感兴趣区域(Region of interest,ROI)提取在生物特征识别中,常用于减少后续处理的计算消耗,提高识别模型的准确性,是生物识别系统中预处理的关键步骤.针对生物识别数据,提出了一种鲁棒的ROI提取方法.方法使用语义分割模型作为基础,通过...感兴趣区域(Region of interest,ROI)提取在生物特征识别中,常用于减少后续处理的计算消耗,提高识别模型的准确性,是生物识别系统中预处理的关键步骤.针对生物识别数据,提出了一种鲁棒的ROI提取方法.方法使用语义分割模型作为基础,通过增加全局感知模块,与分割模型形成对抗结构,为模型提供先验知识,补充全局视觉模式信息,解决了语义分割模型的末端收敛困难问题,提高了模型的鲁棒性和泛化能力.在传统二维(2D)指纹、人脸、三维(3D)指纹和指纹汗孔数据集中验证了方法的有效性.实验结果表明,相比于现有方法,所提出的ROI提取方法更具鲁棒性和泛化能力,精度最高.展开更多
Background The down-regulation of constitutive nitric oxide synthase （cNOS） and up-regulation of inducible nitric oxide synthase （iNOS） are associated with the allergen-provocated airway hyperresponsiveness （AHR...Background The down-regulation of constitutive nitric oxide synthase （cNOS） and up-regulation of inducible nitric oxide synthase （iNOS） are associated with the allergen-provocated airway hyperresponsiveness （AHR）. This study aimed to determine whether their alteration also plays an important role in the AHR induced by lipopolysaccharide （LPS）. Methods Hartley male guinea pigs, weighing between 250 g and 350 g, were injected with LPS at a dose of 1 mg/kg every 24 hours for three days. A non-selective NOS inhibitor, N^-nitro-L-arginine methyl ester （L-NAME）, or a selective inducible NOS inhibitor, aminoguanidine （AG）, were used thirty minutes before each injection of LPS. Airway reactions, nitric oxide （NO） production and inflammatory changes were detected 24 hours after the last dose of LPS. Results AG significantly decreased the NO production in the bronchoalveolar lavage fluid （BALF） and sharply reduced the intensity of bronchoconstdction to histamine challenge. L-NAME also significantly decreased the NO production in the BALF, but had no effect on airway reactions or, perhaps, a tendency to enhance the intensity of AHR. Conclusions The data suggest that inducible NOS contributes to the AHR induced by repetitive intraperitoneal LPS, and constitutive NOS was also involved.展开更多
文摘感兴趣区域(Region of interest,ROI)提取在生物特征识别中,常用于减少后续处理的计算消耗,提高识别模型的准确性,是生物识别系统中预处理的关键步骤.针对生物识别数据,提出了一种鲁棒的ROI提取方法.方法使用语义分割模型作为基础,通过增加全局感知模块,与分割模型形成对抗结构,为模型提供先验知识,补充全局视觉模式信息,解决了语义分割模型的末端收敛困难问题,提高了模型的鲁棒性和泛化能力.在传统二维(2D)指纹、人脸、三维(3D)指纹和指纹汗孔数据集中验证了方法的有效性.实验结果表明,相比于现有方法,所提出的ROI提取方法更具鲁棒性和泛化能力,精度最高.
文摘Background The down-regulation of constitutive nitric oxide synthase （cNOS） and up-regulation of inducible nitric oxide synthase （iNOS） are associated with the allergen-provocated airway hyperresponsiveness （AHR）. This study aimed to determine whether their alteration also plays an important role in the AHR induced by lipopolysaccharide （LPS）. Methods Hartley male guinea pigs, weighing between 250 g and 350 g, were injected with LPS at a dose of 1 mg/kg every 24 hours for three days. A non-selective NOS inhibitor, N^-nitro-L-arginine methyl ester （L-NAME）, or a selective inducible NOS inhibitor, aminoguanidine （AG）, were used thirty minutes before each injection of LPS. Airway reactions, nitric oxide （NO） production and inflammatory changes were detected 24 hours after the last dose of LPS. Results AG significantly decreased the NO production in the bronchoalveolar lavage fluid （BALF） and sharply reduced the intensity of bronchoconstdction to histamine challenge. L-NAME also significantly decreased the NO production in the BALF, but had no effect on airway reactions or, perhaps, a tendency to enhance the intensity of AHR. Conclusions The data suggest that inducible NOS contributes to the AHR induced by repetitive intraperitoneal LPS, and constitutive NOS was also involved.
