This PDF file contains the front matter associated with SPIE Proceedings Volume 8285, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

The modern era of interior-point methods (IPM) dates to 1984, when Karmarkar proposed his algorithm for linear programming. In the years since then, interior-point algorithm and software for linear problems have become quite sophisticated and gained extraordinary interest as an alternative to the sparse simplex based methods. In this paper, we introduce the infeasible primal-dual algorithm which is widely considered the most efficient general purpose IPM. Then many issues that are crucial for the efficient implementation of an interior-point algorithm are addressed in detail. To facilitate the understanding of different implementation strategies, some illustrative numerical results on a subset of problems from the Netlib collection are presented.

In this article we described methods of link spam detection with using of machine learning. We analyzed main factors of link spam, which helps to find them. There is algorithm of link spam detection, based on support vector machines. The methods of link spam detection shows good results

Increased organizational dependence on data warehouse (DW) systems drives management attention towards improving DW systems success. However, the successful implementation rate of DW systems is low and many firms did not achieve intended goals. A recent studies show that improves and evaluates DW success is one of the top concerns facing IT/DW executives. Existing information system (IS) research has studied DW success more from information quality and system quality. Researchers argue in this study that we should also take the relationship quality, which has significant research and practical implications in that it connects to IS success directly. As our first attempt, this study, referring to both IS and marketing literature, examines how communication, coordination, cooperation, commitment, and trust can be achieved to some degrees by high quality relationships between DW parties.

Power of a class reflects the relative importance of the class in an object-oriented software system, and has many implications for resource allocating. And how to measure it is still a problem we face. They paper presents to measure it based the connection structure of a network constructed from the software. To support this, we first use a network model to describe the connection structure. Then some centrality metrics used in social network analysis are introduced to quantify the power of classes. The commons and differences among these centrality metrics are revealed. We think these metrics can be used together to have a more systematic view of the importance of each class.

Breast cancer continues to be a significant public health problem in the world. Early detection is the key for improving breast cancer prognosis. The aim of the research presented here is in twofold. First stage of research involves machine learning techniques, which segments and extracts features from the mass of digital mammograms. Second level is on problem solving approach which includes classification of mass by performance based case base classifier. In this paper we build a case-based Classifier in order to diagnose mammographic images. We explain different methods and behaviors that have been added to the classifier to improve the performance of the classifier. Currently the initial Performance base Classifier with Bagging is proposed in the paper and it's been implemented and it shows an improvement in specificity and sensitivity.

This paper contains the case description of reviewing of different research papers for scientific conference and Models related to the chosen case description. The evaluation of the business models is based on the Sequal quality framework. This paper is based on the case description as provided in later sections of this paper; however the case is extended and focused as necessary to explain the business models.

This paper describes about real-time moving pedestrian object detection and tracking. In this work, a new pedestrian object tracking approach is proposed, which uses a new texture model for better representing the pedestrian object. This approach based primarily on a step of object detection for automatic pedestrian object area extraction, features extraction step for pedestrian object representation and finally tracking step based on region matching method. We propose a new texture model for discriminant pedestrian object representation. This model based on a set of Haralick indexes computed from co-occurrence matrix. The presented approach can handle orientation, object scale, illumination changes, and clutter and complex occlusions.

Remote sensing image have been known to be an important part for environmental analysis. Drought early warning system is one of the few example of remote sensing image applications. One of the oldest tool in remote sensing studies, NDVI is often used for drought detection. Although very essential, remote sensing image requires large storage requirement. A large image data may cause network congestion which certainly affects the aptitude of the drought detection system. An image compression may be used as an approach to this issue. However, this would lead to another issue, image quality. This article emphasize on the effect of image compression through transformation towards remote sensing image. Analysis is conducted through NDVI pixel threshold as well as other complimentary error metric method. Hybrid methods of transformation are presented here for the image transformation process. The experiments performed on test images shows that hybrid transformation is capable of reducing image data and preserving sufficient quality.

Graph Drawing is a branch of computer graphics that addresses the problem of constructing geometric representations of conceptual structures that are modeled by graphs. Crossing reduction is one of the crucial problems in drawing networks that are presented in a hierarchical manner which is solved by repeated layer-by-layer techniques or by global ones. In this paper, we introduce a new efficient algorithm for drawing hierarchical graphs based on the well-known barycenter algorithm. Our algorithm improves the existing layer-by-layer techniques by 15-25% and recent global algorithms by 5% further reduction in the number of crossings.

Video based pupil detection techniques are useful for Perceptual user interfaces and human monitoring. Pupil detection plays a vital role in identification systems. This paper presents an algorithm for pupil detection in video images under active infrared (IR) illumination. At first, using a low pass filter the eyelashes are removed. Following, the thresholding process takes place. Using the Otsu's thresholding method, the proper thresholding value is estimated. Then, using the genetic algorithm, filter band width and threshold value are estimated accurately. After the pupil segmentation process, pupil boundary is modelled. The modeling process is based on circular Hough transform. Experimental results show promising performances on eye video images captured in ideal and non-ideal conditions.

Enterprise Resource Planning systems are the combination of different business IS (Information System) applications that are designed according to the organization requirements. Generally ERP systems are suffering from complex user interface issues. Recent research shows that there is a need for improvement concerning, the user interface from their perspectives. In order to design the software applications that are easy to use, memorize and apply to new problems, we must know the users philosophy and something about learning, reminiscence and problems solving. The Usability engineering is the only way to study the deeds of users while interacting with ERP (Enterprise Resource & Planning). This paper will focus on the users' experiences view of financial module in ERP system.

Image transmission takes place as an important research branch in multimedia broadcasting communication systems in the last decade. Our paper presents image transmission over a FFT-OFDM (Fast Fourier Transform Orthogonal Frequency Division Multiplexing). The need for encryption techniques increase with the appearance of the expression which said that our world became small village, and the use of image application such as conference and World Wide Web which increase rapidly in recent years. Encryption is an effective method for protecting the transmitted data by converting it into a form being invisible over transmission path and visible in receiver side. This paper presents a new hybrid encryption technique based on combination of Backer maps and logistic map. This proposed technique aims to increase PSNR and reduce the noise in the received image. The encryption is done by shuffling the positions of a pixel image using two dimensional Baker maps then encrypt using XOR operation with logistic map to generate cipher image over orthogonal frequency multiplexing (OFDM). The encryption approach adopted in this paper is based on chaotic Baker maps because the encoding and decoding steps in this approach are simple and fast enough for HDTV applications. The experimental results reveal the superiority of the proposed chaotic based image encryption technique using two logistic maps and two dimensional Backer map over normal Backer map.

The paper studied the sub-pixel positioning theory and algorithms of image processing and measurement. we derived the formular and had the boundary points located to Sub-pixel level.

New technologies, especially, the Internet have made a huge impact on knowledge management and information dissemination in education. The web portal as a knowledge management system is very popular topics in many organizations including universities. Generally, a web portal defines as a gateway to online network accessible resources through the intranet, extranet or Internet. This study develops a knowledge portal for the students in the Faculty of Computer Science and Information Technology (FCSIT), University of Malaya (UM). The goals of this portal are to provide information for the students to help them to choose the right courses and major that are relevant to their intended future jobs or career in IT. A quantitative approach used as the selected method for this research. Quantitative method provides an easy and useful way to collect data from a large sample population.

A unified framework for integrating PC cluster based parallel rendering with distributed virtual environments (DVEs) is presented in this paper. While various scene graphs have been proposed in DVEs, it is difficult to enable collaboration of different scene graphs. This paper proposes a technique for non-distributed scene graphs with the capability of object and event distribution. With the increase of graphics data, DVEs require more powerful rendering ability. But general scene graphs are inefficient in parallel rendering. The paper also proposes a technique to connect a DVE and a PC cluster based parallel rendering environment. A distributed multi-player video game is developed to show the interaction of different scene graphs and the parallel rendering performance on a large tiled display wall.

Segmentation of an image plays an important role in image processing applications. In this paper segmentation of pomegranate magnetic resonance (MR) images has been explored. Pomegranate has healthy nutritional and medicinal properties for which the maturity indices and quality of internal tissues play an important role in the sorting process in which the admissible determination of features mentioned above cannot be easily achieved by human operator. Seeds and soft tissues are the main internal components of pomegranate. For research purposes, such as non-destructive investigation, in order to determine the ripening index and the percentage of seeds in growth period, segmentation of the internal structures should be performed as exactly as possible. In this paper, we present an automatic algorithm to segment the internal structure of pomegranate. Since its intensity of stem and calyx is close to the internal tissues, the stem and calyx pixels are usually labeled to the internal tissues by segmentation algorithm. To solve this problem, first, the fruit shape is extracted from its background using active contour model (ACM). Then stem and calyx are removed using morphological filters. Finally the image is segmented by fuzzy c-means (FCM). The experimental results represent an accuracy of 95.91% in the presence of stem and calyx, while the accuracy of segmentation increases to 97.53% when stem and calyx are first removed by morphological filters.

Cosmic Movement Detection is concerned with the difficult task of taking the images of sky through high powered telescope or satellite transmitted images and performing image processing in order to discover new galaxies, stars and other cosmic objects or to describe already known galaxies, stars and other cosmic objects. Description meant to describe the type of cosmic object under consideration, whether it's already been recognized previously or not, whether it's moving close to Earth or moving away from Earth. It has several applications astrophysics, astronomy and astroscience. Automating this process to a computer requires the use of various image processing techniques. The method which the present paper describes is based on Doppler Effect of light i.e. red and blue shifting property. Several factors like poor illumination, noise disturbance, viewpoint-dependence, Climate factors, Transmission and Imaging conditions can affect the algorithm working. This paper reports an algorithm for Cosmic Movement Detection using image processing and Doppler Effect of light. The present paper suggests Cosmic Movement can be detected checking the color of the galaxy and it can help in determining whether it's moving towards Earth or its moving away.

This paper primarily exploits some of the more obscure, but inherent properties of camera and image to propose a simpler and more efficient way of perceiving depth. The proposed method involves the use of a single stationary camera at an unknown perspective and an unknown height to determine depth of an object on unknown terrain. In achieving so a direct correlation between a pixel in an image and the corresponding location in real space has to be formulated. First, a calibration step is undertaken whereby the equation of the plane visible in the field of view is calculated along with the relative distance between camera and plane by using a set of derived spatial geometrical relations coupled with a few intrinsic properties of the system. The depth of an unknown object is then perceived by first extracting the object under observation using a series of image processing steps followed by exploiting the aforementioned mapping of pixel and real space coordinate. The performance of the algorithm is greatly enhanced by the introduction of reinforced learning making the system independent of hardware and environment. Furthermore the depth calculation function is modified with a supervised learning algorithm giving consistent improvement in results. Thus, the system uses the experience in past and optimizes the current run successively. Using the above procedure a series of experiments and trials are carried out to prove the concept and its efficacy.

This paper present an affine projection algorithm (APA) using grouping selection of input vectors. To improve the performance of conventional APA, the proposed algorithm adjusts the number of the input vectors using two procedures: grouping procedure and selection procedure. In grouping procedure, the some input vectors that have overlapping information for update is grouped using normalized inner product. Then, few input vectors that have enough information for for coefficient update is selected using steady-state mean square error (MSE) in selection procedure. Finally, the filter coefficients update using selected input vectors. The experimental results show that the proposed algorithm has small steady-state estimation errors comparing with the existing algorithms.

This paper aims at locating base station (BS) rationally to satisfy the most customs by using the least BSs. An ameliorated GA is proposed to search for the optimum solution. In the algorithm, we mesh the area to be planned according to least overlap length derived from coverage radius, bring into isometric grid encoding method to represent BS distribution as well as its number and develop select, crossover and mutation operators to serve our unique necessity. We also construct our comprehensive object function after synthesizing coverage ratio, overlap ratio, population and geographical conditions. Finally, after importing an electronic map of the area to be planned, a recommended strategy draft would be exported correspondingly. We eventually import HongKong, China to simulate and yield a satisfactory solution.

In this paper, an image resolution enhancement algorithm using dual-tree complex wavelet transform and neural network is proposed. The low resolution image is used as the input signal of the neural network and the network estimates all the 16 wavelet subbands of the corresponding high resolution image. The high resolution image is then obtained using the inverse dual-tree complex wavelet transform. Directional selectivity of the DT-CWT increases the accuracy of subband estimation done by the neural network. Subjective comparisons and objective quality assessments indicate additional improvement over the conventional bicubic and bilinear interpolation techniques and even new complex counterpart methods.

Segmentation is one of the fundamental issues of image processing and machine vision. It plays a prominent role in a variety of image processing applications. In this paper, one of the most important applications of image processing in MRI segmentation of pomegranate is explored. Pomegranate is a fruit with pharmacological properties such as being anti-viral and anti-cancer. Having a high quality product in hand would be critical factor in its marketing. The internal quality of the product is comprehensively important in the sorting process. The determination of qualitative features cannot be manually made. Therefore, the segmentation of the internal structures of the fruit needs to be performed as accurately as possible in presence of noise. Fuzzy c-means (FCM) algorithm is noise-sensitive and pixels with noise are classified inversely. As a solution, in this paper, the spatial FCM algorithm in pomegranate MR images' segmentation is proposed. The algorithm is performed with setting the spatial neighborhood information in FCM and modification of fuzzy membership function for each class. The segmentation algorithm results on the original and the corrupted Pomegranate MR images by Gaussian, Salt Pepper and Speckle noises show that the SFCM algorithm operates much more significantly than FCM algorithm. Also, after diverse steps of qualitative and quantitative analysis, we have concluded that the SFCM algorithm with 5×5 window size is better than the other windows.

The paper presents a new design method and power lines communication circuit consolidates their power and data in a single bus. This method is based on power electronic topology and can export new application circuit. Our-the bus's principle are briefly introduced. Buck-by considering the circuit transmission line, the characteristics of the steady-state circuit influence and dynamic switch characteristics are analyzed. The validity of this method is verified by the experiment results.

This paper analyzes the problems existing in the operation of the interference in the process of network communication electronic equipment. According to the typical problems in the electromagnetic compatibility and the anti-interference general network engineering practice, such as grounding technical problems, the application of the circuit separation technology, cable interference suppression ability, improve the quality of power supply, etc. The new technology and the application of new technology are introduced in this paper.

Content-based image retrieval using low-level features such as color, texture and shape is one of the major challenges in image processing and computer vision. Color as the most important low-level feature, has very wide applications in image retrieval systems. In this paper a new content-based image retrieval method using color feature of image regions is expressed. To this end, at the first step, images are partitioned to five fixed regions include center, top, bottom, left and right. Then the color autocorrelogram for each region is computed separately and kept as a feature vector to compare different images similarities. Because of the importance of the images center, weight of the center region is doubled when comparing similarity of images. For comparing other regions, difference of most similar regions is computed. This comparison makes the algorithm more invariant to rotation and to somehow changing the viewing angle, than the similar works. By combining the output of this method with the output of global color histogram-based retrieval method, system performance and accuracy of the results are more improved.

As we all know, the origin of modern logistics was in the United States, developed in Japan, became mature in Europe, and expanded in China. This is a historical development of the modern logistics recognized track. Due to China's economic and technological development, and with the construction of Shanghai International Shipping Center and Shanghai Yangshan International Deepwater development, China's modern logistics industry will attain a leap-forward development of a strong pace, and will also catch up with developed countries in the Western modern logistics level. In this paper, the author explores the flexibility of China's modern logistics management techniques to extend the use, and has certain practical and guidance significances.

An enhanced filtering method in Contourlet transform domain is proposed in this paper. According to local properties, SAR image will be firstly divided into homogeneous area, non-homogeneous area and edge area. Then average filtering method is applied to the homogeneous area, Contourlet transform de-noising method applied to the non-homogeneous area and edge area retained directly. It's proved that not only can we effectively eliminate the noise but also maintain edge details through this method. Compared with other classical filters, the capability to keep the edge details is stronger.

In this paper, we present a new approach to recognize Tibetan words by analyzing the structure and geometric features. The printed Tibetan words, including consonant and vowel symbols, are firstly generated by FontLister tool and through pre-processing, such as skeletonization and deburring. Then the words are splitted into the single characters, which can easily recognized by feature extraction and pattern matching. We adopt six features according to the geometry and stroke structure. Finally, the Tibetan words are automatically "understood" and converted into Latin alphabet transcription, which make more people read. The experiments show this approach has a recognization accuracy over 89%.

In wireless communication, channel equalization is one of the most challenging tasks because the broadcast channels are often subjected to multipath fading and several bandwidth limitations. Our project is mainly concerned with the Single Carrier Long Term Evolution (SC-LTE) uplink system model and its performance analysis when Fractionally Spaced Constant Modulus Algorithm (FS-CMA) and Recursive Least Squares Algorithm (RLS) are implemented. FS-CMA (CMA (1, 2) and CMA (2, 2)) and RLS algorithms are used individually to eliminate noise caused by multipath fading and Additive White Gaussian Noise (AWGN) in the stationary channel. The performance of conventional FS-CMA and RLS algorithm are compared using Bit Error Rate (BER) vs. Signal to Noise Ratio (SNR), Convergence, Equalizer output, Peak to Average Power Ratio (PAPR) vs. SNR.

In this paper, a new exemplar-based framework is presented, which treats image completion, texture synthesis and texture Analysis in a unified manner. In order to be able to avoid the occurrence of visually inconsistent results, we pose all of the image-editing tasks in the form of a discrete global optimization problem. The objective function of this problem is always well-defined, and corresponds to the energy of a discrete Markov Random Field (MRF). For efficiently optimizing this MRF, a novel optimization scheme, called Priority-BP, is then proposed, which carries two very important extensions over the standard Belief Propagation (BP) algorithm: "priority-based message scheduling" and "dynamic label pruning". These two extensions work in cooperation to deal with the intolerable computational cost of BP, which is caused by the huge number of labels associated with our MRF. In an Experimental results on a wide variety of input images are presented, which demonstrate the effectiveness of our image-completion framework for tasks such as object removal, texture synthesis, text removal and texture Analysis.

In order to find out the structural changes of land use / cover in Farm of Xinjiang Production and Construction Group(XPCG), Using the satellite remote sensing images of the 148th group in 2002 and 2007, Through the image processing information extracting, we have done the dynamic monitoring for the changes of land use, and done the quantitative analysis for the characteristics of land use changes in this period of 5 years. The writer also discusses the driving principle of land use changes from the natural factors and social economic factors. It can provide several scientific basises for government and the relevant departments in the process of industrialization and management of urban planning.

Over the last few decades, damage identification methods of civil and mechanical structures have been drawing much interest from various fields. This article presented a novel time-frequency based approach for detecting structural damage in pipeline. This is done by applying lamb wave technique and wavelet transforms for detection of change in the structural response data. Lamb waves were propagated through pipes and the received signals were subjected to wavelet transforms. The viability of the proposed approach is demonstrated using simulation examples. The pipes were modeled using finite -element method. This damage detection technique may serve the purpose of structural health monitoring because of long -distance inspection capability in situations where spatially distributed measurement of structural response is difficult.

This paper proposes an online competitive learning algorithm, briefly denoted as KACL (Kernel Averaging Competitive Learning), using kernel functions and quadtree structure for clustering analysis and remotely sensed image segmentation. Initially, KACL constructs a quadtree with a pre-specified scale from online input data and then locates all clusters by moving or self-splitting the nodes of the quadtree. The complexity of the quadtree is decided by the distribution of data. In the learning rule of KACL, we use the local means of vectors but not single vector so as to avoid the movement of learning prototypes among different clusters. We give the mathematical properties of KACL and present the proof of its convergence. KACL avoids the dead node problem and presetting of the number of clusters. Two experiments are separately carried out on Gaussian mixture data sets and remote sensing images and the results have shown good performance.

Personal identification technology as security systems is developing rapidly. Traditional authentication modes like key; password; card are not safe enough because they could be stolen or easily forgotten. Biometric as developed technology has been applied to a wide range of systems. According to different researchers, vein biometric is a good candidate among other biometric traits such as fingerprint, hand geometry, voice, DNA and etc for authentication systems. Vein authentication systems can be designed by different methodologies. All the methodologies consist of matching stage which is too important for final verification of the system. Neural Network is an effective methodology for matching and recognizing individuals in authentication systems. Therefore, this paper explains and implements the Neural Network methodology for finger vein authentication system. Neural Network is trained in Matlab to match the vein features of authentication system. The Network simulation shows the quality of matching as 95% which is a good performance for authentication system matching.

These days there are millions of digital images stored in image database. People will take a lot of time to find the right image for their query. Similarly, using keyword image retrieval may produce wrong result due to incorrect annotation. An image retrieval system is one of the tools that can be used for searching and retrieving images from a large database of digital images. In general, images features (color, texture, shape) are extracted to represent the images. This project focuses on color image features by applying global color histogram matching technique for HSV color space. The CBIR system is tested through query image of about 100 images from database. The retrieval obtained was good. From the result of precision and recall, we suggest some importing to consider for the future research.

The dynamic range of many natural scenes is far greater than the dynamic range of the imaging devices. These scenes present a challenge to the consumer digital cameras. The well-known technique to capture the full dynamic range of the scene is by fusing multiple images of the same scene. Usually people combine three or five different exposures to get the full dynamic range of the scene. Some cameras like Pentax K-7, always combines fixed exposures together to produce the output result. However, this should be adaptive to the scene characteristics. We propose an optimal solution for dynamically selecting the exposure sets.

In the recent years wavelet transform (WT) has been widely used for image compression. As WT is a sequential process, much time is required to transform data. Here a new approach has been presented where the transformation process is executed concurrently. As a result the procedure runs first and the time of transformation is reduced. Multiple threads are used for row and column transformation and the communication among threads has been managed effectively. Thus, the transformation time has been reduced significantly. The proposed system provides better compression ratio and PSNR value with lower time complexity.

Defect detection on industrial product using vision system currently is considered as necessity. Quality control of product requires several key factors. Consistency of detection is one of the crucial factors in quality control. Previous method of defect detection requires human assistance. Vision system is the alternative solution for the inconsistent human-based detection. This article discusses defect detection, defect feature extraction and defect classification of fabric product. The solution proposed is by using statistical filter for defect recognition. Extracted features are GLCM-based features, and the proposed 3D defect feature. Defect classification is carried out using Neural Network. The result shows a promising result towards classifying defect product.

A commercial software package FLUENT is used for numerical simulation of the distribution of the air flow and gas in goaf of the working face with downward ventilation. With the increase of coal seam inclination, the phenomenon of airflow stagnant appears first on the upper part of the goaf, then backflow starts and eddy forms a certain distance away from the working face. When the coal seam slope increases to a certain angle, the air leakage in working face entirely reverses and flows into the gob from the bottom part of the mining face, and returns to working face from the upper part, particularly the upper corner of the working face. As the coal seam angle increases, the natural ventilation pressure in goaf increases too. When the natural ventilation pressure is less than the ventilation pressure differential in working face, the bigger of coal seam angle, the smaller of leakage air volume, and the closer of high-concentration gas migrate from goaf to working face. When the natural ventilation pressure in goaf is greater than the ventilation pressure differential in working face, with the increase of coal seam angle, the natural ventilation pressure and the air leakage volume increases, the range of leakage air backflow increases, high-concentration methane in the goaf will move to deep and total amount of gas in goaf decreases.

This paper proposes an affine projection algorithm (APA) with selective input vectors, which based on the concept of selective-update in order to reduce estimation errors and computations. The algorithm consists of two procedures: input- vector-selection and state-decision. The input-vector-selection procedure determines the number of input vectors by checking with mean square error (MSE) whether the input vectors have enough information for update. The state-decision procedure determines the current state of the adaptive filter by using the state-decision criterion. As the adaptive filter is in transient state, the algorithm updates the filter coefficients with the selected input vectors. On the other hand, as soon as the adaptive filter reaches the steady state, the update procedure is not performed. Through these two procedures, the proposed algorithm achieves small steady-state estimation errors, low computational complexity and low update complexity for colored input signals.

In this paper, a novel algorithm is proposed for fingertips detection in view of two-handed static hand pose recognition. In our method, finger tips of both hands are detected after detecting hand regions by skin color-based segmentation. At first, the face is removed in the image by using Haar classifier and subsequently, the regions corresponding to the gesturing hands are isolated by a region labeling technique. Next, the key geometric features characterizing gesturing hands are extracted for two hands. Finally, for all possible/allowable finger movements, a probabilistic model is developed for pose recognition. Proposed method can be employed in a variety of applications like sign language recognition and human-robot-interactions etc.

Wavelet analysis and its application has found much attention in recent times. It is vastly applied in many applications such as involving transient signal analysis, image processing, signal processing and data compression. It has gained popularity because of its multiresolution, subband coding and feature extraction features. The paper describes efficient application of wavelet analysis for image compression, exploring Daubechies wavelet as the basis function. Wavelets have scaling properties. They are localized in time and frequency. Wavelets separate the image into different scales on the basis of frequency content. The resulting compressed image can then be easily stored or transmitted saving crucial communication bandwidth. Wavelet analysis because of its high quality compression is one of the feature blocks in the new JPEG2000 image compression standard. The paper proposes Daubechies wavelet analysis, quantization and Huffman encoding scheme which results in high compression and good quality reconstruction.

We propose an algorithm that can be used by amateur astronomers to analyze the images acquired during solar eclipses. The proposed algorithm analyzes the image, detects the eclipse and produces results for parameters like magnitude of eclipse, eclipse obscuration and the approximate distance between the Earth and the Moon.

Computerized lip reading, or speech reading, is concerned with the difficult task of converting a video signal of a speaking person to written text. It has several applications like teaching deaf and dumb to speak and communicate effectively with the other people, its crime fighting potential and invariance to acoustic environment. We convert the video of the subject speaking vowels into images and then images are further selected manually for processing. However, several factors like fast speech, bad pronunciation, and poor illumination, movement of face, moustaches and beards make lip reading difficult. Contour tracking methods and Template matching are used for the extraction of lips from the face. K Nearest Neighbor algorithm is then used to classify the 'speaking' images and the 'silent' images. The sequence of images is then transformed into segments of utterances. Feature vector is calculated on each frame for all the segments and is stored in the database with properly labeled class. Character recognition is performed using modified KNN algorithm which assigns more weight to nearer neighbors. This paper reports the recognition of vowels using KNN algorithms

In this paper, we propose a robust blind watermarking algorithm based on quantization of distance between wavelet coefficients for copyright protection. We divide wavelet coefficients into some blocks and obtain the first, second, and third maximum coefficients in each block. Then we do quantization between the first, and second maximum coefficients according to binary watermark bits, which effectively makes watermark to be resist against attacks. Also, we need neither the original image nor watermark to extract the embedded watermark by using the block-based watermarking method. The algorithm as a watermarking system has appropriate performance duo to imperceptibility. In addition, experimental results also show that the proposed method is quite robust under either non-geometry or geometry attacks.

Growing research in computer vision obliges reliable and efficient target tracking methods. This letter proposes a review on target estimation and tracking algorithms grouped into two namely deterministic and nondeterministic based algorithms. Various issues affecting success and/or failure of such algorithms are discussed as well as presenting current references (fifty five in total) on the topic. In conclusion, no single algorithm can be ideal for all tracking problems. It is hoped that with only two groups of tracking and estimation algorithms, this work will guide tracking system designers on choosing a suitable algorithms for use.

This article shows a new method to detect blood vessels in the retina by digital images. Retinal vessel segmentation is important for detection of side effect of diabetic disease, because diabetes can form new capillaries which are very brittle. The research has been done in two phases: preprocessing and processing. Preprocessing phase consists to apply a new filter that produces a suitable output. It shows vessels in dark color on white background and make a good difference between vessels and background. The complexity is very low and extra images are eliminated. The second phase is processing and used the method is called Bayesian. It is a built-in in supervision classification method. This method uses of mean and variance of intensity of pixels for calculate of probability. Finally Pixels of image are divided into two classes: vessels and background. Used images are related to the DRIVE database. After performing this operation, the calculation gives 95 percent of efficiency average. The method also was performed from an external sample DRIVE database which has retinopathy, and perfect result was obtained

In this paper, a novel compression algorithm for Chinese document images is proposed. Initially, documents are segmented into readable components such as characters and punctuation marks. Similar patterns within the text are found by shape context matching and grouped to form a set of prototype symbols. Text redundancies can be removed by replacing repeated symbols by their corresponding prototype symbols. To keep the compression visually lossless, we use a multi-stage symbol clustering procedure to group similar symbols and to ensure that there is no visible error in the decompressed image. In the encoding phase, the resulting data streams are encoded by adaptive arithmetic coding. Our results show that the average compression ratio is better than the international standard JBIG2 and the compressed form of a document image is suitable for a content-based keyword searching operation.

Image segmentation is an important and difficult step for automatic rock particle size distribution estimation. In this paper, we propose a method for the segmentation rock images in a machine vision system using the Voronoi diagram. Typically, rock edge detection is achieved using the watershed transform. Marker-driven watershed segmentation extracts seeds indicating the presence of rocks at specific image locations. The marker locations are then set to be regional minima within the topological surface, which is normally the gradient or thresholds of the original input image. In contrast, our approach uses the generalized Voronoi diagram through Euclidian distance transform for the segmentation of the rock image after extracting the markers. In order to avoid the difficult step of segmentation, we also investigate an image-classification system for rock particulate size estimation using two-level wavelet decomposition. Experimental results show that using the Voronoi diagram is not only more robust than watershed for rock particulate size estimation but also less computationally complex. The Voronoi diagram and watershed methods are superior to the image classification method.

Direct volume rendering has been received much attention since it need not to extract geometric primitives for visualization and its performance is generally better than surface rendering. Transfer functions, which are used for mapping scalar field to optical properties, are of vital importance in obtaining a sensible rendering result from volume data. Though traditional color transfer functions are in RGB color space, HSL color space that conveys semantic meanings is more intuitive and user-friendly. In this paper, we present a novel approach aims to emphasize and distinguish strong boundaries between different materials. We achieve it by using data value, gradient magnitude and dimension of the volumetric data to set opacity. Then, through a linear map from data value, gradient magnitude and second derivative to hue, saturation and lightness respectively, a color transfer function is obtained in HSL color space. Experimental tests on real-world datasets indicate that our method could achieve desirable rendering results with revealing important boundaries between different structures and indicating data value's distribution in the volume by using different colors.

A method used for improving the performance of the digital watermarking based on the modification of image pixels is proposed in this paper. In the embedding process, the watermark bits are partially embedded into the blue component of the image pixels. The criteria used to determine whether a watermark bit embedding is performed or not is derived from the variance of neighbor pixels of the embedding pixel. The same criteria are used in the watermark extraction process to identity positions of the watermarked pixels. The experimental result show the improved performance in term of NC, compared to the previous methods. The improved robustness of the proposed method against various common imaging processing based attacks e.g. additive Gaussian noise, JPEG compression standard and image cropping at various strengths is also presented and compared.

This paper presents preliminary development of the LFR which use a normal laptop with a web cam to guide itself through a track. The designed LFR is insensitive to environmental factors such as darkness, lighting, camera distortion or line color. This development is based on computer vision enhancement using digital image processing. It is accomplished through the following stages: Firstly, the acquired RGB image using the web cam is converted to another color coordinates for testing and comparing to choose the best color space. After that, the image contrast is enhanced using adaptive histogram equalization, and then the image is filtered using wiener filter. Finally morphological operations are carried out. The results are evaluated qualitatively and quantitatively from the points of peak signal-to-noise ratio (PSNR), entropy, and image smoothness. The results show that the proposed method is promising for vision enhancement of LFR.

In this paper, an adaptive approach for bilateral filtering, based on estimation of the range filtering parameter using homogeneity measurements is introduced for image denoising. For each pixel, the homogeneity measurement is performed utilizing its local neighbors considering different directions, and the range parameter is estimated from the variance of the most homogeneous blocks. In order to reduce the residual noise, the filtering operation is carried out in an iterative manner. A stopping criterion using the structural similarity index is provided. Experiments are carried out using various standard images and the results show that the proposed method is more effective in reducing additive white Gaussian noise as compared to several recently introduced denoising techniques in terms of the peak signal-to-noise ratio and structural similarity index with a reduced computational requirement.

This paper presents a fire flame detection using CCTV cameras based on image processing. The scheme relies on color segmentation and space-time analysis. The segmentation is performed to extract fire-like-color regions in an image. Many methods are benchmarked against each other to find the best for practical CCTV camera. After that, the space-time analysis is used to recognized fire behavior. A space-time window is generated from contour of the threshold image. Feature extraction is done in Fourier domain of the window. Neural network is used for behavior recognition. The system will be shown to be practical and robust.

Accurate attenuation correction of emission data is mandatory for quantitative analysis of PET images. One of the main concerns in CT-based attenuation correction(CTAC) of PET data in multimodality PET/CT imaging is misalignment between PET and CT images. The aim of this study, is to proposed a hybrid method which is simple, fast and accurate, for registration of PET and CT data which affected from respiratory motion in order to improve the quality of CTAC. The algorithm is composed of three methods: First, using B-spline Free Form Deformation to describe both images and deformation field. Then applying a pre-filtering on both PET and CT images before segmentation of structures in order to reduce the respiratory related attenuation correction artifacts of PET emission data. In this approach, B-spline using FFD provide more accurate adaptive transformation to align the images, and structure constraints obtained from prefiltering applied to guide the algorithm to be more fast and accurate. Also it helps to reduce the radiation dose in PET/CT by avoiding repetition of CT imaging. These advances increase the potential of the method for routine clinical application.