Laser weapons and laser three-dimensional imaging require high precision in target tracking, at the same time, it also need to face the situation where the target is interfered by the laser. To meet this need, in this paper, we propose a high-precision eccentric tracking algorithm based on kernel correlation filtering, which can be used to solve the problem. We simulated on Visual Studio and CCS platforms and the result show that the Average of Center Location Error (ACLE) can reach 0.74 pixels. For comparison, the ACLE of Tracking algorithm Based on Template Matching (TBTM) is 3.7pixels. In addition, our algorithm can run at 178 Frames-per-Second, the fake code is provided in the paper (see Algorithm 1).