Current Issue : January - March Volume : 2014 Issue Number : 1 Articles : 6 Articles
Over the years, maritime surveillance has become increasingly important due to the recurrence of piracy. While\r\nsurveillance has traditionally been a manual task using crew members in lookout positions on parts of the ship, much\r\nwork is being done to automate this task using digital cameras coupled with a computer that uses image processing\r\ntechniques that intelligently track object in the maritime environment. One such technique is level set segmentation\r\nwhich evolves a contour to objects of interest in a given image. This method works well but gives incorrect\r\nsegmentation results when a target object is corrupted in the image. This paper explores the possibility of factoring in\r\nprior knowledge of a ship�s shape into level set segmentation to improve results, a concept that is unaddressed in\r\nmaritime surveillance problem. It is shown that the developed video tracking system outperforms level set-based\r\nsystems that do not use prior shape knowledge, working well even where these systems fail....
We present a simple yet highly efficient method to register range and color images. This method does not rely\r\nupon calibration parameters nor does it use visual features analysis. Our assumption is that if the transformation\r\nthat registers the images is a mathematical function, we can approximate with little number of samples. To this\r\nend, thin-plate spline-based interpolations are used in this paper. Therefore, the registration of one point in our\r\nmethod takes only the solving of a nonlinear function. Drastically enhanced performances in the computational\r\nprocessing are attained under this condition. In fact, we show that ultimately our computational algorithm is\r\nindependent of the complexity of the mathematical model underlying it. Finally, this paper reports on the results of\r\nexperiments conducted with various range camera models that endorse the proposed method. Eventually, three\r\nkey features can be derived from our method: practicality, accuracy, and wide applicability....
In large-scale multimedia event detection, complex target events are extracted from a large set of\r\nconsumer-generated web videos taken in unconstrained environments. We devised a multimedia event detection\r\nmethod based on Gaussian mixture model (GMM) supervectors and support vector machines. A GMM supervector\r\nconsists of the parameters of a GMM for the distribution of low-level features extracted from a video clip. A GMM is\r\nregarded as an extension of the bag-of-words framework to a probabilistic framework, and thus, it can be expected to\r\nbe robust against the data insufficiency problem. We also propose a camera motion cancelled feature, which is a\r\nspatio-temporal feature robust against camera motions found in consumer-generated web videos. By combining\r\nthese methods with the existing features, we aim to construct a high-performance event detection system. The\r\neffectiveness of our method is evaluated using TRECVID MED task benchmark....
A lossy compression method is one where compressing data and then decompressing it retrieves data that may well be different from the original, but is close enough to be useful in some way. Lossy compression is most commonly used to compress multimedia data (audio, video, still images), especially in applications such as streaming media and internet telephony. In lossless compression schemes, the reconstructed image, after compression, is numerically identical to the original image. However lossless compression can only achieve a modest amount of compression. Lossless compression is preferred for archival purposes and often medical imaging, technical drawings, clip art or comics. This paper outlines the comparison of compression methods such as JEPG, JEPG 2000 with SPIHT encoding on the basis of compression ratio and compression quality. The comparison of these compression methods are classified according to different medical images like MRI and CT. For JPEG based image compression RLE and Huffman encoding techniques are used by varying the bits per pixel. For JPEG 2000 based image compression SPIHT encoding method is used. The DCT and DWT methods are compared by varying bits per pixel and measured the performance parameters of MSE, PSNR and compression ratio. In JPEG 2000 method, compared the different wavelets like Haar, CDF 9/7, CDF 5/3 etc and evaluated the compression ratio and compression quality. Also varied the decomposition levels of wavelet transform with different images....
This paper presents recommended techniques for choosing video sequences for subjective experiments. Subjective\r\nvideo quality assessment is a well-understood field, yet scene selection is often driven by convenience or content\r\navailability. Three-dimensional testing is a newer field that requires new considerations for scene selection. The\r\nimpact of experiment design on best practices for scene selection will also be considered. A semi-automatic\r\nselection process for content sets for subjective experiments will be proposed....
We propose a person verification method using behavioral patterns of human upper body motion. Behavioral\r\npatterns are represented by three-dimensional features obtained from a time-of-flight camera. We take a statistical\r\napproach to model the behavioral patterns using Gaussian mixture models (GMM) and support vector machines. We\r\nemploy the maximum likelihood linear regression adaptation method to estimate GMM parameters with a limited\r\namount of data. Experimental results show that it reduced by 28.6% the relative equal error rates from a system using\r\nthe maximum likelihood estimation with 25 samples per subject. We also demonstrate that the proposed approach is\r\nrobust against variations in body motion over time....
Loading....