PROJECT TITLE :

Lower Bound on Transmission Using Non-Linear Bounding Function in Single Image Dehazing

ABSTRACT:

Scattered light from air particles reduces the clarity of a photograph taken in poor weather (e.g. haze, fog, mist, smog). One fuzzy image can be restored to its original clarity using a single image dehazing method (SID). The SID is a difficult problem because of its ambiguity. SID is typically solved using the atmospheric scattering model (ATSM). ATSM's transmission and ambient light are two of its most important metrics. SID's accuracy and effectiveness are dependent on the precision of the transmission and atmospheric light values used in the SID process. For transmission estimates, this method calculates the least colour channel difference between a clear and a hazy image. To limit dehazing reconstruction error, the translation problem provides a lower bound on transmission. Bounding function (BF) and a quality control parameter determine the bottom bound. In order to accurately assess transmission, a non-linear model is developed. Using the proposed quality control parameter, dehazing can be fine-tuned. Dehazing methods currently in use are compared to the accuracy of the suggested approach for transmission. Dehazed photos can be compared visually, and the proposed method's efficiency can be confirmed by objective evaluations.