Self-thermal radiation compensation for autofocus algorithm in infrared optical system

Self-thermal radiation is a key factor of reducing image quality of infrared optical system and can cause the autofocus process inefficient. A study on infrared images affected by self-thermal radiation arising from temperatures of system under test was presented to optimize autofocus algorithm. Experimental results showed that conventional autofocus functions like Gray-level Variance(GV), Laplacian of Gaussian (LoG) Operator and Discrete Cosine Transform(DCT) methods had low accuracy and sensitivity for determining the optimal imaging plane from a series of infrared images, when the optical system worked in high temperature environments. Integrating self-thermal radiation compensation algorithm to correct the image data, we put forward a new autofocus algorithm. The autofocus function was based on wavelet packet transform to extract high-frequency information from infrared images, and performances of the new algorithm were proved by Athermalization-Efficiency Assessment System.