Polarimetric dehazing method for dense haze removal based on distribution analysis of angle of polarization

Abstract

Many dehazing methods have proven to be effective in removing haze out of the hazy image, but few of them are adaptive in handling the dense haze. In this paper, based on the angle of polarization (AOP) distribution analysis we propose a kind of polarimetric dehazing method, which is verified to be capable of enhancing the contrast and the range of visibility of images taken in dense haze substantially. It is found that the estimating precision of the intensity of airlight is a key factor which determines the dehazing quality, and fortunately our method involves a high precision estimation inherently. In the experiments a good dehazing performance is demonstrated, especially for dense haze removal. We find that the visibility can be enhanced at least 74%. Besides, the method can be used not only in dense haze but also in severe sea fog.

Generalized Polarimetric Dehazing Method Based on Low-Pass Filtering in Frequency Domain

“…This experimental result illustrates the effectiveness of the proposed polarimetric dehazing method. As a comparison, Figure 4d is the reconstructed image using our previously proposed polarimetric dehazing method [12]. Both methods can reconstruct high-quality dehazed images, but it is necessary to tune the bias parameters manually in our previous method.…”

“…From Figure 4b, it can be seen that the nearer scene (trees) has a brighter value, while the farther scene (building) has a relatively darker value. (c) the dehazed image using our proposed polarimetric dehazing method; (d) the dehazed image using the dehazing method proposed in [12]; (e) the dehazed image using the dehazing method proposed by Y. Schechner [24].…”

“…From Equation (6), it can be seen that the DoP of the airlight can only be calculated from the sky area, where there is no direct light component in S 0 . However, the calculated DoP of the sky area is also severely influenced by the noise [12]. In the frequency domain of hazy images, the airlight is mostly located in the low-frequency region, while the direct light and noise are usually located in the high-frequency region.…”

“…These techniques have also been combined together, such as polarimetric imaging with image processing [9,10] and polarimetric imaging with image fusion [11], to further improve the performance. These techniques can effectively enhance the contrast of hazy images captured in scattering media, and even increase the imaging distance [2,12]. Among these dehazing techniques, image-processing techniques typically need only one hazy image to do the dehazing process [13][14][15].…”

“…As a result, the DoP of the scattering light usually cannot be estimated accurately, and the quality of the dehazed images is unstable accordingly. 2 of 12 To overcome this problem, the DoP of the scattering light is usually tuned by a bias parameter to improve the quality of the dehazed image [5,12,18]. However, this bias parameter is affected by the camera itself and is related to different scattering media, so it is not a fixed value and must be tuned manually in the dehazing process.…”