LAUSR

Since traditional target tracking algorithms employ artificial features, they are not robust enough to describe the appearance of a target. Therefore, it is difficult to apply them to complex scenes. Moreover, the traditional target tracking algorithms do not measure the confidence level of the response. When the confidence level is low, the appearance model of the target is easily disturbed, and the tracking performance is degraded. This paper proposes the Multiple Features and Average Peak Correlation Energy (MFAPCE) tracking algorithm. The MFAPCE tracking algorithm combines deep features with color features and uses average peak correlation energy to measure confidence level. The algorithm uses multiple convolution layers and color histogram features to describe the target appearance. The response is obtained by optimizing the context information using a correlation filter framework. The average peak correlation energy is used to determine the final confidence level of the response and thus determines whether to update the model. The experiments showed that the MFAPCE algorithm improves the tracking performance compared with traditional tracking algorithms.