In order to solve the problem that the conventional tracker is not adapted to the abrupt motion, a tracking algorithm based on the improved salp swarm algorithm (ISSA) was proposed. Visual tracking is considered to be a process of locating the optimal position through the interaction between leaders and followers in successive images. Firstly, the adaptive mechanism of leader and follower is introduced into the original salp swarm algorithm (SSA) to balance the exploitation and exploration of the algorithm. This method can improve the accuracy and effect of tracking. Secondly, the golden‐sine algorithm was used to update the position of followers, considering that the SSA had a single spatial search mode for followers and was easy to fall into the local optimum. By comparing with 19 classical tracking algorithms, qualitative and quantitative analysis is carried out to verify the tracking effect of the proposed method. A large number of experimental results show that the algorithm proposed here has good performance in visual tracking, especially for mutation motion tracking.

Salp Swarm Algorithm is a new intelligent optimization algorithm. Because of it is fewer control parameters and convenient operation, it has attracted the attention of researchers from all circles. However, due to the lack of complex iterative process, it has some disadvantages, such as low optimization precision and poor population diversity in the late iteration. To solve these problems of Salp Swarm Algorithm, we proposed a Salp Swarm Algorithm based on mutual learning mechanism. In this article, the improved Salp Swarm Algorithm uses the iteration factor of tangent change to update the population position, which balances the global exploration and local development ability of the algorithm. At the same time, the introduction of mutual learning mechanism in the local development stage solves the problem of poor population diversity in the later iteration of Salp Swarm Algorithm, and improves the convergence accuracy of the algorithm. Finally, 23 classical and CEC2014 benchmark functions are used to evaluate the effectiveness of the proposed algorithm. The experimental results show that the improved Salp Swarm Algorithm has better optimization accuracy and stability compared with the algorithm of Salp Swarm, Moth Flame Optimization, Grasshopper Optimization, and Ant Lion Optimization.

Kernel Correlation Filter (KCF) tracker has shown great potential on precision, robustness and efficiency. However, the candidate region used to train the correlation filter is fixed, so tracking is difficult when the target escapes from the search window due to fast motion. In this paper, an improved KCF is put forward for long-term tracking. At first, the moth-flame optimization (MFO) algorithm is introduced into tracking to search for lost target. Then, the candidate sample strategy of KCF tracking method is adjusted by MFO algorithm to make it has the capability of fast motion tracking. Finally, we use the conservative learning correlation filter to judge the moving state of the target, and combine the improved KCF tracker to form a unified tracking framework. The proposed algorithm is tested on a self-made dataset benchmark. Moreover, our method obtains scores for both the distance precision plot (0.891 and 0.842) and overlap success plots (0.631 and 0.601) on the OTB-2013 and OTB-2015 data sets, respectively. The results demonstrate the feasibility and effectiveness compared with the state-of-the-art methods, especially in dealing with fast or uncertain motion.