Earticle

The aim of image restoration is to restore the image affected by degradations to the most desired form. It comprises a set of techniques applied to the degraded image to remove or reduce the cause of degradations. This study focuses on Astronomy images. Astronomy images suffer from mainly two types of degradations: atmospheric turbulence blur and additive white Gaussian noise. This study presents a new method to restore astronomy images by proposing a hybrid method that combines three techniques to restore a degraded image. The first technique is phase preserving algorithm used for the denoising operation. Then a normalization operation is employed to provide the image its natural grayscale intensity. After that Richardson Lucy deblurring algorithm is used to deblur the image depending on the Point Spreading Function (PSF) determined earlier. When the deblurring process is completed, the anticipated image will be in the most desirable form.

Power spectral analysis of the heart rate and blood pressure variations has commonly used to provide indices of autonomic cardiovascular modulation, but the effect of changing posture from lying to standing on these variations and the interpretation of their power spectra is still largely in dispute. It was due to the reason that till now no study was made yet that clearly outlines the variations in terms of RR-interval and blood pressure series from lying to standing position. Thus the aim of this paper lies in the application of classifying the subjects based on their RR-intervals, systolic and diastolic blood pressure series, prior to spectral analysis, at two different physical activity related postures. In this paper K-Nearest Neighbor algorithm has been proposed as a classifier for classifying the subjects based on lying and standing postures. Here we also studied the classification accuracy achievable with a KNN classifier using three different methods (i) Euclidean (ii) City block and (iii) Correlation of calculating the nearest distance in order to propose the optimal one. Further an attempt has been made to evaluate each of these methods for five different values of K=1, 3, 5, 7 and 9 in order to propose the best fit value of K for classifying the subjects. After performing the comparative analysis between these three methods of distance metrics and for different values of K, it is found that K=1 is the best choice out of 3, 5, 7 and 9 and Correlation has been emerged as one of the optimal method for computing the nearest distance with highest classification accuracy of 98.60 % with K=1 for lying and 99.95 % with K=1 for standing postures.

Ensemble classifier is a combining approach to improve the accuracy of the simple classifiers. In this article, we introduced a new method for static handwritten signature verification based on an ensemble classifier. In our introduced method, after pre-processing stage, signature image is convolved with Gabor wavelets to compute the Gabor coefficients in different scales and directions. Three different feature sets are extracted from resulting Gabor coefficients using statistical approaches. A nearest neighbor classifier classifies each feature set by an adaptive method. The proposed ensemble classifier combines the output of the three simple classifiers, which are essentially the same. Although these simple classifiers looks the same, but the different input feature set and the adaptive thresholds related to each classifier makes them to be different with each other. Therefore, from the viewpoint of the classifiers combination, the proposed method can be considered as a feature level combination type. The proposed method was evaluated by applying on two datasets: Persian and South African signature datasets. Experimental results shown our proposed method has the lowest error rate in comparison with other methods.

Automatic extraction of the index of broadcast streams from radio and television has become a challenging research topic over the last years. The automatic classification of audio types, such as speech, music, noises/atypical events etc, has found numerous applications. In this paper we study the evaluation of different machine learning algorithms, which have successfully been used in other classification tasks, on the task of classification of audio broadcast news. The audio classification scheme consists of pre-processing, audio parameterization with established audio features, and classification to acoustic events. The experimental evaluation was carried out using the Voice of America broadcast recordings database for the Greek language. The experimental results indicated that the best performance, approximately 92% of accuracy, was achieved by the classification scheme using the boosting technique with decision trees.

The objective of this paper is to propose a new algorithm in face detection that has the capabilities of detecting the face with different poses and under different conditions. This objective is obtained in different stages and using different proposed algorithms. Firstly, a robust segmentation algorithm is proposed to extract and segment the skin region from the image. Secondly, different filtering steps are applied to this segmented image to obtain the face candidate region only. After that, Feature-Based approach is used to detect the features from this candidate face which can work in real-time with minimal training in contrast to other approaches such as image-based approach. Finally, some rules is applied in order to judge if this candidate is profile face or not either the profile face is right or left. Experimental results show that the proposed method is robust under a wide range of lighting conditions, different poses and different races. These results are taken from three different face databases. The proposed method is implemented using Matlab version 7.6 software and gives a correct detection rate reach 90 %.

In this paper, rotation-extension CORDIC methods, i.e. double-rotation and triple- rotation, are proposed for the objective of improving the performance and accuracy of the CORDIC computational algorithm in radix-2. In the double-rotation and triple-rotation methods, the convergences of the CORDIC computations are accelerated by duplicating and triplicating the micro-rotation angles to be 2and 3, respectively. The non-redundant mechanism, where a rotation direction is in a set of 1 and -1, depending on an intermedi- ate converging parameter (either y or z), is applied to constant scaling factors. Convergence range and accuracy of elementary functions hardware performed by using the CORDIC methods in rotation mode and vectoring mode on the circular, hyperbolic, and linear co- ordinate systems are examined, investigated and compared to Matlab/Simulink simulation results. The comparison results show that the proposed CORDIC methods provide higher computational accuracy than the conventional one at the same number of iterations. A high precision CORDIC algorithm is introduced and evaluated for VLSI implementation. Finally, speed and area performance of the CORDIC hardware based on the pipeline (un- folded) digit-parallel architecture of the proposed CORDIC methods are compared to the CORDIC methods previously published in the literature.

So far, numerous methods have been proposed for face recognition based on single image person. In this paper, we introduce a new recognition algorithm based on feature vectors of three types of moments and fuzzy combination of them. We used an ethnic database in our experiments that are distinctly different from other famous databases such as AR and FERET. In ethnic database women have Hijab that is important feature in Muslim community such as Arabian countries. Also most of men have beard or mustache in these nations. Experimental results show that these features increase the recognition rate. Really, these characteristics eliminate the false recognition between male and female in the process. Subsets of 100 images from ethnic, AR and FERET are used in our experiments. Proposed method achieves 96% accuracy for ethnic database that is higher than two other databases.

This paper proposes a novel approach for processing digital mammograms to detect micro-calcifications. They may be so small that they are almost undetectable visually, but it could be indicators of a possible malignancy. An analysis algorithm based on optical holographic property of images and clustering principles are proposed to detect the micro-calcifications. This process consists of three stages. In the first stage the mammographic patterns are subjected to optical holographic analysis. The resulting images are passed to the second stage, in which morphological operations are performed. The third stage detects the malignant portions of the mammographic pattern using unsupervised texture classification by extracting laws features. Texture classification is an important image processing task with a broad application range. Many different techniques for texture classification have been explored. This paper explores the unsupervised classifications of digital mammograms using K-means and Fuzzy C-means approaches. Results show that the proposed techniques detect the malignant portions of the breast very well thus enabling earlier detection of tumor.

The human-computer interaction depended computing mode is evolved to an intelligent adaptive context aware computing. The corresponding middleware should be adaptive, reflective, dynamic reconfigurable and physical/virtual context aware. In this paper the design of a context-aware middleware is discussed. Some key issues involved, e.g. virtual and meta sensors, reflective context model, multi-agent mechanism, an extension of Web Service-Open Web Service, etc. are analyzed and studied. An experimental system and its design are introduced.

In this work, we have presented a hierarchical scheme for classifying music data. Instead of dealing with large variety of features, proposed scheme relies on MFCC and its variants which are introduced at the different stages to satisfy the need. At the top level music is classified as song (music with voice) and instrumental (music without voice) based on MFCC. Subsequently, instrumental signals and songs are classified based on instrument type and genres respectively. Hierarchical approach has been followed for such detailed categorization. Using two-stage process, instrumental signals are identified as one of the four types namely, string, woodwind, percussion or keyboard. Wavelet and MFCC based features are used for this purpose. For song classification, at first level signals are categorized as classical or non-classical(popular) ones by capturing the MFCC pattern present in the high sub-band of wavelet decomposed signal. At second level, we consider the task of further classification of popular songs into various genres like Pop, Jazz, Bhangra (an Indian genre) based on amplitude variation pattern. RANSAC has been utilized as the classifier at all stages. Experimental result indicates the effectiveness of the proposed schemes.

An effecting compression algorithm removes the redundancy of image signal, when we want to represent high quality videos and images with lower bit rate. Removal of statistical correlation and insignificant components of image signal make the corresponding videos and images highly compressed. This paper represents a new algorithm to measure the blocking artifacts of videos by analyzing the distortions of local properties of image signals like dominant edge magnitude and direction. We also want to incorporate light weighted human vision measurement system like edge information to measure video artifacts in real time. Then simple bucket filling approach is applied, where the particular bucket contains the maximum value also indicating the block boundaries that are passed to the report module. After computing distortion measure a proposed detection approach is used to capture the distorted frames. Extensive experiments on various videos show that the new algorithm is very much efficient and faster to measure the independent lossy compressed block errors in real time video error detection applications.