CAJ | 학술논문

针对复杂环境下非刚体目标轮廓跟踪存在跟踪失败的问题，提出一种基于在线学习的Snake模型及其轮廓跟踪算法。利用跟踪-学习-检测( TLD)机制实现目标快速跟踪，通过跟踪结果在线更新Snake模型约束，进而提高目标轮廓跟踪的准确性。初始化阶段，在GrabCut算法的基础上，将待跟踪目标分成若干个子块，并在后续跟踪过程中，利用TLD实现各子目标的定位跟踪，形成目标的轮廓置信图。同时针对各子目标提取特征，产生正负样本，更新各子目标跟踪模型。应用置信图建立参数化Snake模型的约束条件，进而得到目标轮廓。实验结果表明，该算法能适应光暗变化与较为复杂坏境下的跟踪，并获得精确的轮廓。
침대복잡배경하비강체목표륜곽근종존재근종실패적문제，제출일충기우재선학습적Snake모형급기륜곽근종산법。이용근종-학습-검측( TLD)궤제실현목표쾌속근종，통과근종결과재선경신Snake모형약속，진이제고목표륜곽근종적준학성。초시화계단，재GrabCut산법적기출상，장대근종목표분성약간개자괴，병재후속근종과정중，이용TLD실현각자목표적정위근종，형성목표적륜곽치신도。동시침대각자목표제취특정，산생정부양본，경신각자목표근종모형。응용치신도건립삼수화Snake모형적약속조건，진이득도목표륜곽。실험결과표명，해산법능괄응광암변화여교위복잡배경하적근종，병획득정학적륜곽。
For non-rigid target contour tracking in a complicated environment has tracking failure problems,this paper proposes a snake model and its contour tracking algorithm based on online learning. The algorithm utilizes the Tracking-Learning-Detection( TLD) mechanism to achieve the goal of fast tracking,and updates snake model constraints through the tracking results to improve the accuracy of the target contour tracking. In the phase of initialization,the target to be tracking is divided into several blocks on the basis of GrabCut algorithm, and the algorithm realizes the sub-targets locating and tracking by the use of TLD in the subsequent tracking process, which forms the confident map of target outline. At the same time,the algorithm produces positive and negative samples and updates each target tracking model for each target feature extraction. The constraint of parameterized snake model is built through confident map and the contour of target is obtained. Experimental results show that the algorithm can adapt to the changing light and dark,and even more complex tracking environment,and obtains precise contour.