为了解决车辆行驶中面对各种复杂环境车道线检测算法精度不高的问题,提出一种基于改进的UFS网络检测算法(Ultra Fast Structure-aware Deep Lane Detection,UFS).首先,采用改进的Gamma校正对待检路面图像进行校正,降低光照、阴影等的影...为了解决车辆行驶中面对各种复杂环境车道线检测算法精度不高的问题,提出一种基于改进的UFS网络检测算法(Ultra Fast Structure-aware Deep Lane Detection,UFS).首先,采用改进的Gamma校正对待检路面图像进行校正,降低光照、阴影等的影响,以提升夜间图像纹理特征。然后引入非局部神经网络模块(Non-Local Block),充分提取图像全局特征,以提高检测可靠性。最后对改进后的算法使用Tusimple、CULane数据集进行测试。结果表明:改进后的模型在物体遮挡、光照变化、阴影干扰等复杂场景下,提升了对复杂噪声与多元场景的处理能力,车道分割的准确率有所改善,具有较好的鲁棒性。展开更多
Energy efficiency and energy-proportional computing have become a central focus in modern supercomputers. These supercomputers should provide high throughput per unit of power to be sustainable in terms of operating c...Energy efficiency and energy-proportional computing have become a central focus in modern supercomputers. These supercomputers should provide high throughput per unit of power to be sustainable in terms of operating cost and failure rates. In this paper, a power-bounded strategy is proposed that maximizes parallel application performance under a given power constraint. The strategy dynamically allocates power to core, uncore, and memory power domains within a node to maximize performance under a given power budget. Experiments on a 20-core Haswell-EP platform for a real-world parallel application GAMESS demonstrate that the proposed strategy delivers performance within 4% of the best possible performance for as much as 25% reduction in the minimum power budget required for maximum performance.展开更多
文摘为了解决车辆行驶中面对各种复杂环境车道线检测算法精度不高的问题,提出一种基于改进的UFS网络检测算法(Ultra Fast Structure-aware Deep Lane Detection,UFS).首先,采用改进的Gamma校正对待检路面图像进行校正,降低光照、阴影等的影响,以提升夜间图像纹理特征。然后引入非局部神经网络模块(Non-Local Block),充分提取图像全局特征,以提高检测可靠性。最后对改进后的算法使用Tusimple、CULane数据集进行测试。结果表明:改进后的模型在物体遮挡、光照变化、阴影干扰等复杂场景下,提升了对复杂噪声与多元场景的处理能力,车道分割的准确率有所改善,具有较好的鲁棒性。
文摘Energy efficiency and energy-proportional computing have become a central focus in modern supercomputers. These supercomputers should provide high throughput per unit of power to be sustainable in terms of operating cost and failure rates. In this paper, a power-bounded strategy is proposed that maximizes parallel application performance under a given power constraint. The strategy dynamically allocates power to core, uncore, and memory power domains within a node to maximize performance under a given power budget. Experiments on a 20-core Haswell-EP platform for a real-world parallel application GAMESS demonstrate that the proposed strategy delivers performance within 4% of the best possible performance for as much as 25% reduction in the minimum power budget required for maximum performance.
文摘目的:评估UF5000检测尿液中酵母样真菌(yeast-like cells,YLC)结果的准确性,并研究该参数能否用于筛查真菌性尿路感染。方法:选取318例疑似尿路感染患者的尿液标本,分别进行UF5000检测和细菌、真菌培养,显微镜镜检记录尿液中是否存在酵母菌。分别以镜下见到酵母菌和培养出真菌作为金标准,ROC曲线分析进行准确性评估。根据是否培养出真菌,以及真菌菌落生长的数量对尿液标本进行分组,非参数检验用于比较不同组间YLC水平的差异。结果:UF5000检测YLC真阳性的曲线下面积(Area under curve,AUC)为0.863,最佳cut off值2.80/μL。UF5000检测YLC预测真菌性尿路感染的AUC为0.913,最佳cutoff值8.95/μL。均有良好的敏感性(95.45%~97.37%)和阴性预测值(98.62%~98.73%)。组间比较显示真菌组YLC水平明显高于阴性组和细菌组(P值均为0.000),而真菌组内亚组仅≥105CFU/mL组YLC水平与105CFU/mL组差异有统计学意义(P=0.027)。结论:选取2.8/μL作为YLC的阳性标准,比厂家提供的6.0/μL更适用于本实验室的临床工作。选取8.95/μL作为最佳cutoff值时,UF5000检测的YLC结果可有效地筛查真菌性尿路感染。
