CAJ | 학술논문

在采用背景减除法提取运动目标的基础上，利用人体的方向梯度直方图HOG 特征，采用线性支持向量机SVM分类器，检测出人体区域。采用粒子滤波方法，进行人体跟踪，研究遮挡问题。针对长时间遮挡问题进行分析，对方向梯度直方图HOG 特征结合粒子滤波的方法进行改进。通过可信度，来判断当前跟踪人体，决定是否进行目标关联，从而提高人体被遮挡情况下跟踪的准确性。
재채용배경감제법제취운동목표적기출상，이용인체적방향제도직방도HOG 특정，채용선성지지향량궤SVM분류기，검측출인체구역。채용입자려파방법，진행인체근종，연구차당문제。침대장시간차당문제진행분석，대방향제도직방도HOG 특정결합입자려파적방법진행개진。통과가신도，래판단당전근종인체，결정시부진행목표관련，종이제고인체피차당정황하근종적준학성。
Based upon the background subtraction method to extract a moving target, the human area is detected by using human feature for histogram of orientation gradient (HOG) and a linear support vector machine (SVM) classifier. The method of particle filter is used to track human and to research occlusion problem. By analyzing the problem of a long occlusion, the histogram of orientation gradient (HOG) feature combined with the particle filtering method has been improved. The current tracking human target is judged by credibility. It is decided whether to carry out the goals associating. The accuracy of tracking is improved.