [2][3][4][5][6]9,10], image compression techniques [8] and video compression techniques [7] where common ground is the need for low bit-rate transmission, quick and robust adaptation to varying source statistics and optimization to fast changing noisy channels. The concept of CSACOMQ can be generalised to combine other AMQ techniques with the COMQ algorithm, and can also be applied to more specialised channels models, such as the Flat (i.e., non-selective in frequency) Fading Rayleigh Channel, for mobile/wireless systems.…”

“…Applications of VQ and A VQ include modern audio codecs, such as the AMR-WB+ [2], the Broadvoice audio codec [3], the CELT audio codec [4] and the OPUS interactive audio codec [5,6]. A VQ has been applied, among others, in video compression [7], where the compression scheme is based on adaptive vector quantization of multi-wavelet coefficients, and in image compression [8], where a fuzzy self-adaptive particle swarm optimization algorithm is described extracting a near-optimum VQ codebook. In [9], an A VQ scheme for speech coding is presented, where the VQ of the Line Spectral Frequency (LSF) parameters of speech is optimized for the probability density function (p.d.f.)…”

Enhancement of Image Segmentation osing Automatic Histogram Thresholding

“…[2] The algorithm assumes that the image contains two classes of pixels following bi-modal histogram (foreground pixels and background pixels), it then calculates the optimum threshold separating the two classes so that their combined spread (intraclass variance) is minimal. [13] METHOD DESCRIPTION In Otsu's method we exhaustively search for the threshold that minimizes the intra-class variance (the variance within the class), defined as a weighted sum of variances of the two classes:…”

“…Examples using different histogram thresholding Methods were shown. [2] BülentSankur, Mehmet Sezgin had conducted an exhaustive survey of image thresholding methods with a view to categorize them, express them under a uniform notation, indicate their differences or similarities, and finally as a basis for performance comparison. They had been categorized into six groups according to the information they are exploiting, such as: Histogram shape-based methods, clustering-based methods, entropy-based methods, object attribute-based methods, spatial methods and local methods.…”