We present a combined real-time face region tracking and highly accurate face recognition technique for an intelligent surveillance system. High-resolution face images are very important to achieving accurate identifi...
详细信息
We present a combined real-time face region tracking and highly accurate face recognition technique for an intelligent surveillance system. High-resolution face images are very important to achieving accurate identification of a human face. Conventional surveillance or security systems, however, usually provide poor image quality because they use only fixed cameras to record scenes passively. We have implemented a real-time surveillance system that tracks a moving face using four pan-tilt-zoom (PTZ) cameras. While tracking, the region-of-interest (ROI) can be obtained by using a low-pass filter and background subtraction with the PTZ. Color information in the ROI is updated to extract features for optimal tracking and zooming. FaceIt/sup /spl reg//, which is one of the most popular face recognition software packages, is evaluated and then used to recognize the faces from the video signal. Experimentation with real human faces showed highly acceptable results in the sense of both accuracy and computational efficiency.
We investigated the relationship between the intracranial pulse pressure (ICP/sub PP/) and the mean intracranial pressure (ICP/sub M/). In adult patients, several research groups have described a linear relationship b...
详细信息
We investigated the relationship between the intracranial pulse pressure (ICP/sub PP/) and the mean intracranial pressure (ICP/sub M/). In adult patients, several research groups have described a linear relationship between ICP/sub PP/ and ICP/sub M/ within the range of cerebral autoregulation. Current monitoring and therapy are mainly based on the mean ICP/sub M/, since it is believed that the ICP/sub M/ contains most of the information provided by the other pulse morphology metrics. In this paper we attempt to answer whether there is further information within the ICP morphology not explained by ICP/sub M/ that might be of prognostic significance. We screened ICP records of 42 patients admitted to the Pediatric Intensive Care Unit at Doernbecher Children's Hospital for segments in which the ICP/sub M/ varied at least 5 mmHg during a 1-hour period. We found 54 segments in 9 different pediatric TBI patients (ages 0.2-17.8 years, mean=9.9 years). ICP/sub PP/ and ICP/sub M/ were calculated for each pulse using an automatic pressure detection algorithm. The coefficient of linear correlation r was > 0.70 in 43/54 segments (p < 0.001), which indicates that there exists a linear relationship between ICP/sub PP/ and ICP/sub M/. However, we found r > 0.90 only in 16/54 segments (p=NS) . This result and visual inspection of ICP/sub PP/ vs. ICP/sub M/ density plots suggest that ICP pulse pressure is not fully explained by the ICP M.
Gene selection, cancer classification and functional gene classification are three main concerns and interests by biologists for cancer detection, cancer classification, and understanding the functions of genes from t...
详细信息
ISBN:
(纸本)0769519571
Gene selection, cancer classification and functional gene classification are three main concerns and interests by biologists for cancer detection, cancer classification, and understanding the functions of genes from the molecular level of tissues, where the large number of genes and relatively small number of experiments in gene expression data generate a great challenge. After a brief introduction of support vector machine(SVM) for classification, this paper presents recent SVM approaches for gene selection, cancer classification and functional gene classification followed by analysis on the advantages and limitations of SVM on these applications.
In this paper we present a methodology on how to use directive elements in an adaptive array methodology. Typically one uses isotropic elements having practically no gain then the signal level is increased by putting ...
详细信息
In this paper we present a methodology on how to use directive elements in an adaptive array methodology. Typically one uses isotropic elements having practically no gain then the signal level is increased by putting hundreds and thousands of these elements together. In this paper we demonstrate a methodology where the elements can be arbitrarily spaced and may even be non-planar. In addition it is shown how to deal with nonuniformly spaced and non-planar arrays. We illustrate these principles in a direction of arrival (DOA) estimation utilizing directive elements.
This paper investigates the application of a knowledge-based approach, founded on semantic networks, to the automatic land use mapping assisted by low resolution satellite images. Like the visual photo-interpretation,...
This paper investigates the application of a knowledge-based approach, founded on semantic networks, to the automatic land use mapping assisted by low resolution satellite images. Like the visual photo-interpretation, the automatic image interpretation considers scene and sensors knowledge, delivered by an expert photo-interpreter, as well as additional information about the region like the digital elevation model, the position of the emergent rocks, the mapping of the water bodies and the road-network. By this means, the analysis of a scene can be automatically performed, mimicking the reasoning of the photo-interpreter. The implementation of such proposal employed the GEOAIDA [J. Buckner et al., June 2001] system, a flexible environment for image interpretation developed at the University of Hanover, which exploits semantic networks to structure the domain specific knowledge. In the reported experiments, a multispectral SPOT 3 XS image was analysed resulting were evaluated and compared with a manually made reference map of the investigated scene. The automatically obtained results were evaluated and compared with a manually made reference map of the investigated scene. The experimental results demonstrate the potential of a knowledge-based approach for low resolution satellite images interpretation.
This paper describes the UIcluster software tool, which partitions expressed sequence tag (EST) sequences and other genetic sequences into "clusters" based on sequence similarity. Ideally, each cluster will ...
详细信息
The paper reports new results on developing parallel algorithms for multiprocessor scheduling with use of cellular automata (CAs). The simpliest case when a multiprocessor system is limited to two processors, but with...
详细信息
The problem of parallel and distributed function optimization is considered. Two coevolutionary algorithms with different degrees of parallelism and different levels of a global coordination are used for this purpose ...
In this paper, ultrasound breast image segmentation is improved by using the volumetric data available in neighboring slices. The new algorithm extends the EM/MPM framework to 3D by including pixels from neighboring f...
详细信息
An error-resilient coding scheme is proposed for the transmission of images over unreliable channels. Forward Error Correction is used in conjunction with the error-resilient source coder for the protection of the com...
详细信息
暂无评论