期刊文献+
共找到5篇文章
< 1 >
每页显示 20 50 100
基于Radon变换和Gabor变换鉴别运动模糊方向角 被引量:4
1
作者 加春燕 王昕 《华北科技学院学报》 2012年第3期19-22,27,共5页
对于匀速直线运动的模糊图像,准确鉴别运动方向角是图像复原的关键。分析了运动模糊图像的频谱特征,介绍了Radon变换的数学原理及用其估计运动模糊方向角的思路、步骤和数值实验结果。由于实际拍摄的模糊图像在很多时候频谱特征不够明显... 对于匀速直线运动的模糊图像,准确鉴别运动方向角是图像复原的关键。分析了运动模糊图像的频谱特征,介绍了Radon变换的数学原理及用其估计运动模糊方向角的思路、步骤和数值实验结果。由于实际拍摄的模糊图像在很多时候频谱特征不够明显,导致用Radon变换鉴别角度出现大的误差,为此,提出了基于Gabor变换的一种改进算法。算法运用"窗口"聚焦频谱图像中心,较好地消除了噪声干扰并克服了Radon变换的弊端,数值实验结果验证了该算法的有效性。 展开更多
关键词 运动模糊方向鉴别 运动模糊频谱特征 RADON变换 GABOR变换
下载PDF
基于鉴别矢量角嵌入的人脸识别
2
作者 孔万增 朱善安 《浙江大学学报(工学版)》 EI CAS CSCD 北大核心 2008年第11期1889-1893,共5页
针对基于构造邻接图的降维人脸识别问题,提出了一种鉴别矢量角嵌入的识别方法.构造了一幅有正/负连接边的邻接图,同类样本之间为正连接边,不同类样本的k近邻为负连接边.连接边权系数的测度采用矢量角代替矢量模,不但省去了传统方法中对... 针对基于构造邻接图的降维人脸识别问题,提出了一种鉴别矢量角嵌入的识别方法.构造了一幅有正/负连接边的邻接图,同类样本之间为正连接边,不同类样本的k近邻为负连接边.连接边权系数的测度采用矢量角代替矢量模,不但省去了传统方法中对热核权函数t参数的估计,而且降低了由于图像样本间的亮度差异对识别率造成的影响.样本数据保持邻接矢量角从高维空间嵌入至低维子空间,在分类识别中采用了角度最近邻分类器.Yale库和UMIST人脸库上的人脸识别实验结果表明,该算法比其他算法有更好的识别率. 展开更多
关键词 人脸识别 鉴别矢量 正/负连接边 最近邻
下载PDF
微波波道配置方法浅论
3
作者 冶刚 《中国无线电》 2008年第7期52-53,共2页
提出一种微波波道配置方法,并理论推导出该方法实现微波波道复用的条件。
关键词 波道复用 干扰限值 鉴别角
下载PDF
Soluble forms of extracellular cytokeratin 18 may differentiate simple steatosis from nonalcoholic steatohepatitis 被引量:20
4
作者 Yusuf Yilmaz Enver Dolar +8 位作者 Engin Ulukaya Semra Akgoz Murat Keskin Murat Kiyici Sibel Aker Arzu Yilmaztepe Selim Gurel Macit Gulten Selim Giray Nak 《World Journal of Gastroenterology》 SCIE CAS CSCD 2007年第6期837-844,共8页
AIM: To investigate whether serum levels of two soluble forms of extracellular cytokeratin 18 (M30-antigen and M65-antigen) may differentiate nonalcoholic steatohepatitis (NASH) from simple steatosis in patients with ... AIM: To investigate whether serum levels of two soluble forms of extracellular cytokeratin 18 (M30-antigen and M65-antigen) may differentiate nonalcoholic steatohepatitis (NASH) from simple steatosis in patients with nonalcoholic fatty liver disease (NAFLD). METHODS: A total of 83 patients with suspected NAFLD and 49 healthy volunteers were investigated. Patients with suspected NAFLD were classified according to their liver histology into four groups: definitive NASH (n = 45), borderline NASH (n = 24), simple fatty liver (n = 9), and normal tissue (n = 5). Serum levels of caspase-3 generated cytokeratin-18 fragments (M30-antigen) and total cytokeratin-18 (M65-antigen) were determined by ELISA. RESULTS: Levels of M30-antigen and M65-antigen were significantly higher in patients with definitive NASH compared to the other groups. An abnormal value (> 121.60 IU/L) of M30-antigen yielded a 60.0% sensitivity and a 97.4% specificity for the diagnosis of NASH. Sensitivity and specificity of an abnormal M65-antigen level (> 243.82 IU/L) for the diagnosis of NASH were 68.9% and 81.6%, respectively. Among patients with NAFLD, M30-antigen and M65-antigen levels distinguished between advanced fibrosis and early-stage fibrosis with a sensitivity of 64.7% and 70.6%, and a specificity of 77.3% and 71.2%, respectively. CONCLUSION: Serum levels of M30-antigen and M65-antigen may be of clinical usefulness to identify patients with NASH. Further studies are mandatory to better assess the role of these apoptonecrotic biomarkers in NAFLD pathophysiology. 展开更多
关键词 STEATOSIS STEATOHEPATITIS Cytokeratin 18 M30-antigen M65-antigen
下载PDF
Direct linear discriminant analysis based on column pivoting QR decomposition and economic SVD
5
作者 胡长晖 路小波 +1 位作者 杜一君 陈伍军 《Journal of Southeast University(English Edition)》 EI CAS 2013年第4期395-399,共5页
A direct linear discriminant analysis algorithm based on economic singular value decomposition (DLDA/ESVD) is proposed to address the computationally complex problem of the conventional DLDA algorithm, which directl... A direct linear discriminant analysis algorithm based on economic singular value decomposition (DLDA/ESVD) is proposed to address the computationally complex problem of the conventional DLDA algorithm, which directly uses ESVD to reduce dimension and extract eigenvectors corresponding to nonzero eigenvalues. Then a DLDA algorithm based on column pivoting orthogonal triangular (QR) decomposition and ESVD (DLDA/QR-ESVD) is proposed to improve the performance of the DLDA/ESVD algorithm by processing a high-dimensional low rank matrix, which uses column pivoting QR decomposition to reduce dimension and ESVD to extract eigenvectors corresponding to nonzero eigenvalues. The experimental results on ORL, FERET and YALE face databases show that the proposed two algorithms can achieve almost the same performance and outperform the conventional DLDA algorithm in terms of computational complexity and training time. In addition, the experimental results on random data matrices show that the DLDA/QR-ESVD algorithm achieves better performance than the DLDA/ESVD algorithm by processing high-dimensional low rank matrices. 展开更多
关键词 direct linear discriminant analysis column pivoting orthogonal triangular decomposition economic singular value decomposition dimension reduction feature extraction
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部