由于FREAK算法存在不具备尺度不变性的缺陷,特征点匹配策略单一容易出现匹配效果不理想的问题,鉴于SIFT和RANSAC算法思想,本文提出一种改进的FREAK算法:SFREAK(SIFT and FREAK).首先,生成高斯差分金字塔图像,使检测出来的特征点具有尺...由于FREAK算法存在不具备尺度不变性的缺陷,特征点匹配策略单一容易出现匹配效果不理想的问题,鉴于SIFT和RANSAC算法思想,本文提出一种改进的FREAK算法:SFREAK(SIFT and FREAK).首先,生成高斯差分金字塔图像,使检测出来的特征点具有尺度不变性;然后采用FREAK描述符对特征点进行描述,获得二进制描述子;最后,在特征点匹配过程中通过Hamming距离匹配进行粗匹配,并结合RANSAC算法对匹配点对进行提纯,实现两幅图像的特征点匹配.实验结果表明,本文提出的改进算法有效解决了FREAK不具备尺度不变性的缺陷,在图像发生尺度变化时,SFREAK算法特征点匹配准确率达到95.7%,相比于FREAK提高了61.9%.另外,本文改进的算法与传统SIFT、FREAK算法相比,表现出更好的鲁棒性.展开更多
Background Previous studies reported interleukin-27 （IL-27）, interferon-y （IFN-γ）, or adenosine deaminase （ADA） alone plays a helpful role in diagnosing tuberculous pleural effusion （TPE）. The present study a...Background Previous studies reported interleukin-27 （IL-27）, interferon-y （IFN-γ）, or adenosine deaminase （ADA） alone plays a helpful role in diagnosing tuberculous pleural effusion （TPE）. The present study aims at comparing the diagnostic accuracy of pleural IL-27, IFN-γ, and ADA, and investigate the diagnostic accuracy of the combination of IL-27, IFN-γ, or/ and ADA for differentiating TPE from pleural effusions with the other etiologies. Methods The concentrations of IL-27, IFN-γ and ADA were simultaneously determined in pleural fluids and sera from 40 patients with TPE; 26 with malignant pleural effusion, seven with infectious pleural effusion, and eight with transudative pleural effusion by enzyme linked immunosorbent assay and colorimetric method. The corresponding biochemical indexs were also simultaneously determined. Results The concentrations of pleural IL-27 and IFN-γ in the tuberculous group were significantly higher than those in the malignant, infectious, and transudative groups. The concentrations of ADA in TPE were significantly higher than those in MPE or transudative effusions, while much lower than those in infectious effusions. Among these three biomarkers, IL- 27 was the most effective for TPE diagnosis, with the cut off value of 900.8 ng/L. IL-27 had a high sensitivity of 95% and specificity of 97.6% for differential diagnosis of TPE from non-TPEs. Combinations of IL-27, IFN-γ and ADA measurements further increased the sensitivity or specificity up to 100%. Conclusions Compared to non-TPEs, IL-27, IFN-γ and ADA all simultaneously increased in TPE; and among these three rapid detection methods, IL-27 appeared to be the best for distinguishing tuberculous from non-TPEs, especially from MPE. Combinations of the three markers （IL-27, IFN-γ and ADA） yielded the highest sensitivity and specificity. These findings suggest that the applications of a new biomarker, IL-27, alone or with IFN-γ and ADA, may contribute to more efficient diagnosis strategies in the management of tuberculous pleurisy.展开更多
文摘由于FREAK算法存在不具备尺度不变性的缺陷,特征点匹配策略单一容易出现匹配效果不理想的问题,鉴于SIFT和RANSAC算法思想,本文提出一种改进的FREAK算法:SFREAK(SIFT and FREAK).首先,生成高斯差分金字塔图像,使检测出来的特征点具有尺度不变性;然后采用FREAK描述符对特征点进行描述,获得二进制描述子;最后,在特征点匹配过程中通过Hamming距离匹配进行粗匹配,并结合RANSAC算法对匹配点对进行提纯,实现两幅图像的特征点匹配.实验结果表明,本文提出的改进算法有效解决了FREAK不具备尺度不变性的缺陷,在图像发生尺度变化时,SFREAK算法特征点匹配准确率达到95.7%,相比于FREAK提高了61.9%.另外,本文改进的算法与传统SIFT、FREAK算法相比,表现出更好的鲁棒性.
基金This study was supported by grants from the National Natural Science Foundation of China，Natural Science Foundation of Guangxi Zhuang Autonomous Region (No.0728137).The authors have declared that no conflict of interest exists
文摘Background Previous studies reported interleukin-27 （IL-27）, interferon-y （IFN-γ）, or adenosine deaminase （ADA） alone plays a helpful role in diagnosing tuberculous pleural effusion （TPE）. The present study aims at comparing the diagnostic accuracy of pleural IL-27, IFN-γ, and ADA, and investigate the diagnostic accuracy of the combination of IL-27, IFN-γ, or/ and ADA for differentiating TPE from pleural effusions with the other etiologies. Methods The concentrations of IL-27, IFN-γ and ADA were simultaneously determined in pleural fluids and sera from 40 patients with TPE; 26 with malignant pleural effusion, seven with infectious pleural effusion, and eight with transudative pleural effusion by enzyme linked immunosorbent assay and colorimetric method. The corresponding biochemical indexs were also simultaneously determined. Results The concentrations of pleural IL-27 and IFN-γ in the tuberculous group were significantly higher than those in the malignant, infectious, and transudative groups. The concentrations of ADA in TPE were significantly higher than those in MPE or transudative effusions, while much lower than those in infectious effusions. Among these three biomarkers, IL- 27 was the most effective for TPE diagnosis, with the cut off value of 900.8 ng/L. IL-27 had a high sensitivity of 95% and specificity of 97.6% for differential diagnosis of TPE from non-TPEs. Combinations of IL-27, IFN-γ and ADA measurements further increased the sensitivity or specificity up to 100%. Conclusions Compared to non-TPEs, IL-27, IFN-γ and ADA all simultaneously increased in TPE; and among these three rapid detection methods, IL-27 appeared to be the best for distinguishing tuberculous from non-TPEs, especially from MPE. Combinations of the three markers （IL-27, IFN-γ and ADA） yielded the highest sensitivity and specificity. These findings suggest that the applications of a new biomarker, IL-27, alone or with IFN-γ and ADA, may contribute to more efficient diagnosis strategies in the management of tuberculous pleurisy.
