基于多测量融合的粒子滤波跟踪算法

复杂背景下运动目标的可靠跟踪,是计算机视觉领域中一个极具挑战性的问题。提出了一种融合颜色和纹理信息的粒子滤波跟踪算法,在粒子滤波的测量阶段,使用颜色直方图对目标进行颜色描述,用梯度方向向量对目标进行纹理描述。对这两种信息,分别用Bhattacharyya系数和欧几里德距离比较粒子与参考模板的相似性。为解决目标变化和遮挡问题,采用了模板更新策略。实验结果表明该方法是稳健的,能够在复杂的背景下对运动目标进行有效、可靠的跟踪。     

Holmes’ front: constructing a new face of battle for America’s Civil War

ABSTRACT

In his seminal study of the changing nature of warfare between Agincourt and the Somme, military historian John Keegan proposed that future historians might consider combatants’ emotions in their assessments of the impact and nature of conflict. Recent years have witnessed the emergence of the history of emotions as an analytical approach, but rarely, if ever, is this directed toward the study of military history, far less the history of insurgencies and counter-insurgencies. This paper examines America’s civil war (1861–1865) as a case study of the ways in which an emotional history approach might illuminate not the physical experiences of but rather the immediate and longer-term reactions to counter-insurgency conflict through a focus on one specific individual, the future Supreme Court Justice Oliver Wendell Holmes, Jr. It proposes that Holmes, whilst not a man of the ranks, nevertheless can reveal the wider ramifications of civil war and its emotional impact, both individually and culturally. As a relatively limited internecine war, one not fought by professional armies but by volunteer forces, America’s civil war highlights the ways in which the soldier’s response points us toward the kind of emotional revolution that has, to date, mainly been located within the European nations.