PTTEP has initiated an aquatic animal monitoring system as part of its Ocean for Life strategy, focusing on conservation surveys and tracking of rare aquatic animals.
An underwater camera was installed on a jacket leg of a PTTEP platform to record underwater life. The footage then was analyzed by artificial intelligence (AI) software using an object-detection method to categorize the animals. A machine learning algorithm provided more accuracy. The AI software also can shorten the videos by cutting out any periods not showing aquatic life.
For the detection algorithm, a large amount of data in the form of images with labels was used to train software to recognize target objects. The AI software was able to detect and identify nine species of aquatic animals—fish, turtles, whales, dolphins, sharks, seals, sea lions, stingrays, and seahorses. Using the AI software, the raw video file can be shortened by up to 85% by removing periods in which no aquatic life is shown in the original video and tracking only animal life in the video frame.