A new target recognition method based on local fuzzy threshold is presented to solve the problems of serious background interference

the absence of reference map for complex terrain object and low contrast between target and background. Firstly

multi-threshold algorithm is designed and saliency map is plot based on the establishment of multi-scale space. Secondly

Itti model is improved and candidate target filtering model is constructed by using local fuzzy threshold method based on image fuzzy rate. Lastly

the detected results are precisely matched using Integral Nprod to determine the right one. The results of the experiments indicate that the match rate of the algorithm increases nearly 20% compared with Hausdorff distance algorithm and time consumption decrease 75%

and increases nearly 40% compared with Integral Nprod algorithm and time consumption decreas 50%.The new algorithm has the advantages of high match rate

high speed and high accuracy for FLIR target in complicated background.