Stereo Depth Detection Using Parallax and Sharpness Blur Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image capturing systems face limitations in obtaining accurate depth information over a wide range of distances, with methods using parallax being restricted by sensor size and baseline length, and those using optical-wavelength differences being limited by the ratio of sharpness/blur, which fails to effectively determine depth at near and far distances.
Innovation Solution
An image capturing system comprising two modules with different optical-wavelength capabilities and a depth map processing device that generates depth maps based on parallax and sharpness/blur differences, allowing for a combined depth map to enhance the depth detection range and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If parallax method is used to obtain depth information, then depth information can be obtained, but the depth detection range is limited by baseline length and sensor size
Solution Approach 1:
The patent combines two different depth measurement methods (parallax method and sharpness/blur difference method) into a unified system. The parallax method is used for near-distance objects while the sharpness/blur method is used for far-distance objects, and the system integrates both depth maps to achieve comprehensive depth detection across a wide range of distances, thereby resolving the limitation of restricted depth detection range.
Solution Approach 2:
The patent segments the depth detection task into two distinct methods based on distance ranges. The first image capturing module with parallax is used for near-distance depth detection, while the second image capturing module with multiple apertures is used for far-distance depth detection. This segmentation allows each method to operate within its optimal range, overcoming the baseline and sensor size limitations of the parallax method alone.
2Measurement precision
If multi-aperture method with different optical-wavelength bands is used, then depth information can be obtained, but the depth detection is limited by sharpness/blur difference ratio
Solution Approach 1:
The patent dynamically switches between different depth measurement methods based on the distance to the object. For near-distance objects, the system uses the parallax method from the first image capturing module. For far-distance objects, it transitions to the sharpness/blur difference method using the second image capturing module with multiple apertures. This dynamic adaptation allows the system to maintain accurate depth detection across varying distances despite the limitations of each individual method.
3Device complexity
If single method is used for depth detection, then system complexity is reduced, but depth detection accuracy across wide range is compromised
Solution Approach 1:
The patent creates a universal depth detection system that can handle both near-distance and far-distance objects accurately. The first image capturing module with parallax and the second image capturing module with multiple apertures work together as a unified system. The depth map processing device intelligently selects and integrates depth information from both modules, making the system universally applicable across a wide range of distances while maintaining high accuracy, thereby justifying the increased complexity.
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
The system achieves a greater depth detection range and accuracy, enabling effective focus across various distances, improving the practical application of image capturing systems by combining depth maps from different optical-wavelength images.
Implementation Method 1
the first lens is used for imaging the scene to the first image sensor
Implementation Method 2
the first aperture is used to limit the amount of light with a first optical-wavelength range
Implementation Method 3
the second lens is used for imaging the scene to the second image sensor
Implementation Method 4
the second aperture is used to limit the amount of light with the first optical-wavelength range
Implementation Method 5
the third aperture is used to limit the amount of light with a second optical-wavelength range
Implementation Method 6
generates a second depth map according to the difference/ratio of sharpness/blur between the second image and the third image
Data Source
AI summary
An image capturing system comprises a first image capturing module, a second image capturing module and a depth map processing/generating device. The first image capturing module having a first lens, a first aperture and a first image sensor generates a first image corresponding to a first optical-wavelength range. The second image capturing module having a second lens, a second aperture, a third aperture and a second image sensor generates a second image of the first optical-wavelength range and a third image of a second optical-wavelength range. The depth map processing/generating device generates a first depth map according to the parallax of the scene observed in the first image and the second image, and generates a second depth map according to the difference/ratio of sharpness/blur between the second image and the third image, therefore wider range of resolving the object distance and accuracy of the image could be both considered.