AR Measurement Using Depth Images for Vertical Plane Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AR measurement technologies using monocular cameras suffer from slow initialization speeds and large measurement errors, particularly when measuring vertical planes with limited texture or colors similar to adjacent horizontal planes, leading to incorrect plane recognition.
Innovation Solution
Establish a three-dimensional coordinate system based on a real environment using a depth image from a camera, obtain pose data of the terminal, and determine measurement points in this system to accurately measure geometric parameters without relying on plane recognition, enabling fast initialization and improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If plane recognition is used for AR measurement, then measurement functionality is provided, but measurement precision deteriorates due to incorrect plane recognition on vertical planes with limited texture or similar colors
Solution Approach 1:
The patent introduces a depth image as an intermediary data source to bridge the gap between camera images and 3D spatial information. Instead of directly recognizing planes from 2D images (which causes accuracy issues), the depth image serves as a mediator that provides accurate depth data for vertical planes, enabling precise measurement without relying on problematic 2D plane recognition algorithms.
Solution Approach 2:
The patent replaces the traditional 2D image-based plane recognition mechanism with a depth image-based measurement system. This substitution eliminates the fundamental limitation of 2D recognition on vertical planes by using depth information that directly captures the 3D spatial structure, thereby improving measurement precision without sacrificing ease of operation.
2Ease of operation
If monocular camera with plane recognition is used, then AR measurement is enabled, but initialization speed becomes slow
Solution Approach 1:
The patent performs preliminary action by capturing a depth image before the actual measurement process. This pre-captured depth image serves as ready-to-use 3D spatial reference data, eliminating the need for time-consuming real-time plane recognition during measurement. The depth image is obtained in advance and stored, allowing instant use during measurement operations.
Solution Approach 2:
The patent creates a 3D spatial model copy from the depth image that represents the real-world environment. This digital copy serves as a pre-computed reference framework that can be immediately applied for measurements without requiring real-time processing. The depth image acts as a template or blueprint that accelerates the measurement initialization process.
3Ease of operation
If 2D image plane recognition is used, then measurement function is provided, but measurement precision deteriorates on vertical planes with limited texture or colors similar to adjacent horizontal planes
Solution Approach 1:
The patent transitions from 2D image processing to 3D depth-based processing by introducing the depth dimension. The depth image provides Z-axis information that is completely absent from 2D images, enabling the system to distinguish vertical planes from horizontal planes based on depth gradients rather than relying on texture or color patterns that cause recognition failures in 2D space.
Solution Approach 2:
The patent changes the fundamental measurement parameter from 2D image characteristics (color, texture) to 3D depth characteristics. By using depth images, the system measures geometric parameters based on depth variations and spatial relationships rather than relying on surface appearance, fundamentally changing how measurement accuracy is achieved and eliminating the limitations of 2D-based approaches.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Provided is a method for measuring a geometric parameter of an object. The method includes: (101) obtaining a first depth image of a real environment photographed by a camera component, and establishing, in accordance with the first depth image, a three-dimensional coordinate system based on the real environment; (102) obtaining pose data of a terminal, obtaining a second depth image and a two-dimensional image of an object to be measured, and displaying the two-dimensional image on a display interface of the terminal; (103) receiving a measurement point selection instruction, determining a coordinate in the three-dimensional coordinate system of a measurement point selected in accordance with the measurement point selection instruction based on the pose data, the second depth image and the two-dimensional image, and determining a geometric parameter of the object to be measured based on the coordinate; and (104) displaying, in the two-dimensional image displayed on the display interface, a measurement mark of the measurement point and the geometric parameter of the object to be measured. The measurement accuracy of Augmented Reality (AR) measurement technology is improved. Further provided are an apparatus for measuring a geometric parameter of an object and a terminal.