期刊文献+
共找到3篇文章
< 1 >
每页显示 20 50 100
基于遮挡标记的目标检测算法
1
作者 帖军 宋威 +2 位作者 尹帆 郑禄 杨欣 《中南民族大学学报(自然科学版)》 CAS 2020年第3期302-308,共7页
为解决自然环境下被遮挡目标检测精度低的问题,提出一种基于遮挡标记的目标检测算法,此方法对图像数据进行标注,分析被遮挡目标的图像特征损失,提出遮挡补偿机制;基于YOLO V3模型构建二分类器对目标和背景进行区分,从而完成目标检测任务... 为解决自然环境下被遮挡目标检测精度低的问题,提出一种基于遮挡标记的目标检测算法,此方法对图像数据进行标注,分析被遮挡目标的图像特征损失,提出遮挡补偿机制;基于YOLO V3模型构建二分类器对目标和背景进行区分,从而完成目标检测任务.实验结果表明:改进后的YOLO V3模型在不同的遮挡目标个数占比下,平均检测精度均明显高于传统的YOLO V3模型,在自然环境下检测被遮挡目标,能达到更好的效果. 展开更多
关键词 目标检测 遮挡标记 补偿机制
下载PDF
Bounding box extraction from spherical hologram of elementary object to synthesize hologram of arbitrary three-dimensional scene with occlusion consideration(Invited Paper) 被引量:1
2
作者 Jae-Hyeung Park Hong-Gi Lim 《Chinese Optics Letters》 SCIE EI CAS CSCD 2014年第6期86-90,共5页
A novel method to extract a bounding box that contains the three-dimensional object from its spherical hologram is proposed. The proposed method uses the windowed Fourier transform to obtain the angular distribution o... A novel method to extract a bounding box that contains the three-dimensional object from its spherical hologram is proposed. The proposed method uses the windowed Fourier transform to obtain the angular distribution of the quasi-collimated beams at each position in the spherical hologram and estimates the bounding box by accumulating the quasi-collimated beams in the volume inside the spherical hologram. The estimated bounding box is then used to realize occlusion effect between the objects in the synthesis of the three-dimensional scene hologram. 展开更多
关键词 Bounding box extraction from spherical hologram of elementary object to synthesize hologram of arbitrary three-dimensional scene with occlusion consideration
原文传递
Multi-Object Tracking Strategy of Autonomous Vehicle Using Modified Unscented Kalman Filter and Reference Point Switching
3
作者 WANG Muyuan WU Xiaodong 《Journal of Shanghai Jiaotong university(Science)》 EI 2021年第5期607-614,共8页
In this study,a multi-object tracking(MOT)scheme based on a light detection and ranging sensor was proposed to overcome imprecise velocity observations in object occlusion scenarios.By applying real-time velocity esti... In this study,a multi-object tracking(MOT)scheme based on a light detection and ranging sensor was proposed to overcome imprecise velocity observations in object occlusion scenarios.By applying real-time velocity estimation,a modified unscented Kalman filter(UKF)was proposed for the state estimation of a target object.The proposed method can reduce the calculation cost by obviating unscented transformations.Additionally,combined with the advantages of a two-reference-point selection scheme based on a center point and a corner point,a reference point switching approach was introduced to improve tracking accuracy and consistency.The state estimation capability of the proposed UKF was verified by comparing it with the standard UKF in single-target tracking simulations.Moreover,the performance of the proposed MOT system was evaluated using real traffic datasets. 展开更多
关键词 multi-object tracking(MOT) light detection and ranging(LiDAR)sensor unscented Kalman filter(UKF) object occlusion
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部