An UAV tracking method based on multi-feature fusion and scale adaptive is proposed.UAV has characteristics of small appearance and fast speed

so it’s easy to occur rotation

shielding

light changes and other problems in the process of movement.An improvement of Kernel correlation filtering（KCF）tracking algorithm for fast moving objects is proposed.Multi-feature fusion strategy is introduced to fuse the complementary histogram of oriented gradients（HOG）and color histogram

and add idea of image pyramid

31 kinds of scales are provided for matching to achieve accurate tracking.UAV flight sequence are collected for experimental verification

and compared with other three tracking algorithms.The experimental results show that the success rate and accuracy of this method are prior to other methods.The area under the maximum curve reaches 0.561 and 0.707

respectively

the mothed can effectively track UAV.