Earticle

This paper investigates the performance of image transmission with unequal error protection (UEP) schemes based on irregular Low-Density Parity-Check (LDPC) codes. Firstly, the UEP is achieved by mapping different bits of the image bytes to different positions of LDPC codes, i.e. the more important bits in the image are mapped to the variable node with higher degrees in irregular LDPC codes, after the channel coding, we map the information bits into a power efficient QAM constellation and map the parity check bits into a spectral efficiency 16QAM constellation. Simulation results show that the UEP scheme is effective.

Due to growing demand for wireless broadband communication and congestion on bandwidth of RF spectrum, optical wireless has received considerable attention in the communication world. Optical wireless or free space optics is a good alternative solution to existing RF communication system. The quality of FSO transmission is characterized by BER. Performance degradation faced by the FSO system can be mitigated by using spatial diversity techniques. This work focuses on performance analysis of 4X4 multiple transmitter/receiver combination which is integrated with different types of amplifiers. By using the feasible system parameters for 4X4 multiple transceiver FSO system integrated with preamplification, maximum range of 425 km can be achieved for acceptable received power under clear weather condition.

Image holistic scene understanding based on image intrinsic characteristics and conditional random fields is proposed. The model integrates image scene classification, image semantic segmentation and object detection. 1) For the scene classification, we use method of PHOW feature extraction plus KPCA dimensional reduction to obtain feature information for each image. 2) For object detection section, saliency detection and segmentation characteristics of the image object detection is useful. We propose the method by integrating image segmentation information got by the method proposed in literature [1]. 3) For the semantic segmentation: (1) For the unary potentials, we incorporating HOG, RGB color histogram and LBP features by the methods proposed in literature [2]; (2) The image manifold structural features can better reflect the importance between hyper-pixel regions and eventually boost accuracy. Therefore, we add the higher-order potential item to reflect inherent manifold image feature of each super pixel region. The experiments testify that model performance has raised on all three sub-tasks.

Focal intention of this paper is to emphasize the appliance of the basic image statics for image restoration, de-blurring, de-noising, enhancement, edge detection, edge sharpening, finding edge position and many more root level appliances of image processing and computer vision. For the evaluation and study of restoration, de-blurring and de-noising, some noise molds are discussed and then for the estimation of the results of statistics, some amount of different types of noise have been added to the image and then the process of filtration is performed for scrutiny the effect. For scrutiny of enhancement some de-enhanced or low contrast image are used. Some image quality measures i.e., Mean Square Error, Peak Signal to Noise Ratio Discussion and conclusions drawn from the experimental results are the comparative study of these appliances of image processing. This paper provide the detail study of selected appliances of image processing and computer vision experimentally, evaluate the performance, compare result on literature and give trend what can be done for new and better loom.

In order to obtain rapid fusion speed, an image fusion algorithm based on Non-subsampled Shearlet Transform (NSST) and Compressed Sensing (CS) is presented. The source images are decomposed with NSST. Based on local area energy, the low-frequency coefficients are fused. The high-frequency coefficients are compressed, fused and reconstructed with CS. Based on global gradient, the measurements of high-frequency coefficients are fused. The inverse NSST is used to get the final fused image. During the fusion course, only the compressed data of the high-frequency coefficients are fused, so the fusion effects can’t be affected. At the same time, the running time can be reduced. In this paper, the multi-focus images are used to verify the algorithm effectiveness. The simulation results indicate that the fusion image can be achieved without prior knowledge of the original images. Although the fusion quality is sacrificed when the sampling rate becomes higher, the algorithm can significantly reduce the time cost and hardware requirements. The algorithm provides an idea on how to satisfy the real time requirements in the fusion system.

Digital images are inseparable part of our life. Images are used at various places like medical imaging, crime scene investigation, forensic analysis, courts etc. but due to ubiquitous accessibility of image editing software, images are no longer trusted. Digital images are losing their credibility. For checking authenticity of digital images forgery detection methods are required. One of the most frequent image forgery is copy-move. In this forgery, a region of the original image is used for producing a manipulated image by performing post-processing operations over copied segment before pasting it to original image. The main principle of finding such type of forgery is based on finding resemblance present in different segments of image. Image is divided in blocks then feature vectors are extracted corresponding to different blocks of image. Sorting techniques are applied to find similarity between blocks. In case of natural images which may have similar regions, shift vectors are calculated to decrease false matches. Several methods are suggested by researchers to detect such type of forgery. In this paper, a survey on frequency-based methods is presented for detecting copy-move forgery in images.

In order to resolve shadow interference and slow image processing speed in the visual navigation of unmanned vehicles on city roads, a new lane detection algorithm based on the Inverse Perspective Mapping(IPM) of vertical sub-image reconstruction using local bands was proposed. The lane’s IPM aerial image was obtained using the projective transformation of three-dimensional images of city roads. The ROI portion of the IPM map was decomposed and analyzed using the sym3 wavelet, and after analyzing and comparing the experimental results the first and second levels of the vertical sub images were selected for the reconstruction, compression, and removal of shadows from the original image. The Canny algorithm was proposed and adopted in order to extract the edge features of the reconstructed images according to real-time road image quality. A modified, polar angle, constraint-based, fast Hough transform was used to locate the candidate lanes. Finally, the main control point, straight-line-fitting algorithm was used to fit the final lanes, which achieved the precise location and recognition of the lane lines. The California Polytechnic lane data set, a public data platform, which is now widely used in road visual recognition fields around the world, was selected for the testing and verification of the algorithm. The results of the experiment and actual operation indicated that, despite having a storage space of approximately one tenth the size of similar algorithms, this detection algorithm effectively solves the issue of shadow removal, meets the real-time requirements of roadway image processing systems for unmanned vehicles with a recognition time of less than 20ms, and is robust.

The paper introduces the Keren registration method and points out its disadvantage which means it will become inaccuracy on the large scale parameters. To reduce the error on large scale parameters of Keren registration, a two step method is proposed, which the phase correlation algorithm is used to estimate the large translation and rotation angle roughly and the improved Keren algorithm is used to estimate accurately the small translation and rotation angle. The experimental results show that the two step method makes less absolute error of angle than Keren method in the situation of large translation and rotation angle. A new method of estimating the standard deviation of noise is introduced to the robust certainty function, which reduces the impact of noise in the process of interpolation using normalized convolution algorithm. By the edge detection of fusion image in the first stage of the interpolation process of normalized convolution algorithm, a calculation method of long axis and short axis of the structure self-adaptive function is improved. The experimental results show that the proposed interpolation method can improve the performance of the original algorithm and enhance the effect of image super-resolution reconstruction.

In advanced image processing, aerial image processing plays an important role in object extraction such as building extraction, road detection etc. The aerial images captured are usually bound to suffer from Gaussian noise, salt and pepper noise, speckle noise etc. Therefore obtaining of aerial image with high accuracy is very difficult task. A flawless aerial image is inevitable for further object extraction process. There are a number of filtering techniques to detach the noise for preserving the integrity of captured aerial image. In this paper we have applied mean filter, median filter, wiener filter, wavelet transform and curvelet transform for removal of various level of Gaussian noise, salt and pepper noise and speckle noise added separately in an aerial image. The performance of both the transforms and filtering methods are compared in terms of Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE).

In this paper we present an image segmentation algorithm based on the gradient histogram threshold image improvement. And the MATLAB simulation result shows that: each kind of image segmentation algorithm has its own advantages, disadvantages and scope. Therefore, it is necessary to judge and analysis the image first before we do image segmentation, then we can choose an appropriate segmentation algorithm and it makes the result of the image segmentation satisfactory.

The ant colony algorithm is an algorithm which is used to find the optimal path. As a kind of bionic evolutionary algorithm, the ant colony algorithm is inspired by the real ant colony foraging mechanisms. Firstly, this paper introduces the basic model of the ant colony algorithm. Then, aiming at the shortcomings of the ant colony algorithm, we propose a new probability formula of the optimal path and the new formula of the pheromone update. In addition, we combine the traditional ant colony algorithm with the local search algorithm and propose the improved ant colony algorithm. It is the LSACA algorithm. In the experimental analysis, we set and analyze the parameters of the algorithm. Then, we compare with the traditional algorithm to prove the feasibility and the effectiveness of the algorithm.

Since the ant colony algorithm is proposed, it has achieved the remarkable achievements in many fields. With the development of the times, the traditional ant colony algorithm exposes its limitations for solving the questions. In this paper, we improve the ant colony algorithm. And we combine the ant colony algorithm with the genetic algorithm. Then, we propose the GAPSPAC algorithm. The algorithm combines the advantages of the genetic algorithm and the ant colony algorithm. And it overcomes the disadvantages to improve the efficiency of solving the questions. In the last experiment, we can see the algorithm has the better problem solving ability and the stability.

The aim of this paper is to develop an efficient and a powerful Matlab based graphical user interface to address the problem of image segmentation. We propose two approaches for segmenting images: An automatic marker controlled watershed segmentation for segmenting an entire image or a scene and a semiautomatic graph cut based segmentation using fixation points. Automatic Watershed segmentation with a Sobel edge detector is used to detect the gradient of an input image resulting in an image less sensitive to noise. To deal with the usual problem of over segmentation using watershed, marker controlled watershed transformation is applied further for segmenting an image. Fixation based graph cut segmentation allows the user to analyze the input image displayed on the screen and specify some hard constraints indicating the object of interest or target object by using the mouse interaction. Experiments are done on the publically available dataset and the results of the supervised evaluation methods are observed to be satisfactory and are demonstrated along with the manually segmented reference image or a ground truth image obtained from segmentation evaluation database

Fuzzy clustering is a popular unsupervised learning method used in cluster analysis which allows a point in large data sets belongs to two or more clusters. Prior work suggests that Particle Swarm Optimization based approach could be a powerful tool for solving clustering problems. In this paper, we propose a data clustering algorithm based on modified adaptive particle swarm optimization. We choose to use artificial bee colony algorithm combined with PSO technique to modify the traditional clustering methods due to its fast convergence and the presence of adaptive mechanisms based on the evolutionary factor. On the one hand, Particle Swarm Optimization is proven to be an effective and robust technique for fuzzy clustering. On the other hand, the artificial bee colony algorithm has the capability to generate diversity within the swarm when the guide bees are in the exploration mode. Through numerical analysis and experimental simulation, we verify that our algorithm performs much better compared with other state-of-the-art algorithms. Future research schedule is also discussed in the final part.

The vertical or horizontal edges don’t dominate in some frames of video sequence, so the conventional discrete cosine transform (DCT) may not be the best choice for those frames. Directional DCT framework behaves better than conventional DCT in coding performance for images where directional edges dominate. In addition, the all phase biorthogonal transform (APBT), which is used in image compression instead of DCT, can also help to improve the performance of compression. In the light of directional DCT and APBT, directional APBT (D- APBT) is proposed and applied to H.263 video coding. Experimental results show that this framework can indeed improve the coding performance remarkably.

Many of the standard compression methods are based, at their lowest level, in coding digital words of variable length or VLC, Huffman type, designed in 1952 or in any of its versions, as canonical. This article presents an VLC with greater efficiency than Huffman encoding, named as “Leveling”, which uses two variants, “Leveled Reordering” for low redundancy, e.g. for text and, “Segmented Leveling” for middle and high redundancy, for image processing. Leveling, developed by Javier Joglar in 1995, uses the concepts of “meaning” and “ordering” of the VLC codes generated, to get optimum performance in terms of “compression ratio”, higher than any other non-adaptive VLC.

In this paper, a fast bounded parametric margin V -support vector machine (BP-V- SVM) for classification is proposed. Different from the parametric margin V -support vector machine (par-V -SVM), the BP-V -SVM maximizes a bounded parametric margin, and consequently the successive overrelaxation (SOR) technique could be used to solve our dual problem as opposed solving the standard quadratic programming problem (QPP) in par-V -SVM. Numerical experiments on several benchmark data sets and NDC data sets demonstrate the feasibility and effectiveness of the proposed algorithm.

In order to improve the convergence and diversity of multiobjective optimization algorithms, the harmonic average distance is employed to improve the aggregating function combined L-rank value. Selection model and searching scheme of artificial bee colony algorithm and diversity maintaining scheme are improved in this paper. This novel many objectives optimization method based on improved artificial bee colony algorithm (ABC) in this paper is compared with other three many objectives optimization methods on 3 to 8 objectives DTLZ. Simulation results show that the proposed algorithm is superior to other algorithms in the diversity and convergence of solutions.

In this paper, modeling and simulative analysis of a Free Space Optical (FSO) communication link has been performed using different system parameters. Analysis has been performed by observing the Q Factor and SNR of the received signal under different data transmission rates, link length, divergence angle, and transmitted power levels of the information signal using OPTISYSTEM simulation software. It has been observed from the results that as the FSO link distance increases, the Q Factor of the received signal decreases and thus BER value increases. Also, as the data transmission rate of the system in increased, the Q Factor of received signal decreases. On increasing the value of divergence angle, the Q Factor of received signal decreases and thus degrades the performance of the system.

A parallel FM signal demodulation system is designed in this paper. All FM signals from 90MHz to 110MHz could be captured, separated and restored into original information. The method of digital channelized receiving is used in the system to divide the broadband signals into independent ones according to their center frequency. This design could be used in the field of electron reconnaissance and multichannel communication system. We realized the design on a software radio platform based on the structure of SCA (Software Communication Architecture), and the advantages of this ideal is discussed in the paper.

Uyghur is one of the most populous and civilized groups with Turkic ethnicity and mainly located Xinjiang Uyghur Autonomous Region of China. Uyghur language belongs to the Karluk branch of the Turkic language family in Altaic language system, and holds agglutinative characteristics in morphological structure. Named Entity Recognition (NER) is an Information Extraction task that has become an essential part of Natural Language Processing (NLP) tasks, such as Machine Translation and Information Retrieval. In this paper, as a subtask of NER, the importance of Uyghur Named Entity Recognition (UPNR) task is demonstrated, the main characteristics of the Uyghur language are highlighted, and the aspects of standardization in annotating named entities are illustrated. Moreover, the approaches used in Uyghur NPNR field are explained and the features of common tools used in Uyghur NPNR are described. A brief review of the state of the art of Uyghur NPNR research is discussed, too. Finally, we present our conclusions. Throughout the presentation, illustrative examples are used for clarification.

Image retargeting aims to change the resolution and aspect ratio of an image to fit it into different devices. The key and most challenging issue for this task is how to balance the tradeoff between preserving the important contents and minimizing the visual distortions. In this paper, we present an effective image retargeting method called concepts-based mesh parameterization. Specifically, we first construct a structure-preserved mesh image. Based on this mesh representation, the image retargeting problem is formulated into a constrained image mesh parameterization problem, finding a homomorphous target mesh with the desired device size. To emphasize different concepts and minimize visual distortion, different concepts are first detected from the image, and are associated into the image mesh. Then the problem is transformed into solving a linear equations system. The target image is rendered by texture mapping. Experiments demonstrates the effectiveness of our concept-based algorithm.

The goal of pan-sharpening is to increase both spatial and spectral resolution of multispectral images. Intensity-hue-saturation (IHS) is one of widely efficient image fusion methods used in recent years. The drawback of IHS is spectral distortion in its results which can be improved by use of wavelet decomposition in IHS-based pan-sharpening methods. Employing Wavelet transforms enhances the resolution of Multispectral (MS) images while maintaining the spectral properties. This paper presents an adaptive IHS-based fusion using "à trous" wavelet (ATW) decomposition based on injecting weighted high frequency components of high spatial panchromatic (PAN) image obtained through à trous decomposition into resampled version of the MS images. Furthermore, the parameters used in the proposed algorithm are optimized through the genetic and Teaching-Learning algorithms. Finally, the proposed method is evaluated using the IKONOS and Landsat ETM+ images and compared to the other conventional methods to confirm its superiority.

As a kind of critical information in images and computer vision tasks, contour has a general application due to the robustness to illumination. However, it is far from meeting the current requirement because of expensive computation and noise in the real-world application. This paper proposes a novel approach of extracting contours in images. Firstly, we can obtain the salient points on the target contour by use of the accuracy property of Chord-to-point distance under combination of an ellipse model method. Then, a hyperbolic curve segment is used to fit the salient points, which can represent the target contour. The extensive experiments show that our method has better robustness and gives more exact approximation to the original target contour. Our work provides more selections for the practical application of contour completion.

Landslide deformation signal processing is significant for landslide stability analysis. Global Position System (GPS) control networks were built to monitor landslide deformation and acquire landslide displacement time series. It was difficult to predict landslide displacement because of the highly non-linear and non-stationary characteristics contained in displacement time series. A Wavelet Analysis - Radial Basis Function Neural Network (WA-RBFNN) model was proposed to overcome this problem. Firstly, monthly cumulative displacement time series was decomposed into different frequency components using wavelet analysis. Then a RNFNN model was established to forecast each frequency component values. The final prediction results were obtained through the sum of the predictive values of each frequency component. GPS monitoring points ZG325 and ZG326 on Baijiabao landslide in the Three Gorges Reservoir Area were used as study cases. A single RBFNN model was also built as comparison. The experimental results show that GPS control network can monitor landslide deformation accurately and the WA-RBFNN model is of high prediction accuracy. What is more, WA-RBFNN model has better prediction effect than a single RBFNN model.

The Active Shape Model (ASM) is one of the most popular local texture models for face alignment. It applies in many fields such as locating facial features in the image to classify or make measurements, face synthesis etc. This paper has proposed some improvements on the classical ASM to increase the performance of the model in the face alignment application. Here, Only 15 ‘land marking’ points are used as a parameter to generate an example shape . In this paper it has been shown several practical examples where we have manually built such example shapes and used them for further processing.