| I&T Solution |
AI-Enabled Automated Vessel Air-Draught Detection to Identify the Highest Point and Waterline of the Vessels based on two cameras using Stereo Vision and YOLOv7 algorithm
(REF: S-1567) |
| Trial Project |
|
| Solution Feature |
- Installation of at least two CCTV cameras with software algorithms to capture vessel images.
- Utilization of the YOLOv7 algorithm to detect vessels quickly and accurately using two cameras.
- Depth Estimation between object and camera to detect the highest point and waterline of the vessels. Calculation of vessel distance using a method based on the position of the vessel in both cameras, geometric derivations, and additional technical data
- Integration of a Graphical User Interface (GUI) to configure the AI-module, accept original images and videos from a network shared drive, and provide output in metric scale (meter) indicating the vessel's height above the waterline.
- Real-time analysis and processing of the images and videos within 2 seconds to deliver accurate height measurements.
|
| Trial Application and Expected Outcome |
- Conduct a trial application of the AI-enabled module with stereo vision depth estimation using captured vessel images and videos.
- Verify the accuracy of the depth estimation by comparing the detected highest point and waterline
- Assess the effectiveness of the stereo vision technique in accurately estimating the distance between the vessel and the camera.
- Validate the usability of the Graphical User Interface (GUI) for configuring the module and obtaining height measurements in metric scale.
- Determine the overall reliability, efficiency, and safety improvements achieved by the AI-enabled automated vessel air-draught detection system.
|
| Additional Solution Information |
Alpha AI_pitch deck_Vessel.pdf
|
| Info on I&T Solution Provider |
|
.png)
.png)
