Container ID Mark Localization Using Corner Reference Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image recognition software struggles with inefficiency and errors in identifying freight container identification marks due to the abundance of miscellaneous information printed on containers, leading to slow processing and incorrect identification.
Innovation Solution
A system and method that locates the freight container identification mark by using two reference points (top left and top right container corner castings) to define a specified area, allowing efficient image analysis within this area, excluding other information as noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image recognition software analyzes the entire image to identify all texts and numbers, then it can detect all information on the freight container, but the processing becomes too complicated and redundant, reducing identification speed
Solution Approach 1:
The patent divides the freight container surface into a predetermined designated area where the identification mark is located and other areas. The system processes only the designated area containing the identification mark, segmenting the image analysis task to exclude redundant information from other areas, thus improving processing speed while maintaining accuracy.
Solution Approach 2:
The patent extracts only the relevant designated area containing the identification mark from the entire freight container image. By taking out and analyzing only this specific region rather than the complete image, the system reduces processing complexity and increases identification speed without sacrificing accuracy.
2Measurement precision
If conventional image recognition software processes all texts and numbers on the freight container, then it can identify all printed information, but it becomes more prone to interference from miscellaneous information, reducing identification reliability
Solution Approach 1:
The patent segments the image into a designated area where the identification mark is located and other areas containing miscellaneous information. By processing only the designated area, the system eliminates interference from unrelated texts and numbers, improving both accuracy and reliability of identification.
Solution Approach 2:
The patent extracts and analyzes only the designated area containing the identification mark, taking out the relevant information from the context of other miscellaneous printed information. This extraction process prevents interference from unrelated elements, ensuring reliable and accurate identification.
3Quantity of substance
If the system analyzes the entire freight container image, then it can capture all printed information, but the processing time increases, reducing operational efficiency
Solution Approach 1:
The patent segments the image analysis to focus only on the designated area containing the identification mark. This segmentation allows the system to capture complete identification information while excluding time-consuming analysis of other printed information on the container, thus reducing processing time.
Solution Approach 2:
The patent extracts only the designated area with the identification mark from the complete freight container image. This extraction enables the system to obtain all necessary identification information without the time loss of processing unrelated printed information, improving operational efficiency.
Data Source
AI summary
The present invention provides a freight container identification mark locator system and method thereof. The method includes the following steps: receiving an image having a freight container with a freight container identification mark from a camera unit; identifying a first reference point and a second reference point from the freight container in the image; obtaining a first coordinate of the first reference point and a second coordinate of the second reference point according to a coordinate model; calculating and outputting a representative coordinate for a specified area in the image according to the first coordinate and the second coordinate; wherein the specified area includes the freight container identification mark. The present invention allows subsequently operations to only identify the freight container identification mark in the specified area according to the representative coordinate, and thus subsequently allowing the freight container identification mark in the image to be identified efficiently.