Image Sensor Discontinuity-Preserving Smoothing for High Dynamic Range
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image sensors face challenges in capturing high dynamic range optical images due to signal saturation and clipping, leading to loss of details in both bright and dark areas, and struggle to accurately render these images on low dynamic range displays and prints, resulting in halo artifacts and reduced image quality.
Innovation Solution
An electronic image sensor with a discontinuity-preserving smoothing system, compensation system, and blending system that processes the image signal to produce a smooth image while preserving sharp discontinuities, compensating for illumination-induced variations, and blending the signals to reduce noise and dynamic range, thereby enhancing detail visibility in both bright and dark regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If conventional image sensors directly capture high dynamic range optical images, then the sensor can record the full range of light intensities, but signal saturation and clipping occur in both bright and dark areas causing loss of image details
Solution Approach 1:
The patent segments the captured light signal into multiple intensity ranges using a microlens array that directs different intensity ranges to different photodetectors. Bright signals are directed to one set of photodetectors while dark signals are directed to another set, allowing simultaneous capture of both highlights and shadows without saturation or clipping.
Solution Approach 2:
The patent adds a spatial dimension to the signal capture process by using a microlens array positioned between the lens and photodetector array. This microlens array creates multiple optical paths that spatially separate signals of different intensities before they reach the photodetectors, effectively adding a dimensional separation mechanism.
2Adaptability or versatility
If conventional image sensors attempt to capture both bright and dark regions simultaneously, then the sensor can record the full scene, but one region becomes saturated while the other loses detail
Solution Approach 1:
The patent applies local quality by assigning different photodetectors to different signal intensity ranges. Some photodetectors are optimized for capturing bright signals while others are optimized for dark signals, allowing each detector to operate in its optimal range and maintain high measurement precision across the entire scene.
3Ease of operation
If tone mapping is applied to convert high dynamic range images to low dynamic range displays, then the image can be displayed on standard devices, but naive signal scaling destroys details due to quantization and clipping
Solution Approach 1:
The patent performs preliminary action by capturing signals in multiple intensity ranges simultaneously at the sensor level before any tone mapping or display conversion is needed. This preliminary separation of signals ensures that detail information is preserved in the captured data, allowing for high-quality tone mapping later without loss of details.
Data Source
AI summary
The present invention provides an image processing method. In one embodiment, the method provides discontinuity-preserving image smoothing and segmentation, noise reduction, reduction of image variations particularly those variations caused by illumination conditions, exposure compensation, and dynamic range compression. The present invention also provides an electronic image sensor that is able to detect high dynamic range optical images and produce electronic images with reduced dynamic range.