We consider here a change detection problem: to find regions of change on a test image with respect to a reference image. Unlike the state-of-the-art change detection and background subtraction algorithms that compute...
详细信息
ISBN:
(纸本)9781424442195
We consider here a change detection problem: to find regions of change on a test image with respect to a reference image. Unlike the state-of-the-art change detection and background subtraction algorithms that compute only local (pixel location-based) changes, we propose to minimize a novel region-based energy functional based on Bhattacharya coefficient involving histograms of image features. the optimization of the proposed energy functional simply consists of two very efficient searches if a crude segmentation such as a bounding box around the region of change is sufficient. Also, it allows variational optimization via level set-based curve evolution for supervised binary image labeling. the framework is demonstrated to cope well with considerable camera motion and shafts of objects between the test and the reference images. We illustrate encouraging results on finding bounding box around abnormality from brain MRI, object detection for maritime surveillance, and segmenting oil-sand particles from conveyor belt images.
In this paper, we propose a novel framework for automated analysis of surveillance videos. By analysis, we imply summarizing and mining of the information in the video for learning usual patterns and discovering unusu...
详细信息
ISBN:
(纸本)9781424442195
In this paper, we propose a novel framework for automated analysis of surveillance videos. By analysis, we imply summarizing and mining of the information in the video for learning usual patterns and discovering unusual ones. We approach this video analysis problem by acknowledging that a video contains information at multiple levels and in multiple attributes. Each such component and co-occurrences of these component values play an important role in characterizing an event as usual or unusual. therefore, we cluster the video data at multiple levels of abstraction and in multiple attributes and view these clusters as a summary of the information in the video. We apply cluster algebra to mine this summary from multiple perspectives and to adapt association learning for automated selection of components because of which the event is unusual. We also propose a novel incremental clustering algorithm.
A novel nonlinear cooperative approach to image denoising and restoration is presented. Samples from the image field with similar characteristics are first grouped into clusters by first performing image decomposition...
详细信息
ISBN:
(纸本)9781424442195
A novel nonlinear cooperative approach to image denoising and restoration is presented. Samples from the image field with similar characteristics are first grouped into clusters by first performing image decomposition based on the Mumford-Shah model using a total variational framework and performing fuzz), c-means clustering within each image partition. Samples within each cluster are then aggregated using an cooperative Bayesian estimation method based on information from all the samples to provide a nonlinear estimate of the original image. the proposed method exploits information redundancy within each cluster to denoise and restore the original image. Furthermore, the proposed cooperative Bayesian estimation method is capable of suppressing noise and reducing image degradation while preserving image detail by utilizing intra-cluster statistics. the experimental results using different types of images demonstrate that the proposed algorithm provides state-of-the-art image denoising performance in terms of both peak signal-to-noise ratio (PSNR) and subjective visual quality
In shape recognition, a multiscale description provides more information about the object, increases discrimination power and immunity to noise. In this paper, we develop a new multiscale Fourier-based object descript...
详细信息
ISBN:
(纸本)9781424442195
In shape recognition, a multiscale description provides more information about the object, increases discrimination power and immunity to noise. In this paper, we develop a new multiscale Fourier-based object description in 2-D space using a low-pass Gaussian filter (LPGF) and a high-pass Gaussian filter (HPGF), separately. Using the LPGF, at different scales, represents the inner and central part of an object more than the boundary. On the other hand using the HPGF, at different scales, represents the boundary and exterior parts of an object more than the central part. Our algorithms are also organized to achieve size, translation and rotation invariance. Evaluation indicates that representing the boundary and exterior parts more than the central part using the HPGF performs better than the LPGF based multiscale representation, and in comparison to Zernike moments and elliptic Fourier descriptors with respect to increasing noise.
Detection of representative frames, also called key-frames, is essential for efficient indexing, browsing and retrieval of video data and also for video summarization. Once a video stream is segmented into shots, the ...
详细信息
ISBN:
(纸本)9781424442195
Detection of representative frames, also called key-frames, is essential for efficient indexing, browsing and retrieval of video data and also for video summarization. Once a video stream is segmented into shots, the representative frames or key-frames for the shot are selected. the number of such frames in a shot may vary depending on the variation it? the content. thus, for a wide variety of shots automatic selection of suitable number of representative frames still remains a challenge. In this work, we propose a novel scheme for key-frame detection by dividing an available shot into subshots using hypothesis testing and majority voting. Each subshot is supposed to be uniform in terms of visual content. then for each subshot, the frame rendering the highest fidelity is extracted as the key-frame. Experimental result shows that the scheme works satisfactorily for a wide variety of shots.
Complete design and implementation of a robust palm biometrics recognition and verification system has been presented. the paper attempts to scientifically develop a comprehensive set of hand geometry features and dev...
详细信息
ISBN:
(纸本)9781424442195
Complete design and implementation of a robust palm biometrics recognition and verification system has been presented. the paper attempts to scientifically develop a comprehensive set of hand geometry features and develop an original algorithm from a fundamental level to robustly compute a selected set of features so as to minimize palm placement effect. these features are combined with chrominance features, to achieve recognition and verification accuracy, significant with respect to the current state in palm biometrics research. the algorithm has been kept robust, simple and computationally efficient while the implementation is relatively inexpensive. the experiments on recognition, Principal Components Analysis (PCA) and verification, on a dataset of 100 users strongly confirm the utility of robustly calculating a comprehensive set Of scientifically selected hand geometry features. the results uncover the potential of hand geometry features and confirm that the system can be used in medium to high security environments.
the GPUs pack high computation power and a restricted architecture into easily available hardware today. they are now used as computation co-processors and come with programming models that treat them as standard para...
详细信息
ISBN:
(纸本)9781424442195
the GPUs pack high computation power and a restricted architecture into easily available hardware today. they are now used as computation co-processors and come with programming models that treat them as standard parallel architectures. We explore the problem of real time ray casting of large deformable models (over a million triangles) on large displays (a million pixels) on an off-the-shelf GPU in this paper Ray casting is an inherently, parallel and highly compute intensive operation. We build a GPU-efficient three-dimensional data structure for this purpose and a corresponding algorithm that uses it for fast ray casting. We also present fast methods to build the data structure on the SIMD GPUs, including a fast multi-split operation. We achieve real-time ray-casting of a million triangle model onto a million pixels on current Nvidia GPUs using the CUDA model. Results are presented on the data structure building and ray casting on a number of models. the ideas presented here are likely to extend to later models and architectures of the GPU as well as to other multi core architectures.
We present a novel learning-based framework for detecting interesting events in soccer videos. the input to the system is a raw soccer video. We have learning at three levels - learning to detect interesting low-level...
详细信息
ISBN:
(纸本)9781424442195
We present a novel learning-based framework for detecting interesting events in soccer videos. the input to the system is a raw soccer video. We have learning at three levels - learning to detect interesting low-level features from image and video data using Support Vector Machines (hereafter SVMs), and a hierarchical Conditional Random Field(hereafter CRF-) based methodology to learn the dependencies of mid-level features and their relation withthe low level features, and high level decisions ('interesting events') and their relation withthe mid-level features: all on the basis of training video data. Descriptors are spatio-temporal in nature - they can be associated with a region in an image or a set of frames. Temporal patterns of descriptors characterise an event. We apply this framework to parse soccer videos into Interesting (a goal or a goal miss) and Non-Interesting videos. We present results of numerous experiments in support of the proposed strategy.
We propose a novel framework for object detection and localization in images containing appreciable clutter and occlusions. the problem is cast in a statistical hypothesis testing framework. the image under test is co...
详细信息
ISBN:
(纸本)9781424442195
We propose a novel framework for object detection and localization in images containing appreciable clutter and occlusions. the problem is cast in a statistical hypothesis testing framework. the image under test is converted into a set of local features using affine invariant local region detectors, described using the popular SIFT descriptor Due to clutter and occlusions, this set is expected to contain features which do not belong to the object. We sample subsets of local features from this set and test for the alternate hypothesis of object present against the null hypothesis of object absent. Further, we use a method similar to the recently proposed spatial scan statistic to refine the object localization estimates obtained from the sampling process. We demonstrate the results of our method on the two datasets TUD Motorbikes and TUD Cars. TUD Cars database has background clutter TUD Motorbikes dataset is recognized to have substantial variation in terms of scale, back-ground, illumination, viewpoint and occlusions.
this paper proposes efficient and robust methods for tracking a moving object at multiple spatial and temporal resolution levels. the efficiency comes from optimising the amounts of spatial and temporal data processed...
详细信息
ISBN:
(纸本)9781424442195
this paper proposes efficient and robust methods for tracking a moving object at multiple spatial and temporal resolution levels. the efficiency comes from optimising the amounts of spatial and temporal data processed. the robustness results from multi-level coarse-to-fine state-space searching. Tracking across resolution levels incurs a accuracy-versus-speed trade-off. For example, tracking at higher resolutions incurs greater processing cost, while maintaining higher accuracy in estimating the position of the moving object. We propose a novel spatial multi-scale tracker that tracks at the optimal accuracy-versus-speed operating point. Next, we relax this requirement to propose a multi-resolution tracker that operates at a minimum acceptable performance level. Finally, we extend these ideas to a multi-resolution spatio-temporal tracker We show results of extensive experimentation in support of the proposed approaches.
暂无评论