Digital Image Dehazing Using Low-Resolution Dark Channel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems in mobile work machines struggle with haze removal due to high computational requirements and processing times, making real-time implementation challenging during inclement weather conditions.
Innovation Solution
A system and method that downscale input digital images to low-resolution, determine a minimum intensity dark channel and atmospheric light value, and apply a transmission map to generate a de-hazed output image using a low-power GPU, reducing computational load and processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-end processing systems with high computational power are used for haze removal, then haze removal quality is improved, but device complexity and cost increase
Solution Approach 1:
The patent changes the resolution parameter of the input image from high-resolution to low-resolution before processing. This parameter change reduces the computational complexity of haze removal algorithms while maintaining acceptable visual quality, thereby resolving the contradiction between haze removal quality and device complexity
Solution Approach 2:
The patent creates a low-resolution copy of the original high-resolution image for processing. This copying approach allows the complex haze removal algorithm to operate on a simplified version, reducing computational requirements while the final output can be scaled back to original resolution if needed
2Measurement precision
If high computational power haze removal techniques are applied, then visibility improvement is enhanced, but processing time increases
Solution Approach 1:
The patent changes the resolution parameter to low-resolution, which directly reduces the number of pixels that need to be processed. This parameter modification enables real-time or near-real-time haze removal while still providing sufficient visibility improvement for practical applications
3Measurement precision
If conventional haze removal algorithms are used on high-resolution images, then de-hazing effectiveness is improved, but power consumption increases
Solution Approach 1:
The patent modifies the resolution parameter from high to low, which reduces the computational workload of haze removal algorithms. This parameter change directly lowers power consumption since fewer calculations are required, while still achieving acceptable de-hazing effectiveness for mobile work machine applications
Data Source
AI summary
A method for removing haziness in a digital image is provided. The method includes receiving an input digital image having some haze content from an image capturing device. The input digital image is downscaled to obtain a low-resolution image. Further, a minimum intensity dark channel is determined for each local patch of the low-resolution image to obtain a dark channel image corresponding to the low-resolution image. Furthermore, a transmission map of the low-resolution image is determined based on the dark channel image. Moreover, an atmospheric light value associated with the low-resolution image is also determined. The method further includes applying the determined transmission map and the atmospheric light value associated with the low-resolution image to the input digital image to generate a de-hazed output image and displaying the generated de-hazed output image on a display unit.