Visual Positioning via Iterative Marker Structure Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional visual positioning technologies face inaccuracies due to vague images of visual code markers, leading to incorrect detection of corner positions and subsequent positioning errors in electronic devices.
Innovation Solution
A visual positioning apparatus and method that uses an image sensor and processor to identify marker images, select feature points, update marker structures, and calculate positions in space by iteratively refining feature points to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional visual positioning technology uses an electronic product with camera function to sense a visual code marker, then the positioning function is provided, but the image of the visual code marker may be vague resulting in inaccurate marker image identification and corner position detection
Solution Approach 1:
The patent applies preliminary action by pre-establishing a marker structure model with predetermined vertices and feature points before actual positioning. The processor reconstructs the marker structure by matching detected corner positions with the pre-defined marker structure, allowing for accurate feature point identification even when the captured image is vague. This pre-prepared structural framework enables reliable positioning without requiring perfectly clear images.
Solution Approach 2:
The patent implements feedback through an iterative optimization process where the processor repeatedly adjusts feature point positions based on the marker structure model and detected corner positions. The system continuously refines the positioning results by comparing detected features with the expected marker structure, using this feedback to improve accuracy progressively until convergence is achieved.
2Adaptability or versatility
If the image of the visual code marker is vague, then the positioning function can still be provided, but the marker image cannot be identified accurately and the four corner positions cannot be detected accurately
Solution Approach 1:
The patent introduces the marker structure model as an intermediary between the vague captured image and the final positioning result. Instead of directly extracting corner positions from the potentially vague image, the system uses the pre-defined marker structure with known geometric relationships as an intermediary framework. This marker structure acts as a reference that guides the accurate identification of corner positions and feature points even when the original image quality is poor.
Solution Approach 2:
The patent applies parameter changes by transforming the positioning problem from direct image analysis to a multi-parameter optimization problem. The system adjusts multiple parameters including feature point positions, marker structure vertices, and correspondence relationships between detected corners and model features. By changing from a single-step detection approach to a multi-parameter iterative optimization approach, the system achieves accurate positioning despite vague images.
Data Source
AI summary
A visual positioning apparatus, method, and non-transitory computer readable storage medium thereof are provided. The visual positioning apparatus derives an image by sensing a visual code marker in a space and performs the following operations: (a) identifying an identified marker image included in the image, (b) searching out the corner positions of the identified marker image, (c) deciding a marker structure of the identified marker image according to the corner positions, wherein the marker structure includes vertices, (d) selecting a portion of the vertices as first feature points, (e) searching out a second feature point for each first feature point, (f) updating the vertices of the marker structure according to the second feature points, (g) selecting a portion of the updated vertices as the third feature points, and (h) calculating the position of the visual positioning apparatus according to the third feature points.