The document proposes an efficient algorithm to enhance visibility in road scene images captured during inclement weather. It uses a hybrid dark channel prior technique along with color analysis to estimate atmospheric light and transmission map. This is then used along with 3D geometric models to recover scene radiance and remove haze. The proposed method accurately estimates atmospheric light position and improves visibility while being computationally efficient compared to existing techniques. It has applications in intelligent transportation systems for traffic surveillance and vehicle detection/tracking.
Seal of Good Local Governance (SGLG) 2024Final.pptx
An Efficient Visibility Enhancement Algorithm for Road Scenes Captured by Intelligent Transportation Systems
1. An Efficient Visibility Enhancement
Algorithm for Road Scenes Captured by
Intelligent Transportation Systems
2. Abstract
• The visibility of images of outdoor road scenes will
generally become degraded when captured during
inclement weather conditions.
• The performance of the proposed method has been
proven through quantitative and qualitative
evaluations. Experimental results demonstrate that
the proposed haze removal technique can more
effectively recover scene radiance while demanding
fewer computational costs than traditional state-of-
the-art haze removal techniques.
3. Introduction
• Visibility in road images can be degraded due to
natural atmospheric phenomena such as haze, fog,
and sandstorms.
• This visibility degradation is due to the absorption
and scattering of light by atmospheric particles.
• Road image degradation can cause problems for
intelligent transportation systems such as traveling
vehicle data recorders and traffic surveillance
systems, which must operate under a wide range of
weather conditions.
6. One Image + Depth + Texture
• Assumes textures of the scene are
• Given (From Satellite Or Aerial Photos)
• Requires User Interaction To Align The 3D Model With The Scene
• Very Accurate Results
8. Optical Model
• In computer vision and pattern analysis, the
optical model is widely used to describe the
digital camera information of a hazy image
under realistic atmospheric conditions in the
RGB color space as
10. Haze Removal Using Dark Channel
Prior
• The dark channel prior is a state-of the- art image
restoration technique by which to remove haze from a
single image. In order to estimate the amount of haze
in an image, dark channel J dark can be expressed as
11. Statistics of the dark channel
( )
dark
J x
ur r
Except for the sky region, the intensity of is low and tends to be
zero
14. Disadvantages of Existing systems
• Low image quality.
• It not estimate the position of the atmospheric
light .
• In these not used in 3D geometric model .
15. Proposed system
• In this we propose an efficient visibility
enhancement algorithm Road Scenes Captured
17. HDCP Module
• Hybrid dark channel prior (HDCP)
• HDCP module can produce a restored image
that is not underexposed by using a procedure
based on the dark channel prior technique.
• Equation can be rewritten via the HDCP as
19. Estimating the Transmission Map
• In a single hazy image, these dark
channel values can provide a direct and
accurate estimation of haze transmission
and atmospheric light Ac in the RGB
color space as
21. To estimate of atmospheric light
Pick the top 0.1% brightest pixels in the dark channel
22. Color Analysis Module
• The particles of sand in the atmosphere caused
by sandstorms absorb specific portions of the
color spectrum.
• This phenomenon leads to color shifts in images
captured during such conditions, resulting in
different color channel distributions.
23. • Each RGB color channel for a typical image,
which is described as
24. • Each RGB color channel in order to avoid
color shifts in the restored image. This can be
measured as
25. Visibility Recovery module
• In order to produce a high-quality haze-free
image captured in different environments, we
combine the information provided via the
HDCP and CA modules to effectively recover
the scene radiance.
26. Advantages
• Accurately estimate the position of atmospheric light.
• The use of the proposed method is quantitative and
qualitative evaluation of intelligent transportation
systems
• It improves visibility in hazy images.
• It accurately estimates the position of atmospheric light.
• The proposed Technique is produced satisfactory
restore image