We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

When is nearest neighbor meaningful

This chapter describes digital image compression techniques, and the JPEG image compression standard. Digital image compression techniques are based on algorithms such as the Wavelet transform, Fractal images, Vector Quantisation and Discrete Cosine Transform (DCT). The digital image compression technique developed by the Joint Photographic Experts Group (JPEG) is mainly based on the quantisation of the DCT.

Digital Image Compression Techniques

In the field of image processing, there are several problems where the efficient search has to be performed in complex search domain to find an optimal solution Image enhancement which improves the quality of an image for visual analysis and/or machine understanding is one of these problems There is no unique image enhancement technique and it’s measurement criterion which satisfies all the necessity and quantitatively judge the quality of a given image respectively Thus sometimes proper image enhancement problem becomes hard and takes large computational time In order to overcome that problem, researchers formulated the image enhancement as optimization problems and solved using Nature-Inspired Optimization Algorithms (NIOAs) which starts a new era in image enhancement field This study presents an up-to-date review over the application of NIOAs in image enhancement domain The key issues which are involved in the formulation of NIOAs based image enhancement models are also discussed here

A Survey on Nature-Inspired Optimization Algorithms and Their Application in Image Enhancement Domain

Reversible data hiding with automatic contrast enhancement for medical images

A method and apparatus for adaptively encoding a media stream with compound prediction are provided. Adaptively encoding a media stream with compound prediction may include decoding a first compound prediction use indicator for decoding a current frame, on a condition that the first compound prediction use indicator indicates that the encoded video stream includes a compound prediction mode for the current block, decoding the compound prediction mode for the current block from the encoded video stream and decoding the current block using the compound prediction mode, and on a condition that the first compound prediction use indicator indicates that the encoded video stream includes a second compound prediction use indicator, decoding the second compound prediction use indicator from the encoded video stream and decoding the current block based on the second compound prediction use indicator; and outputting or storing the decoded current block.

Adaptively encoding a media stream with compound prediction

Multimedia security facilitates the protection of information in multiple forms of digital data such as text, image, audio and video. Many approaches are available for protecting digital data; these include Cryptography, Steganography, and Watermarking etc. A novel approach is proposed for multimedia data security by integrating Steganography and Visual Cryptography (VC) with the goal of improving security, reliability and efficiency. The proposed technique consists of two phases. The first phase is used to hide the message dynamically in an Cover Image1 by changing the number of bits hidden in RGB channels based on the indicator value. VC schemes conceal the Cover Image2 into two or more images which are called shares. In the second phase two shares are created from a Cover Image2 and the stego image created in the first phase is hidden in these two shares. The shares are safe as they reveal nothing about the multimedia content. The Cover Image2, stego image and the hidden message can be recovered from the shares without involving any complex computation. Experimental results show that the new scheme is simple and retrieves many multimedia contents and accomplishes a high level of security.

Multilevel Multimedia Security by Integrating Visual Cryptography and Steganography Techniques

Machine Learning‐Based Recommender System for Breast Cancer Prognosis

Agricultural productivity is one of the important sectors that influence the Indian economy. One of the greatest challenges that affect agricultural productivity is plant disease which is quite prevalent in almost all crops. Hence, plant disease detection has become a hot research area to enhance agricultural productivity. Automated detection of plant diseases is hugely beneficial to farmers as it reduces the manual workload of monitoring and detection of the symptoms of diseases at a very early stage itself. In this work, an innovative method to categorize the tomato and maize plant leaf diseases has been presented. The efficiency of the proposed method has been analyzed with plant village dataset.

A Novel Method for Plant Leaf Disease Classification Using Deep Learning Techniques

We develop and test a new, two-stage, residual vector quantization algorithm using variable bit-rate encoding In the first stage, we partition the input image into non-overlapping blocks, vector-quantize and code them by a small codebook using the well-known K-means algorithm The novelty in this method is the use of high eigen-valued blocks as initial seeds which serve as good distributors in the formation of clusters and fast convergence We compute the residual vectors and classify them based on threshold values of distortion and variance Vectors above the given threshold require second-stage coding In the second stage, we partition the residual vectors further into small sub blocks and scalar-quantize each sub block to form number patterns instead of performing Direct Vector Quantization (DVQ) These number patterns, which form the secondary codebook, are easily generated without complex calculations by applying basic ideas from combinatorics Both the intra-block and inter-block correlation properties have been exploited to enhance the compression rate This method offers several advantages: 1) the computational complexity is greatly reduced; 2) exhaustive comparisons in DVQ are carried out more efficiently; 3) the picture quality of the reconstructed image is not compromised; and, 4) a reduced bit-rate is achieved

A Pattern-Based Residual Vector Quantization (PBRVQ) algorithm for compressing images

proposed method is a two phase scheme that enhances the performance of K-means vector quantization algorithm for compressing images. In the proposed method, we have explored the possibility of application of statistical parameters for choosing the initial seeds for K-means algorithm. The selection of initial seeds depends on the statistical features of input data set. The novelty in our approach is the judicious selection of initial seeds based on variance, mean, median and mode parameters. Considering mode value of each dimension of the data adds uniqueness to our method. Our approach shows better performance yielding good PSNR and variable bit rate at a very low time complexity. This method is best suited for online web applications that involve massive and rapid image and video transmission.

/pdf/novel-k-means-algorithm-for-compressing-images-35bicr0all.pdf

M. Mary Shanthi Rani

Papers

Multilevel Multimedia Security by Integrating Visual Cryptography and Steganography Techniques

Machine Learning‐Based Recommender System for Breast Cancer Prognosis

A Novel Method for Plant Leaf Disease Classification Using Deep Learning Techniques

A Pattern-Based Residual Vector Quantization (PBRVQ) algorithm for compressing images

Novel K-means Algorithm for Compressing Images