Video Quality Enhancement Using an Optimized Dehazing Algorithm

Abstract Video in open environments faces several problems such as the hazing problem. Hazing is produced due to light scattering from particles in the atmosphere around the scene to be imaged. In this paper, an optimized dehazing algorithm is modified by applying pre-processing on frames before dehazing process to remove noise and limited dynamic range because all frames contain some noise and limited dynamic range due to sensor measurement errors that can be amplified in the haze removal process if ignored. Different types of enhancement techniques are applied to remove noise and limited dynamic range before the dehazing process. Two video sequences are used: a Near Infrared (NIR) and a visible sequence. The hazing effect can be performed according to the hazing mathematical model. The effect of attenuation parameter α and attenuation weight ω on dehazed frames are investigated. The proposed algorithm proves good results in the enhancement of two types of video as shown in the spectral entropy of dehazed frames.