Abstract
In foggy weather, contrast and color of images captured by traffic visual surveillance system are degraded significantly. To improve contrast and enlarge gray range of fog-degraded image, an improved filter variable multi-scale Retinex algorithm is presented. From comparing simulation resuts by four algorithms (global histogram equalization, local histogram equalization, multiscale retinex respectively variable filter retinex), the subjective and objective evaluation is summarized respectively. Experimental results show that the method has good universality and practicality to enhance local texture details of foggy image and suppress noise effectively.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
