Structured Light Depth Imaging via Virtual Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional depth image acquisition technologies face challenges in stability due to dependence on image feature recognition, are prone to errors in varying brightness and color conditions, and suffer from insufficient resolution and limited response rates, particularly in applications like 3D gaming and medical imaging.
Innovation Solution
A depth image acquiring device comprising a projecting device that projects a random pattern of scattered light spots, and at least two image sensing devices that capture these patterns, with an image processing unit to match and verify pixel correspondence across images, generating a disparity image for calculating depth values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If passive image comparison method is used to calculate depth image, then depth information can be obtained, but the method is very liable to influence of changes in color and brightness, causing poor comparison result and greatly compromised stability
Solution Approach 1:
The patent applies active light source projection that emits structured light patterns (such as infrared or visible light) to illuminate the target object. This active illumination creates consistent lighting conditions and projected patterns that are independent of ambient light conditions, thereby eliminating the sensitivity to color and brightness changes that plagues passive image comparison methods. The projected patterns serve as stable reference features for depth calculation regardless of environmental lighting variations.
2Measurement precision
If TOF technology is used for depth sensing, then depth information can be acquired, but the resolution is insufficient (160×120-320×240) and response rate is limited
Solution Approach 1:
The patent projects structured light patterns consisting of multiple distinct spots or lines across the target scene. Each projected spot corresponds to a specific spatial location, and by tracking the displacement of these segmented light spots in captured images, the system achieves high-resolution depth measurement. This segmentation of the projection pattern into discrete measurable elements enables resolution far exceeding conventional TOF sensors.
3Productivity
If active light source is projected and image features are created by projecting color codes, then matching operation can be accelerated by pyramid algorithm, but device complexity increases
Solution Approach 1:
The patent pre-projects structured light patterns with known spatial configurations onto the target scene before performing depth calculation. These pre-projected patterns create predetermined feature points that can be rapidly matched using optimized algorithms. The pyramid algorithm mentioned in the background is applied to this pre-structured data, achieving fast matching by coarse-to-fine processing of the known pattern displacements, thereby accelerating productivity while managing complexity through the use of predictable, pre-planned projection patterns.
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 enhances accuracy and stability by reducing the influence of ambient light and occlusion issues, improving depth image resolution and response rates, and providing more reliable depth information in diverse environmental conditions.
Implementation Method 1
The projecting device projects a projection pattern to an object, in which the projecting device serves as a virtual image sensing device, and an inherent image of the projection pattern serves as a virtual image
Implementation Method 2
The first and the second image sensing device sense the projection pattern projected to the object respectively, to generate a first real image and a second real image
Data Source
AI summary
A depth image acquiring device is provided, which includes at least one projecting device and at least one image sensing device. The projecting device projects a projection pattern to an object. The image sensing device senses a real image. In addition, the projecting device also serves as a virtual image sensing device. The depth image acquiring device generates a disparity image by matching three sets of dual-images formed by two real images and one virtual image, and generates a depth image according to the disparity image. In addition, the depth image acquiring device also generates a depth image by matching two real images, or a virtual image and a real image without verification.