摘要
传统的NMS算法的过滤阈值是人为设定的,由于阈值的选取不当可能会造成漏检和误检。在应用NMS算法时,所有图像的最佳阈值不是完全相同的,根据图像自身信息的不同而发生变化。针对上述问题,提出基于F1值的非极大值抑制阈值自动选取方法,综合考虑检测算法的准确率与召回率,选取使F1值最高的最佳过滤阈值,构建映射关系。测试阶段,利用映射关系和图像信息自动选取对应的过滤阈值。实验结果表明,本文提出的改进版本NMS算法将检测精度mAP值提高了1.1%。与现有的先进算法做对比,证明了本文算法的有效性。
The filtering threshold of the traditional non-maximum suppression(NMS)algorithm is artificially set.However,the improper selection of the threshold may result in leak and error detection.When applying the NMS algorithm,the optimal threshold for all images differs because the information obtained from the image itself changes.Given the aforementioned problems,we propose an automatic selection method of the NMS threshold based on the F1 score,which comprehensively considers the accuracy and recall rates of the detection algorithm and selects the best filtering threshold based on the highest F1 score to establish a relationship map.Experimental results show that the improved version of the NMS algorithm proposed in this study enhances the detection accuracy mAP value by 1.1%.Compared with the existing advanced algorithms,the proposed algorithm has been proven to be more effective.
作者
王照国
张红云
苗夺谦
WANG Zhaoguo;ZHANG Hongyun;MIAO Duoqian(College of Computer Science and Technology,Tongji University,Shanghai 201804,China;Key Laboratory of Embedded System and Service Computing of Ministry of Education,Tongji University,Shanghai 201804,China)
出处
《智能系统学报》
CSCD
北大核心
2020年第5期1006-1012,共7页
CAAI Transactions on Intelligent Systems
基金
国家自然科学基金项目(61573255,61976158,61673301)
国家重点研发计划项目(213).
关键词
计算机视觉
目标检测
非极大值抑制算法
卷积神经网络
深度学习
检测框
F1值
自适应算法
computer vision
object detection
non-maximum suppression algorithm
convolutional neural network
deep learning
detection boxes
F1 value
self-adaptive algorithm