Distance Estimation Device Using Focal-Depth Image Synthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional distance estimation methods using passive sensing techniques, such as depth-from-defocus, suffer from varying accuracy across different focal depths, especially when blurring is large, and require multiple images or complex optical systems, leading to reduced image quality and lost information.
Innovation Solution
A distance estimation device and method that obtain multiple focal-depth images and generate an all-in-focus image, then synthesize distance maps from these images to provide consistent accuracy across focal depths by selecting the most similar blurred images at each focal depth, thereby enhancing distance estimation precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If depth-from-defocus method is used for distance estimation, then distance can be calculated directly from images, but accuracy varies significantly when blurring is large
Solution Approach 1:
The patent dynamically adapts the blur parameter selection based on the actual blurring amount in the image. Instead of using a fixed blur parameter, the system calculates the actual blurring from the image data and selects or adjusts the blur parameter accordingly, making the distance estimation accurate across varying focal depths and blurring conditions
Solution Approach 2:
The patent changes the blur parameter based on the actual blurring amount detected in the image. By adjusting this critical parameter dynamically, the system maintains accurate distance estimation even when the amount of blurring varies significantly, resolving the accuracy consistency issue
2Measurement precision
If multiple images with different focal positions are captured for distance estimation, then distance information can be obtained, but image quality deteriorates and information is lost
Solution Approach 1:
The patent introduces an all-in-focus image as an intermediary element that contains distance information without requiring multiple focused images. This all-in-focus image serves as a mediator that preserves the original image quality while providing the necessary blur variations for accurate distance estimation through proper blur parameter selection
3Device complexity
If conventional distance estimation methods are used, then processing is simpler, but accuracy varies across different regions of the image
Solution Approach 1:
The patent applies local quality by selecting different blur parameters for different regions or conditions. Instead of using a uniform approach, the system adapts the blur parameter to match the local blurring characteristics, ensuring consistent accuracy across different image regions while maintaining relatively simple processing
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach ensures high and uniform distance estimation accuracy across all regions, reducing variations and maintaining image quality by synthesizing distance maps from multiple focal-depth images, thus overcoming the limitations of conventional methods.
Implementation Method 1
an all-in-focus image generation unit which generates an all-in-focus image focused in a focal range wider than each of a focal range of the first focal-depth image and a focal range of the second focal-depth image
Implementation Method 2
a first generation unit which generates, from the all-in-focus image, a plurality of blurred images which are focused at the first focal depth and blurred at the plurality of focal depths
Data Source
AI summary
A distance estimation device (1x) prevents the variation in distance estimation accuracy that has been a problem in distance estimation performed using an all-in-focus image and an arbitrary-focal-depth image. The distance estimation device can estimate a focal distance of any subject with a high degree of accuracy, and includes: a first generation unit (10w) generating a first distance map indicating a focal depth determined from an all-in-focus image and a first focal-depth image; a second generation unit (11w) generating a second distance map indicating a focal depth determined from the all-in-focus image and a second focal-depth image; and a distance map synthesis unit (14) synthesizing the generated first distance map and the generated second distance map to generate a synthetic distance map.