3D Image Depth Resolution via Motion-Gated Multi-Image Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques for extracting 3D information from devices result in low-resolution depth information, degrading the quality of 3D images compared to color images.
Innovation Solution
A device and method that utilize a projector to project light patterns onto an object, with multiple cameras and a processor to obtain and merge depth images by adjusting light parameters and motion sensing, allowing for the creation of high-resolution 3D images by merging depth information from multiple images based on motion thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If depth information is extracted using a single depth image, then the extraction process is simple and fast, but the resolution of depth information is low, degrading the quality of 3D images
Solution Approach 1:
The patent merges multiple depth images captured from different positions to create a composite depth image with higher resolution. The processor combines depth information from multiple sources, integrating them to produce enhanced depth data that overcomes the resolution limitations of individual depth images.
Solution Approach 2:
The patent introduces the dimension of spatial movement by capturing depth images at multiple positions. By moving the device to different locations and capturing depth images from various angles, the system adds spatial diversity to the depth data collection process, enabling higher resolution reconstruction.
2Measurement precision
If multiple depth images are captured and merged, then the resolution of depth information increases, but the time required for image acquisition and processing increases
Solution Approach 1:
The patent employs periodic capture of depth images at multiple positions rather than continuous capture. The device moves to predetermined positions and captures depth images at each position, creating a periodic sampling pattern that reduces total acquisition time while maintaining sufficient resolution.
Solution Approach 2:
The patent pre-determines the positions at which depth images will be captured before actual image acquisition begins. By planning the capture positions in advance, the system optimizes the acquisition process, avoiding unnecessary movements and captures, thereby reducing overall time consumption.
3Measurement precision
If depth images are merged without considering device motion, then processing is faster, but merging accuracy decreases due to motion-induced misalignment
Solution Approach 1:
The patent uses motion sensor data as feedback to adjust the depth image merging process. The motion information provides real-time data about device movement, which is fed back into the merging algorithm to compensate for motion-induced misalignments, ensuring accurate integration of depth images even during device movement.
Solution Approach 2:
The patent dynamically adjusts merging parameters based on motion conditions. When motion is detected, the system changes the merging parameters to account for the movement, such as adjusting alignment transformations or selecting appropriate depth values, thereby maintaining accuracy under varying motion conditions.
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 method enhances the resolution of depth information, improving the quality of 3D images by selectively merging depth data from multiple images based on device motion, resulting in more accurate and detailed 3D representations.
Implementation Method 1
a projector configured to project light having a preset pattern to an object; a first camera configured to sense light reflected and fed back by the object
Implementation Method 2
a first camera configured to sense light reflected and fed back by the object to obtain a depth image
Data Source
AI summary
A device for obtaining an image includes a projector configured to project light having a preset pattern to an object, a first camera configured to sense light reflected and fed back by the object to obtain a depth image, a second camera configured to obtain a color image of the object, a motion sensor configured to sense a motion of the device. The device also includes and a processor configured to obtain a color image and a plurality of depth images of the object, by changing a parameter about the projected light and the light sensed by the first camera. When the motion of the device is less than a threshold, the processor is configured to merge pieces of depth information respectively obtained from the plurality of depth images with each other, and obtain a 3D image of the object, based on the merged depth information and the color image.