摘要
以治安视频监控中电动摩托车的管理需求为背景,提出了一种基于可变形部件模型(DPM)的电动摩托车目标检测方法。结合可变形部件模型多尺度、高效率的优势,实现了电动摩托车目标检测算法。构建了包含1006个正样本和1000个负样本的训练集、测试集样本库,完成了定量的实验测定。实验结果表明,新提出的算法可以在监控视频图像中高效、准确地检测出电动摩托车目标。
With the background of the management needs of scooters in security video surveillance,a target detection method for scooters based on Deformable Part Model(DPM)is proposed.The target detection algorithm for scooters is implemented by combining the advantages of multi-scale and high efficiency of the deformable part model.The training set and test set sample libraries containing 1006 positive samples and 1000 negative samples are constructed,and quantitative experimental measurements are completed.The experimental results show that the newly proposed algorithm can detect scooters targets in surveillance video images with higher efficiency and accuracy.
作者
徐志佳
XU Zhijia(School of Mechanical Engineering,Guiyang University,Guiyang,Guizhou 550005,China)
出处
《自动化应用》
2025年第14期63-65,68,共4页
Automation Application
