期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
基于目标性权值度量的多示例学习目标跟踪 被引量:1
1
作者 滑维鑫 慕德俊 +1 位作者 郭达伟 刘航 《北京航空航天大学学报》 EI CAS CSCD 北大核心 2017年第7期1364-1372,共9页
针对多示例学习(MIL)跟踪算法在包概率计算过程中对示例样本不加以区分导致分类器性能下降,及采用最大化似然函数选择相应的弱分类构造强分类增加了算法复杂度的问题,提出了一种基于目标性权值学习的多示例目标跟踪算法,该算法利用目标... 针对多示例学习(MIL)跟踪算法在包概率计算过程中对示例样本不加以区分导致分类器性能下降,及采用最大化似然函数选择相应的弱分类构造强分类增加了算法复杂度的问题,提出了一种基于目标性权值学习的多示例目标跟踪算法,该算法利用目标性测量每个示例样本对包概率的重要性,根据其目标性测量结果对每个正示例样本赋予相应的权值,从而判别性地计算包概率,提高跟踪精度。同时在弱分类器选择过程中,采用最大化弱分类器与似然函数概率内积的方法从弱分类器池中选择弱分器构造强分类器,减少算法的计算复杂度。通过对不同复杂场景下视频序列的跟踪,实验结果表明,本文所提出的目标性权值学习的多示例目标跟踪算法优于其对比算法,表现出较好的跟踪精度和鲁棒性能。 展开更多
关键词 多示例学习(MIL) 目标性测量 弱分类器选择 包概率计算 目标跟踪
下载PDF
A FRAMEWORK TO MEASURE CLASS COHESION
2
作者 Xu Baowen Chen Zhenqiang Zhou Yuming(Department of Computer Science and Engineering, Southeast University, Nanjing 210096) (State Key Laboratory of Software Engineering, Wuhan University, Wuhan 430072) 《Journal of Electronics(China)》 2003年第1期69-73,共5页
Classes are the basic modules in Object-Oriented (OO) software, which consist of attributes and methods. Thus, in OO environment, the cohesion is mainly about how tightly the attributes and methods of classes cohere w... Classes are the basic modules in Object-Oriented (OO) software, which consist of attributes and methods. Thus, in OO environment, the cohesion is mainly about how tightly the attributes and methods of classes cohere with each other. This letter discusses the relationships between attributes and attributes, attributes and methods, methods and methods of a class,and the properties of these relationships. Based on these properties, the letter proposes a new framework to measure the cohesion of a class. The approach overcomes the limitations of previous class cohesion measures, which consider only one or two of the three relationships in a class. 展开更多
关键词 Object-Orientation(OO) Software measurement Dependence analysis COHESION
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部