Digital Image Halo Correction Using Light Map Convolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Artifacts such as halos caused by light sources degrade the quality of digital images, particularly in three-dimensional scene modeling studios, leading to inhomogeneous brightness modifications and reduced contrast.
Innovation Solution
A method for correcting halos in digital images by generating a digital light intensity map, applying a convolution kernel specific to the shooting device, and removing the corrective value map pixel by pixel to obtain a halo-free image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If light sources are placed inside the shooting field to illuminate the scene, then the illumination intensity is improved, but the image quality deteriorates due to halo artifacts and inhomogeneous brightness modifications
Solution Approach 1:
The patent applies preliminary action by generating a light intensity map before image correction, predicting the halo effect that will occur during image capture. This allows the correction process to preemptively remove the predicted halo artifacts from the captured image, resolving the contradiction between using strong illumination and maintaining image quality.
Solution Approach 2:
The patent converts the harmful halo effect into a benefit by using the same light intensity map that characterizes the problem to generate a correction map. The halo artifact information is transformed into corrective data that, when applied through convolution, removes the unwanted effects while preserving the beneficial illumination.
2Reliability
If conventional visual corrections such as contrast enhancements are applied, then the artistic quality is improved, but the objective image accuracy deteriorates due to subjective manipulation
Solution Approach 1:
The patent implements feedback by using the actually captured light intensity map to generate the correction map, rather than applying predetermined or subjectively chosen corrections. The system measures the real halo effect in the captured image and uses this feedback to create an objective correction that restores image accuracy without arbitrary artistic manipulation.
3Measurement precision
If multiple shooting devices are used for three-dimensional scene modeling, then the modeling accuracy is improved, but the complexity of correcting halos across multiple images increases
Solution Approach 1:
The patent applies universality by creating a single light intensity map that characterizes the illumination across the entire multi-device setup. This universal map can be used to generate correction maps for multiple shooting devices, allowing the same correction approach to work across all devices and simplifying the overall correction process while maintaining high modeling accuracy.
Data Source
AI summary
A method for correcting a halo in a digital image captured using photogrammetry in a 3-D modeling studio, the halo being generated through the interaction of light originating from a light source in the studio with the optic of the shooting device, and manifesting as a local lightening of the digital image, the method comprising the steps of generating a light intensity map characterizing the light source in terms of spatial distribution and light intensity, providing a convolution kernel specific to the shooting device, calculating a convolution product of the light intensity map and the kernel to obtain a corrective value map, and removing the corrective value map from the digital image pixel by pixel to produce a corrected image in which the halo is not present.