In order to improve the reconstruction precision of the temperature field, we propose a temperature field reconstruction algorithm based on the firefly algorithm. The location of the firefly is updated through their behaviors, such as attraction and movement. When the optimal location of the firefly group is determined, the temperature value of each pixel center in the temperature field can be confirmed to reconstruct the temperature field of the tested regions. We reconstruct a one-peak temperature field via simulations, and the reconstruction effect is measured by  maximum absolute value error, mean absolute error and root mean square error. Compared with traditional iterative algorithms, the proposed algorithm has higher precision.

 ,