Problem Statement

Marine researchers finds it difficult to detect the marine debris present underwater manually. Detecting such debris using Sonar currently requires extensive inspection of large volumes of noisy acoustic imagery making the process slow and error-prone.

Solution

AI-powered Underwater Marine Debris Detection using Side Scan Sonar(SSS) to detect the debris and identify them.

In Detail

Develop an automated Computer Vision (CV) system that analyses Side-Scan Sonar(SSS) imagery to distinguish man-made marine debris from natural sea floor features, detect and classify potential hazards, assign confidence scores and geotag detected objects using available sonar metadata. The system should provide an intuitive dashboard for visualisation and generate structured reports for use in marine surveys and clean-up operations.