Improved Staple target tracking method based on local sensitive histogram

• Main Authors: JIN LONGXU, DAI WEICONG, LI GUONING
• Format: Patent
• Language: Chinese; English
• Published: 21.02.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The embodiment of the invention provides an improved Staple target tracking method based on a local sensitive histogram. The improved Starle target tracking method based on the local sensitive histogram comprises the following steps: firstly, calculating a three-channel locality sensitive histogram of a grayscale image; respectively solving and extracting features through two ridge regression equations to train two tracking models; wherein the color classifier is trained by using a histogram extracted from each channel of a locally sensitive histogram; and finally, fusing the classification results of the two tracking models to obtain the position of the target in the current frame, so that the problem of non-ideal tracking effect on a single-channel grayscale image and an infrared image in an existing Staple target tracking method is effectively solved. 本发明实施例提供了一种基于局部敏感直方图改进的Staple目标跟踪方法。该基于局部敏感直方图改进的Staple目标跟踪方法首先计算灰度图像的3通道局部敏感直方图，再通过两个岭回归方程分别求解并提取特征训练两个跟踪模型，其中颜色分类器使用局部敏感直方图的每个通道上提取的直方图训练，最后融合两个跟踪模型的分类结果获得