Current Issue : July - September Volume : 2015 Issue Number : 3 Articles : 5 Articles
Our research focuses on the question of classifiers that are capable of processing images rapidly and accurately\nwithout having to rely on a large-scale dataset, thus presenting a robust classification framework for both facial\nexpression recognition (FER) and object recognition. The framework is based on support vector machines (SVMs) and\nemploys three key approaches to enhance its robustness. First, it uses the perturbed subspace method (PSM) to\nextend the range of sample space for task sample training, which is an effective way to improve the robustness of a\ntraining system. Second, the framework adopts Speeded Up Robust Features (SURF) as features, which is more\nsuitable for dealing with real-time situations. Third, it introduces region attributes to evaluate and revise the\nclassification results based on SVMs. In this way, the classifying ability of SVMs can be improved.\nCombining these approaches, the proposed method has the following beneficial contributions. First, the efficiency of\nSVMs can be improved. Experiments show that the proposed approach is capable of reducing the number of samples\neffectively, resulting in an obvious reduction in training time. Second, the recognition accuracy is comparable to that\nof state-of-the-art algorithms. Third, its versatility is excellent, allowing it to be applied not only to object recognition\nbut also FER....
In recent day cybercrime is too easy to execute, viewing some last decade we can see image ownership authentication has drawn a sharp attention due to easy availability of the internet and inexpensive digital recording. Moreover storage peripherals have created an environment where duplication and misdistribution of the digital content has become easier that leads cybercrime. In this paper, a method for embedding an invisible digital watermark on to host image using alpha channel is described. Simulation results show the effectiveness and robustness of proposed method. In addition here we are going to compress the watermark using compressive sensing to create space by which we can get bandwidth efficiency and rapid transmission....
This paper presents an approach to classify static foreground blobs in surveillance scenarios. Possible application is\nthe detection of abandoned and removed objects. In order to classify the blobs, we developed two novel features\nbased on the assumption that the neighborhood of a removed object is fairly continuous. In other words, there is a\ncontinuity, in the input frame, ranging from inside the corresponding blob contour to its surrounding region.\nConversely, it is usual to find a discontinuity, i.e., edges, surrounding an abandoned object. We combined the two\nfeatures to provide a reliable classification. In the first feature, we use several local histograms as a measure of similarity\ninstead of previous attempts that used a single one. In the second, we developed an innovative method to quantify\nthe ratio of the blob contour that corresponds to actual edges in the input image. A representative set of experiments\nshows that the proposed approach can outperform other equivalent techniques published recently....
In asymmetric stereoscopic video compression,\nthe views are coded with different qualities. According to the\nbinocular suppression theory, the perceived quality is closer\nto that of the higher-fidelity view. Hence, a higher compression\nratio is potentially achieved through asymmetric coding.\nFurthermore, when mixed-resolution coding is applied,\nthe complexity of the coding and decoding is reduced. In\nthis paper, we study whether asymmetric stereoscopic video\ncoding achieves the mentioned claimed benefits. Two sets\nof systematic subjective quality evaluation experiments are\npresented in the paper. In the first set of the experiments, we\nanalyze the extent of downsampling for the lower-resolution\nview in mixed-resolution stereoscopic videos. We show that\nthe lower-resolution view becomes dominant in the subjective\nquality rating at a certain downsampling ratio, and this\nis dependent on the sequence, the angular resolution, and\nthe angular width. In the second set of the experiments,\nwe compare symmetric stereoscopic video coding, qualityasymmetric\nstereoscopic video coding, and mixed-resolution\ncoding subjectively. We show that in many cases, mixedresolution\ncoding achieves a similar subjective quality to that of symmetric stereoscopic video coding, while the computational\ncomplexity is significantly reduced....
There are many methods for detection of point-like features in gray-leveled bitmap images. The problem of defining a\nthreshold for acceptance or rejection of the results is usually neglected or left to experts. In this paper, a novel method\nof estimating suboptimal detection threshold values is proposed. It is based on overlapping the results of two or three\ndifferent methods parametrized with respective thresholds. The quality functions (of two or three variables), whose\nglobal extrema (maximum) approximately correspond to the suboptimal levels of thresholds for the used methods,\nare defined. This method was applied to a series of the bitmaps generated by a radar sensor and by simulated bitmaps....
Loading....