交通异常事件检测是智能交通系统中的关键任务,但现有目标检测算法在该领域的应用尚存在技术瓶颈,针对检测精度不足、模型对复杂场景的适应性差以及缺乏高质量的公开数据集等问题,提出了一种改进的YOLOv6模型,旨在提高交通异常事件(如...交通异常事件检测是智能交通系统中的关键任务,但现有目标检测算法在该领域的应用尚存在技术瓶颈,针对检测精度不足、模型对复杂场景的适应性差以及缺乏高质量的公开数据集等问题,提出了一种改进的YOLOv6模型,旨在提高交通异常事件(如车辆碰撞、单车冲撞和车辆起火)检测的准确性和性能。先将原YOLOv6模型中的损失函数替换为CIoU损失函数,以增强模型的定位精度,后引入CBAM注意力机制,以提高模型对关键特征的关注度,再采用自动混合精度训练策略优化训练过程,最后为了验证改进效果,通过游戏引擎Grand Theft Auto V生成数据集,并对其进行标注,涵盖3类交通异常事件。试验结果表明:1)提出的改进YOLOv6模型在交通异常事件的检测任务中可获得87.2%的平均检测精度,在各项指标上表现优异;2)召回率AR较次优模型提高2.1%,IoU阈值为0.5时,平均精度mAP高出2.6%;IoU阈值为0.5至0.95时,mAP增长3.7%;3)车辆相撞、单车相撞和车辆起火烧毁的精度分别达到79.9%、37.6%和65.6%,均优于次优模型,验证了改进方法的有效性。展开更多
In recent years the photovoltaic community has witnessed the unprecedented development of perovskite solar cells（PSCs） as they have taken the lead in emergent photovoltaic technologies. The power conversion efficien...In recent years the photovoltaic community has witnessed the unprecedented development of perovskite solar cells（PSCs） as they have taken the lead in emergent photovoltaic technologies. The power conversion efficiency of this new class of solar cells has been increased to a point where they are beginning to compete with more established technologies. Although PSCs have evolved a variety of structures, the use of hole-transporting materials（HTMs） remains indispensable. Here, an overview of the various types of available HTMs is presented. This includes organic and inorganic HTMs and is presented alongside recent progress in associated aspects of PSCs, including device architectures and fabrication techniques to produce high-quality perovskite films. The structure, electrochemistry, and physical properties of a variety of HTMs are discussed, highlighting considerations for those designing new HTMs. Finally, an outlook is presented to provide more concrete direction for the development and optimization of HTMs for highefficiency PSCs.展开更多
文摘交通异常事件检测是智能交通系统中的关键任务,但现有目标检测算法在该领域的应用尚存在技术瓶颈,针对检测精度不足、模型对复杂场景的适应性差以及缺乏高质量的公开数据集等问题,提出了一种改进的YOLOv6模型,旨在提高交通异常事件(如车辆碰撞、单车冲撞和车辆起火)检测的准确性和性能。先将原YOLOv6模型中的损失函数替换为CIoU损失函数,以增强模型的定位精度,后引入CBAM注意力机制,以提高模型对关键特征的关注度,再采用自动混合精度训练策略优化训练过程,最后为了验证改进效果,通过游戏引擎Grand Theft Auto V生成数据集,并对其进行标注,涵盖3类交通异常事件。试验结果表明:1)提出的改进YOLOv6模型在交通异常事件的检测任务中可获得87.2%的平均检测精度,在各项指标上表现优异;2)召回率AR较次优模型提高2.1%,IoU阈值为0.5时,平均精度mAP高出2.6%;IoU阈值为0.5至0.95时,mAP增长3.7%;3)车辆相撞、单车相撞和车辆起火烧毁的精度分别达到79.9%、37.6%和65.6%,均优于次优模型,验证了改进方法的有效性。
基金financial support from the Natural Science Foundation of China (grant numbers: 51661135021, 21606039, 91233201, and 21276044)
文摘In recent years the photovoltaic community has witnessed the unprecedented development of perovskite solar cells（PSCs） as they have taken the lead in emergent photovoltaic technologies. The power conversion efficiency of this new class of solar cells has been increased to a point where they are beginning to compete with more established technologies. Although PSCs have evolved a variety of structures, the use of hole-transporting materials（HTMs） remains indispensable. Here, an overview of the various types of available HTMs is presented. This includes organic and inorganic HTMs and is presented alongside recent progress in associated aspects of PSCs, including device architectures and fabrication techniques to produce high-quality perovskite films. The structure, electrochemistry, and physical properties of a variety of HTMs are discussed, highlighting considerations for those designing new HTMs. Finally, an outlook is presented to provide more concrete direction for the development and optimization of HTMs for highefficiency PSCs.
