3D Camera Coded Information Reading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for reading coded information from objects face challenges with perspective distortion and the need for multiple cameras to accurately identify codes on 3D objects, leading to increased complexity and decreased reading performance, especially when objects are close together.
Innovation Solution
The use of three-dimensional cameras to capture images and process them into distortion-free two-dimensional images, allowing for accurate coded information recognition without the need for multiple cameras, as the 3D images inherently provide volume information and enable correct object association.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple 2D cameras are used to capture codes on all faces of 3D objects, then code detection coverage is improved, but system complexity and equipment requirements increase
Solution Approach 1:
The patent transitions from using multiple 2D cameras to a single 3D camera, changing the dimensional capability of the imaging system. The 3D camera captures depth information and spatial relationships, enabling it to detect codes on multiple faces of objects simultaneously without requiring multiple cameras positioned at different angles.
Solution Approach 2:
The 3D camera serves multiple functions that would otherwise require separate devices: it can detect codes on top faces, side faces, and bottom faces of objects within its field of view, and it provides both spatial positioning and code detection capabilities in a single system.
2Speed
If 2D cameras are used to read codes on moving objects, then reading speed is maintained, but perspective distortion reduces measurement precision
Solution Approach 1:
By introducing the third dimension (depth) through 3D imaging, the system can compensate for perspective distortion. The depth information allows the system to understand the spatial relationship between the camera and the code, enabling accurate code reading even when the code is at an angle or varying distance from the camera.
Solution Approach 2:
The system uses the Z-coordinate (depth) information from the 3D camera to dynamically adjust reading parameters. When a code is detected at a certain depth and angle, the system can modify the reading strategy accordingly, maintaining accuracy despite changes in object position, speed, or orientation.
3Measurement precision
If objects are spaced far apart on the conveyor belt, then code reading accuracy is improved, but productivity decreases
Solution Approach 1:
The 3D camera's depth perception capability allows it to distinguish between objects at different Z-positions, enabling accurate code reading even when objects are closely spaced on the conveyor belt. The system can identify which object is in focus and read its code without interference from adjacent objects.
Solution Approach 2:
The Z-coordinate information acts as an intermediary that mediates between closely spaced objects. By using depth information to separate and identify individual objects in the scene, the system can accurately read codes on each object without requiring physical spacing between them.
4Measurement precision
If the speed of movement is reduced, then code reading accuracy is improved, but productivity decreases
Solution Approach 1:
The 3D camera enables dynamic parameter adjustment based on object position and speed. The system can adapt reading parameters such as exposure time, gain, and processing algorithms in real-time based on the depth and velocity information, maintaining accuracy even at higher conveyor speeds.
Data Source
AI summary
The invention relates to a method and a system for reading coded information from an object. The system comprises one or more three-dimensional cameras configured such as to capture three-dimensional images of the object and a processor configured such as to process the captured three-dimensional images. The processor is designed to: identify planes upon which faces of the object lie; extract two-dimensional images that lie on the identified planes; and apply coded information recognition algorithms to at least part of the extracted two-dimensional images.