Mutual information has been used in many clustering algorithms for measuring general dependencies between random data variables, but its difficulties in computing for small size datasets has limited its efficiency for...
详细信息
Mutual information has been used in many clustering algorithms for measuring general dependencies between random data variables, but its difficulties in computing for small size datasets has limited its efficiency for clustering in many applications. A novel clustering method is proposed which estimates mutual information based on information potential computed pair-wise between data points and without any prior assumptions about cluster density function. The proposed algorithm increases the mutual information in each step in an agglomerative hierarchy scheme. We have shown experimentally that maximizing mutual information between data points and their class labels will lead to an efficient clustering. Experiments done on a variety of artificial and real datasets show the superiority of this algorithm, besides its low computational complexity, in comparison to other information based clustering methods and also some ordinary clustering algorithms.
Statistical learning based face detection systems search multiple scale sub-frames of an image or frame of a video stream with a trained classifier to detect face objects. If the frame is large there will be a large n...
详细信息
ISBN:
(纸本)9781424431625
Statistical learning based face detection systems search multiple scale sub-frames of an image or frame of a video stream with a trained classifier to detect face objects. If the frame is large there will be a large number of these sub-frames. Sections of the frame with a regional variance below a predefined threshold do not need to be searched as it is not possible for a face to exist in these spaces. A preprocessing system to eliminate these low variance regions of a frame is presented in this paper. A top down quad-tree deconstruction of the frame is used to accomplish this task. Regional variance is computed and tested to determine if a given quadrant should be further broken down. If below a predefined threshold that region will be eliminated in a mask image. This procedure is continued until all sections are eliminated or a defined tree depth is reached. The resulting mask image is then smoothed and subjected to thresholding, merging the remaining valid search areas. The resulting filled mask is then used to determine whether a given sub-frame should be sent to a Viola-Jones face detection cascade. Preliminary results show promise in reducing the number of sub-frames that must be considered for detection, increasing the speed of the detection system.
A novel architecture for performing hue-saturation-value (HSV) domain enhancement of digital color images with non-uniform lighting conditions is proposed in this paper for video streaming applications. The approach p...
详细信息
A novel architecture for performing hue-saturation-value (HSV) domain enhancement of digital color images with non-uniform lighting conditions is proposed in this paper for video streaming applications. The approach promotes log-domain computation to eliminate all multiplications, divisions and exponentiations utilizing the effective logarithmic estimation modules. An optimized quadrant symmetric architecture is incorporated into the design of homomorphic filter for the enhancement of intensity value. Efficient modules are also presented for conversion between RGB and HSV color spaces. The design is able to bring out details hidden in shadow regions of the image. It is capable of producing 187.86 million outputs per second (MOPs) on Xilinx's Virtex II XC2V2000-4ff896 field programmable gate array (FPGA) at a clock frequency of 187.86 MHz. It can process over 179.1 (1024 X 1024) frames per second and consumes approximately 70.7% and 76.8% less hardware resource with 127% and 280% performance boost when compared to the designs with machine learning algorithm [10], and with separated dynamic and contrast enhancements [11], respectively.
Diabetic retinopathy is the leading cause of blindness in the working age population in the industrialized world. computer assisted analysis has the potential to assist in the early detection of diabetes by regular sc...
详细信息
One of commonly used criteria for finding an endmember set is to assume that for a given number of endmembers, p, a p-vertex simplex with its vertices specified by p endmembers always yields the maximum volume. Since ...
详细信息
One of commonly used criteria for finding an endmember set is to assume that for a given number of endmembers, p, a p-vertex simplex with its vertices specified by p endmembers always yields the maximum volume. Since there are also other criteria which have been widely used for endmember extraction, the issue of interest is "does an endmember set really produce a simplex with maximum volume?" In other words, using the criterion of a simplex with maximum volume is a better and more effective measure than other criteria currently being used by endmember extraction such as orthogonal projection-based pixel purity index (PPI), fully constrained least squares-based spectral unmixing, etc. This paper explores this issue by investigating a number of popular endmember extraction algorithms which are designed by different criteria. An extensive experiment-based study is also conducted for comparative analysis.
Canada is one of the major exporters of wheat in the world. The quality of these exports is well known and factors such as lack of insect infestation are very important. The use of thermal images for subsequent analys...
详细信息
Canada is one of the major exporters of wheat in the world. The quality of these exports is well known and factors such as lack of insect infestation are very important. The use of thermal images for subsequent analysis of temperatures profiles for grain classification and insect detection is a method under investigation. This paper presents an approach for automatic image segmentation of the wheat kernels based on the combined use of wavelet analysis and pulse coupled neural networks. It is shown that using wavelets as a preprocessing technique yields a consistent accurate segmentation in terms of the iteration number in which the network yields reliable edges of the wheat kernels. Subsequent analysis of these segmentations can determine internal qualities such as infestations.
Water diffusion measurements have been shown to be sensitive to tissue cellular size, extra cellular volume, and membrane permeability. Therefore, diffusion tensor imaging (DTI) by MRI can be used to characterize high...
详细信息
ISBN:
(纸本)9781424406715;1424406714
Water diffusion measurements have been shown to be sensitive to tissue cellular size, extra cellular volume, and membrane permeability. Therefore, diffusion tensor imaging (DTI) by MRI can be used to characterize highly cellular regions of tumors versus acellular regions, distinguishing cystic regions from solid regions. An automatic segmentation method is proposed in this paper based on a multi-phase clustering algorithm to segment the brain tumors in a feature space extracted from DTI images. The algorithm is applied on images of a total of 20 patients with 4 different types of tumors. The tumor region segmentation was 92% accurate based on the segmentation results using anatomical images and 100% accurate based on biopsy results. In general, the segmentation results obtained by the proposed method revealed a strong agreement with the biopsy results and anatomical images, providing support for the accuracy and robustness of the proposed feature space and the segmentation procedure.
In earlier work, we have shown the importance of including 3D shape characteristics when analyzing regions of interest (ROIs) in magnetic resonance imaging (MRI) data. Spherical harmonics (SPHARM) based ROI shape desc...
详细信息
In earlier work, we have shown the importance of including 3D shape characteristics when analyzing regions of interest (ROIs) in magnetic resonance imaging (MRI) data. Spherical harmonics (SPHARM) based ROI shape descriptors were proposed and shown to provide important complementary information to traditionally used simple volumetric ROI measures. In this paper we extend our SPHARM shape parameterization technique by using functions defined on concentric spherical shells. We then propose the use of a novel radial transform to obtain unique features even under independent rotations of the constituting shells. These enhanced features enable the analysis of 3D ROIs with complex topologies including those with possible disconnections (e.g. ventricles). We validate the proposed 3D shape descriptors on synthetic data and demonstrate their sensitivity to subtle shape changes in the presence of inter-subject variability. We also apply our approach to real MRI data and detect significant shape changes in the left and right thalamus in Parkinson's disease (PD) patients when compared against normal volunteers, complementing the observed volumetric changes.
Phase-shifting fringe-projection methods have been increasingly used for three-dimensional (3D) object surface modeling to permit full-field measurement. This paper presents a real-time full-field high-resolution 3D s...
详细信息
Phase-shifting fringe-projection methods have been increasingly used for three-dimensional (3D) object surface modeling to permit full-field measurement. This paper presents a real-time full-field high-resolution 3D surface-shape measurement system implemented with an efficient 3D shape measurement pipeline and triangular-pattern phase-shifting based on off-the-shelf components, software synchronization and a single computer-processor. The system projects computer-generated fringe patterns with a triangular intensity profile onto an object via a Digital Light Processing (DLP) projector. The projected patterns are electronically shifted and a CCD camera synchronized with the DLP projector by software captures the images from another direction. The captured images are processed by a single computer to reconstruct the 3D shape using triangular-pattern phase-shifting algorithms, and the model is displayed in real time. The 3D shape acquisition system achieved a speed of 5.6 fps for an image size of 648times494 pixels using the two-step triangular-pattern phase-shifting method, without any hardware synchronization or dual processing.
This paper presents a novel user interaction concept for document image scanning with mobile phones. A high resolution mosaic image is constructed in two main stages. Firstly, online camera motion estimation is applie...
详细信息
This paper presents a novel user interaction concept for document image scanning with mobile phones. A high resolution mosaic image is constructed in two main stages. Firstly, online camera motion estimation is applied to the phone to assist the user to capture small image patches of the document page. Automatic image stitching process with the help of estimated device motion is carried out to reconstruct the full view of the document. Experiments on document images captured and processed with mosaicing software clearly show the feasibility of the approach.
暂无评论