Underwater robotic operation usually requires visual perception(e.g.,object detection and tracking),but underwater scenes have poor visual quality and represent a special domain which can affect the accuracy of visual...Underwater robotic operation usually requires visual perception(e.g.,object detection and tracking),but underwater scenes have poor visual quality and represent a special domain which can affect the accuracy of visual perception.In addition,detection continuity and stability are important for robotic perception,but the commonly used static accuracy based evaluation(i.e.,average precision)is insufficient to reflect detector performance across time.In response to these two problems,we present a design for a novel robotic visual perception framework.First,we generally investigate the relationship between a quality-diverse data domain and visual restoration in detection performance.As a result,although domain quality has an ignorable effect on within-domain detection accuracy,visual restoration is beneficial to detection in real sea scenarios by reducing the domain shift.Moreover,non-reference assessments are proposed for detection continuity and stability based on object tracklets.Further,online tracklet refinement is developed to improve the temporal performance of detectors.Finally,combined with visual restoration,an accurate and stable underwater robotic visual perception framework is established.Small-overlap suppression is proposed to extend video object detection(VID)methods to a single-object tracking task,leading to the flexibility to switch between detection and tracking.Extensive experiments were conducted on the ImageNet VID dataset and real-world robotic tasks to verify the correctness of our analysis and the superiority of our proposed approaches.The codes are available at https://github.com/yrqs/VisPerception.展开更多
Based on the status quantity of water resources in Xi′an region, a commentary on many math models is given for water resources operation and management, and the visual model is provided to solve practical problems. ...Based on the status quantity of water resources in Xi′an region, a commentary on many math models is given for water resources operation and management, and the visual model is provided to solve practical problems. In this model, the information for decision is visible in GIS (Geographic Information Systems) and topological figures. With object orientation methods, the objects are described in proprieties, methods, relations and time periods. The simulation process of the model is developed with Delphi and MapInfo, and the real decision scheme could be examined and practice decision process can be simulated from which. Decision analysis conducted from visual conditions is believable. The exploration to visual model is a beginning of practice research, much more study of which still needs to do.展开更多
机场飞行区现使用的场面监视方法存在着定位偏差较大、不稳定、易跳变、皆为点源定位等问题。针对这些问题,设计了基于视觉图像的飞行区监视方法,实现快速准确的目标检测和轮廓定位,使飞行区监视更加稳定精确。提出了一种基于MobileNetV...机场飞行区现使用的场面监视方法存在着定位偏差较大、不稳定、易跳变、皆为点源定位等问题。针对这些问题,设计了基于视觉图像的飞行区监视方法,实现快速准确的目标检测和轮廓定位,使飞行区监视更加稳定精确。提出了一种基于MobileNetV3和YOLOv5的网络模型(以下称为MobileNetV3-YOLOv5),即在YOLOv5的主干中使用MobileNetV3,来提高对目标的检测速度和准确度;提出了一种基于优化特征点提取的改进定向快速旋转简报(Oriented FAST and Rotated BRIEF,ORB)算法,将图像分割成多个区域,分别提取每个区域的特征点,从而提高目标识别框内区域的特征点识别数量,再进行特征点聚类筛选,最后根据识别目标类型采用最小包围盒进行轮廓划分,得到目标的轮廓定位。试验结果表明:MobileNetV3-YOLOv5方法对比原始YOLOv5模型,在识别目标准确率方面提升5百分点,在效率方面提升14张/s;同时在0~60 m的范围内,轮廓估计误差仅为2.9%;体现了所提出的监视方法的有效性,可以提升飞行区监视定位准确性和运行安全性。展开更多
基金Project supported by the National Natural Science Foundation of China(Nos.61633004,61725305,and 62073196)the S&T Program of Hebei Province,China(No.F2020203037)。
文摘Underwater robotic operation usually requires visual perception(e.g.,object detection and tracking),but underwater scenes have poor visual quality and represent a special domain which can affect the accuracy of visual perception.In addition,detection continuity and stability are important for robotic perception,but the commonly used static accuracy based evaluation(i.e.,average precision)is insufficient to reflect detector performance across time.In response to these two problems,we present a design for a novel robotic visual perception framework.First,we generally investigate the relationship between a quality-diverse data domain and visual restoration in detection performance.As a result,although domain quality has an ignorable effect on within-domain detection accuracy,visual restoration is beneficial to detection in real sea scenarios by reducing the domain shift.Moreover,non-reference assessments are proposed for detection continuity and stability based on object tracklets.Further,online tracklet refinement is developed to improve the temporal performance of detectors.Finally,combined with visual restoration,an accurate and stable underwater robotic visual perception framework is established.Small-overlap suppression is proposed to extend video object detection(VID)methods to a single-object tracking task,leading to the flexibility to switch between detection and tracking.Extensive experiments were conducted on the ImageNet VID dataset and real-world robotic tasks to verify the correctness of our analysis and the superiority of our proposed approaches.The codes are available at https://github.com/yrqs/VisPerception.
基金″Ninth-Five-Year Plan″for Science and Technology in China(95 /91 2 /0 5 /0 2 )
文摘Based on the status quantity of water resources in Xi′an region, a commentary on many math models is given for water resources operation and management, and the visual model is provided to solve practical problems. In this model, the information for decision is visible in GIS (Geographic Information Systems) and topological figures. With object orientation methods, the objects are described in proprieties, methods, relations and time periods. The simulation process of the model is developed with Delphi and MapInfo, and the real decision scheme could be examined and practice decision process can be simulated from which. Decision analysis conducted from visual conditions is believable. The exploration to visual model is a beginning of practice research, much more study of which still needs to do.
文摘机场飞行区现使用的场面监视方法存在着定位偏差较大、不稳定、易跳变、皆为点源定位等问题。针对这些问题,设计了基于视觉图像的飞行区监视方法,实现快速准确的目标检测和轮廓定位,使飞行区监视更加稳定精确。提出了一种基于MobileNetV3和YOLOv5的网络模型(以下称为MobileNetV3-YOLOv5),即在YOLOv5的主干中使用MobileNetV3,来提高对目标的检测速度和准确度;提出了一种基于优化特征点提取的改进定向快速旋转简报(Oriented FAST and Rotated BRIEF,ORB)算法,将图像分割成多个区域,分别提取每个区域的特征点,从而提高目标识别框内区域的特征点识别数量,再进行特征点聚类筛选,最后根据识别目标类型采用最小包围盒进行轮廓划分,得到目标的轮廓定位。试验结果表明:MobileNetV3-YOLOv5方法对比原始YOLOv5模型,在识别目标准确率方面提升5百分点,在效率方面提升14张/s;同时在0~60 m的范围内,轮廓估计误差仅为2.9%;体现了所提出的监视方法的有效性,可以提升飞行区监视定位准确性和运行安全性。
