Pattern-Based Decoding Using Object Distance Ratios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current coding and decoding technologies for 1-D, 2-D bar codes, and micro-codes are limited by the focal length and resolution of sensors, leading to decreased decoding accuracy, especially when decoding at short distances and with sensor tilting.
Innovation Solution
A coding and decoding method using patterns where distances or ratios between objects are set to form codes, allowing sensors to obtain distances or ratios between object-images for accurate decoding, even at short distances and with varying sensor positions or tilts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional bar code decoding methods are used, then decoding can be performed, but decoding accuracy decreases when decoding at short distances due to focal length limitations
Solution Approach 1:
The patent changes the fundamental parameters of code representation from traditional bar code patterns to geometric patterns with specific distance relationships between objects. This allows the code to be decoded based on relative distances rather than absolute pixel measurements, enabling accurate decoding at various distances including short distances without being limited by sensor focal length.
2Measurement precision
If sensor resolution is increased to improve decoding accuracy, then decoding precision improves, but device complexity and cost increase
Solution Approach 1:
The patent uses the concept of copying by creating a coded pattern that encodes information through the spatial relationships between multiple objects. The decoder measures distances between object images and uses these relative measurements to reconstruct the original code, eliminating the need for high-resolution sensors to directly resolve fine bar code details.
Solution Approach 2:
The patent transitions from traditional 1-D or 2-D bar code representations to a multi-object geometric pattern system where information is encoded in the dimensional relationships (distances) between objects. This allows decoding to be performed by measuring distances between object images rather than requiring high-resolution imaging of fine patterns.
3Measurement precision
If sensor position is fixed to maintain decoding accuracy, then measurement precision is maintained, but adaptability to different positions and tilts decreases
Solution Approach 1:
The patent incorporates feedback mechanisms where the system measures actual distances between object images and compares them against expected distance relationships encoded in the pattern. When sensor tilt or position variation causes measurement deviations, the system can detect these discrepancies and adjust the decoding process accordingly, maintaining accuracy across different sensor positions.
Solution Approach 2:
The patent creates a universal coding system where the same pattern-based approach works across multiple sensor positions and orientations. The distance-based encoding allows the code to be decoded accurately regardless of whether the sensor is tilted or positioned at different heights, making the system universally applicable to various capture scenarios without requiring precise sensor positioning.
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 reduces limitations from sensor focal lengths and resolutions, enabling accurate decoding within short distances and compensating for sensor tilt errors, enhancing decoding accuracy and flexibility.
Implementation Method 1
sensing a pattern comprising a plurality of objects and obtaining a plurality of object-images corresponding to the objects
Data Source
AI summary
Disclosed are coding and decoding methods using patterns and a system thereof. The object-images are obtained via detecting a pattern comprising objects. The distances between the object-images are obtained and a code is found based on the obtained distances between the object images. The distances between the object-images are defined as distances between centers of the object-images. As there are at least three objects in a pattern, a code is obtained based on a ratio of distances between object-images, which equals to a ratio of distances between objects. As there are at least four objects in a pattern, it is first determined whether a first ratio of object-images equals to a predetermined ratio of object-images. If no, a second ratio of object-images is obtained via a correcting factor and the first ratio of object-image, and finally a code is obtained based on the second ratio of object-images.