Ship Section Man-Overboard Detection Using 3D Kinematic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current man-overboard monitoring systems on ships face challenges in accurately distinguishing between a person falling overboard and false alarms, such as spray or birds, due to limitations in video-based detection methods, leading to high false alarm rates and inadequate reliability.
Innovation Solution
A monitoring device equipped with cameras and an evaluation device that uses three-dimensional kinematic analysis and artificial intelligence to differentiate between a person and other objects by determining the starting position and kinematic variables of moving objects, reducing false alarms through precise calibration and image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video-based detection methods are used to monitor man-overboard events, then detection capability is provided, but false alarm rate increases due to inability to distinguish between persons and other objects like spray or birds
Solution Approach 1:
The patent transitions from two-dimensional image analysis to three-dimensional spatial analysis by determining the starting position of moving objects in 3D space. This dimensional enhancement allows the system to distinguish between objects at different depths and locations, enabling reliable differentiation between persons falling overboard and other moving objects like spray or birds without increasing false alarm rates
2Area of stationary object
If conventional video surveillance is used, then monitoring coverage is achieved, but false detections increase due to spray, birds, or falling objects being mistaken for man-overboard events
Solution Approach 1:
By extending analysis from 2D image planes to 3D spatial coordinates, the system can identify the starting position of detected objects in three-dimensional space. This allows the monitoring system to maintain broad coverage while filtering out false alarms by verifying whether objects originate from the ship deck area versus other locations in the monitoring volume
Solution Approach 2:
The patent applies different evaluation criteria to different spatial zones within the monitoring area. By focusing kinematic analysis specifically on objects with starting positions in the ship section or railing area, the system maintains high reliability for critical zones while tolerating detections in peripheral areas, effectively reducing overall false alarm rates
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention relates to a monitoring device 1 for monitoring a man-overboard situation in a ship section 4, wherein the ship section 4 is monitored by video technology using at least one camera 5a,5b and the camera 5a,5b is designed to provide surveillance in the form of video data. The monitoring device comprises an evaluation device 7, said evaluation device 7 having an interface for receiving the video data, wherein the evaluation device 7 is designed to detect a moving object in the ship section 4 on the basis of the video data and to determine a kinematic variable of the moving object. The evaluation device 7 is designed to determine a starting point in three dimensions on the basis of the video data and the kinematic variable of the moving object and to evaluate the moving object as a man-overboard event on the basis of the starting point.