MISSILEBORNE IMAGE DEFOGGING METHOD BASED ON SUPER-PIXEL SEGMENTATION AND DUAL CHANNEL FUSION

Aimed at the problem that the dark channel prior dehazing algorithm has poor adaptability in the missile image, a missile image dehazing method based on super-pixel segmentation and dual-channel fusion is proposed. The haze image was preprocessed. The super-pixel segmentation was performed to obtain improved bright and dark channels. The transmittance was estimated by a dual-channel fusion algorithm. A fast weighted guided filtering algorithm was proposed to refine the transmittance. The light value was obtained to obtain the dehazed image. Its brightness and saturation were adjusted. The simulation results show that the proposed algorithm can accurately estimate the atmospheric light value and transmittance of the fog-loaded image, which effectively avoids the problem of color distortion after defogging the bright area. The dehazing images have clear details, good scene contrast restoration, and strong visibility, which is superior to the contrast algorithm in comprehensive evaluation.