Dark channel prior image dehazing method based on multiscale fusion

Abstract

The invention discloses a dark channel prior image dehazing method based on multiscale fusion. The dark channel prior image dehazing method based on multiscale fusion comprises the steps that (1) minimum value filter is conducted on a fog-degraded image through a color channel with a neighborhood size of 1*1 and a color channel with a neighborhood size of 15*15, so that corresponding dark channel images are obtained, and an estimated point transmissivity graph and an estimated block transmissivity graph are calculated based on the dark channel images; (2) edge image smoothness protection based on L0 gradient minimization is conducted on the estimated point transmissivity graph; (3) large-size Gaussian filter is conducted on the estimated block transmissivity graph; (4) with a normalized gradient magnitude as the weight, the estimated point transmissivity graph and the estimated block transmissivity graph are fused to enable a corrected transmissivity graph to be obtained; (5) the lower bound of the transmissivity is estimated based on median filtering; (6) the sky brightness is estimated by means of a 15*15 dark channel image; (7) inverse solution is conducted on a fog-degraded model, and a dehazing result is output. The dark channel prior image dehazing method based on multiscale fusion has the advantages that the dehazing capacity is high, the complexity is low, the image dehazing quality and the image dehazing efficiency can be improved.

Description

technical field [0001] The invention mainly relates to the field of image processing, in particular to a multi-scale fusion-based dark channel prior image defogging method suitable for degraded image enhancement processing in foggy weather. Background technique [0002] Under the influence of fog, images acquired by image acquisition equipment often have problems such as weakened contrast, color degradation, and loss of details, resulting in a sharp decline in image visibility. If it is directly applied to computer vision systems (such as road monitoring, ship navigation, etc.) In the process, it will inevitably have an adverse effect on the robustness and accuracy of the system. In some environments, when only a single degraded image can be obtained, the lack of information makes defogging an ill-conditioned equation solution problem. The existing dehazing enhancement algorithms often add effective constraints based on certain strong prior knowledge and appropriate assumpt...

AI Technical Summary

Problems solved by technology

[0013] (1) Interpolation refinement is performed based on block estimation of a certain scale, and the transmittance estimation of the small-area depth mutation area with foreground interference occlusion is inaccurate, resulting in incomplete defogging;

[0014] (2) In the process of Soft Matting, large-scale sparse linear equations need to be solved, the time complexity and space complexity are huge, and overshoot distortion may also appear at the edge of the sudden change of depth of field;

[0015] (3) The transmittance map obtained by using Soft Matting contains a large number of redundant texture details that do not reflect depth information. Substituting the degradation model for inverse solution will cause the enhancement result to lose texture details

[0016] (4) The lower bound of the transmittance is considered to be a fixed value, and the adaptive difference of the degraded image under different fogging conditions

However, this method requires additional reference (guiding) images, and other problems in the classical method are still not fully resolved

Images