AI Draft Measurement From Video Under Angle and Hull Fouling Errors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing draft measurement methods for ships are prone to errors due to variations in shooting angle, picture quality, and hull surface contamination, and pose safety risks such as musculoskeletal and crush injuries.
Innovation Solution
A draft measurement system using artificial intelligence technology, including a drone or video acquisition device, applies deep learning models for segmentation and edge detection, distortion correction algorithms, and real-time video processing to accurately determine draft values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a trained person directly measures the draft using a draft gauge after approaching draft marks painted on each side of the ship, then the measurement can be performed with simple equipment and procedures, but the measurement accuracy deteriorates due to environmental conditions such as wind, waves, and swells, and the skill level of the measurer
Solution Approach 1:
The patent replaces the mechanical manual measurement system with an automated image processing system. A camera captures images of the ship's draft marks, and computer vision algorithms automatically measure the draft, eliminating the need for manual measurement with draft gauges. This substitution resolves the contradiction by maintaining procedural simplicity while dramatically improving measurement accuracy through automated image analysis that is not affected by environmental conditions or human skill variations.
Solution Approach 2:
The patent creates a digital copy of the physical draft measurement process. Instead of directly measuring with physical tools, the system captures optical images of the draft marks and processes them digitally to obtain measurement results. This copying approach allows the measurement to be performed remotely and automatically, improving accuracy while keeping the overall process simple.
2Reliability
If a trained person boards a workboat to approach draft marks for measurement, then the measurement can be performed directly at the ship, but safety risks increase due to crush injuries, drowning, and musculoskeletal injuries from postural instability
Solution Approach 1:
The patent introduces an intermediary device (camera or drone) to perform the measurement task remotely. Instead of having a person board the workboat and approach the ship, the camera captures images from a safe distance, and the image processing system performs the measurement. This intermediary approach eliminates safety risks while maintaining measurement reliability through automated analysis.
Solution Approach 2:
The patent replaces the mechanical approach of manual measurement from a workboat with an automated remote measurement system. The camera-based system eliminates the need for personnel to board vessels in hazardous conditions, substituting physical presence with optical capture and digital processing, thereby removing safety risks entirely while maintaining measurement quality.
3Productivity
If image processing is performed without artificial intelligence technology, then the processing speed and simplicity are maintained, but measurement accuracy deteriorates due to errors from shooting angle, picture quality, and hull surface contamination
Solution Approach 1:
The patent applies artificial intelligence technology that can adapt to varying parameters such as shooting angles, picture qualities, and surface conditions. The AI model is trained to recognize draft marks under diverse conditions and can adjust its analysis accordingly, maintaining both high processing efficiency and improved measurement accuracy by automatically compensating for environmental variations.
Solution Approach 2:
The AI-based image processing system incorporates feedback mechanisms that allow it to learn from and adapt to different measurement conditions. The system can identify and correct errors caused by varying shooting angles, picture quality, and surface contamination through iterative optimization and validation, maintaining both speed and accuracy.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A draft measurement system and method based on image processing using artificial intelligence technology. The draft measurement system and method can improve the accuracy of draft measurement by preventing errors in image processing that can occur depending on the shooting angle, the picture quality, the degree of surface contamination of a hull, and the like, through application of artificial intelligence technology to images extracted from video data, and can determine a draft value based on video data acquired in real time using technology such as a deep learning model for segmentation for draft mark edge detection and waterplane recognition and an edge detection algorithm and distortion correction algorithm for waterplane detection.