Aiming at the problem of tracking fail ure caused by fuzzy and scale change,an adaptive scale KCF tracking algorithm wi th anti-fuzzy properties is proposed.Firstly,an anti-fuzzy feature detection a nd matching model is constructed.The model masks the image to be matched and extracts the region of interest of the target.Then,SURF and Brisk descriptors are used to detect and describe the feature points in the targ et area.The target position in the fuzzy image is obtained by twice matching the feature points.Secondly,the current imag e definition is judged.If the image clarity is higher than a certain level,the traditional KCF algorithm is used to predict the object position,and the optimal object scale is obtained by comparing the object response peak under different s cales.Otherwise,the anti-fuzzy feature detection and matching model is used to obtain the target position.Finally,the p erformance of the proposed algorithm is tested in OTB-2015dataset.The accuracy of the proposed algorithm is 85.2%, 17.4% higher than the traditional KCF algorithm,and the success rate is 77.8%,which is 23% higher than the traditi onal KCF algorithm.Experimental results show that the proposed algorithm can adaptively change the size of the t racking frame in the tracking process,and can effectively track the target in th e fuzzy image.