Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Ship Identification System
Transcript of Ship Identification System
Goal of the project
There are geostationary cameras located in ports doing only recording, also their videos will be deleted after few months back of storing. The goal of this project is to create a system that can extract useful information from those video streams
Q & A
Objectives of the system
Identify vessel types using live or recorded video stream.
Location of ships from the FOV (Field Of View) coordinates.
Plot and visualize approximate place of the ship on the geographical map.
Generate reports which are described ship information.
Domain of the project.
System will use live video stream from geostationary camera.
Use only for Harbors.
Geographical Map Will represent the area that captured by the geostationary camera
Feasibility of the Project
OpenCV - real-time computer vision library.
Lots of preferences in the computer vision.
User Friendly IDE with support.
Lots of Classifier Training preferences
R. K.J. P. Rathnayaka
H. M. A. U. Bandara
G. S. V. M. Ishan
G. K. L. Abayawardana
H. M. T. B. Wijesuriya
B. M. A. P. Wimalarathna
A. S. F. Rumana
Methodology of Development
Data Gathering for analysis
Discuss with seniors who did related projects
Read Research papers such as “Coding Robin”
Abhishek Kumar Annamraju (2014), TRAINCASCADE AND CAR DETECTION USING OPENCV[Online]
Andrew s sobral (2015), Vehicle Detection, tracking and Counting [online]
Computer Vision Workshop - Project Report [online]
Summarized Data Gathering
Got the video stream of cargo ship from Colombo Port
Downloaded 2000 negative images
Downloaded 2000 positive images
Staff of Colombo Port authority
Identifies the vessel types.
After a ship is detected, a message will be popped-up with an alarm
System will be able to track the path of the ships and update database continuously.
Availability of the system
Availability of the web service
Using AIS System
Operator able to add additional meta-data to generated reports.
System provides the facility of the counting, number of the ships.
Can search and visualize, how ships are located in a given time range.
Automatically store the generated data record.
Domain of the project
Standalone system will run only in Linux distributions
Web site only provide facility to access data for authorized users
Use case Diagram
● Implementation cost
● Hardware and software technology cost
● Daily administrative cost
Builder Creational Pattern
Proposed Design pattern
Proposed Data Design
Training the classifier
Creating connection with database
Connect between OpenCV and Qt IDE
Train a classifier to accurately identify vessels.
Create a user friendly reliable system to analyze data