基于光强变化场景的目标检测与跟踪新方法

为在光强变化场景下实现目标的检测与跟踪，研究了一种新方法。该方法综合集成了Vibe算法与粒子滤波算法的优点，通过在改进Vibe算法中引入光照补偿模型、重构粒子滤波模型，解决了2种方法对光强变化的自适应性难题；在局部三值模式(LTP)中建立了一种自适应阈值方法，并采用线性回归分类方法，实现了目标跟踪。通过开发相应程序并将所提新方法的结果与标准结果进行对比验证，验证结果表明：新方法的偏差小于文中其他3个对照方法的相应偏差，该方法对光强变化场景下的目标检测与跟踪研究有一定帮助。

New method for target detection and tracking by changing illumination

For detecting and tracking the moving objects of interest by changing illumination, a new method is proposed. The proposed method combines improved Vibe algorithm with the particle filter algorithm by introducing illumination feeding model to Vibe algorithm, and reconstructing the particle filter to solve the changing illumination problem. In the phase of moving target tracking, Linear Regression Classifier(LRC) is introduced into the recognition algorithm, as well as the background information. The improved Local Ternary Pattern(LTP) is adopted to carry out feature extraction and recognition. Developing the program and carrying out some experiments, the experimental results demonstrate that the new method can assist detecting and tracking moving target with changing illumination study.