Sea-Surface Oil Detection Using Learned Behavioral Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video surveillance systems are unable to reliably identify sea-surface oil, which can result from offshore oil platform operations or spills, due to constant variations in the maritime environment, leading to false-positive identifications.
Innovation Solution
A method and system combining a camera system and computer vision engine with machine learning capabilities to analyze video frames from LWIR cameras, identifying foreground blobs indicative of sea-surface oil and distinguishing them from noise and false positives by learning expected patterns over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional video surveillance systems use static predefined patterns to identify sea-surface oil, then the system structure remains simple, but the reliability of identification deteriorates due to false positives from environmental variations
Solution Approach 1:
The system transitions from static predefined patterns to dynamic learned patterns that adapt to changing maritime environments. The machine learning engine continuously learns from video data to update patterns of normal oil presence, allowing the system to adapt to environmental variations while maintaining reliability.
Solution Approach 2:
The system performs self-learning and self-updating through the machine learning engine, which automatically analyzes video data and refines patterns without external intervention. This enables the system to improve its own identification reliability while managing the complexity internally.
2Reliability
If the system uses machine learning to distinguish normal from anomalous oil patterns, then false alarms are reduced, but the processing time and computational complexity increase
Solution Approach 1:
The machine learning engine performs preliminary learning during periods when no anomalies are present, building a baseline of normal oil patterns. This preliminary action prepares the system to quickly identify anomalies when they occur, reducing processing time during critical detection phases.
Solution Approach 2:
The system continuously processes video data to update patterns and detect anomalies without interruption. This continuous operation allows the system to maintain up-to-date knowledge of normal conditions while promptly detecting changes, optimizing both reliability and response time.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively generates alerts for anomalous sea-surface oil patches, differentiating between normal and abnormal observations, thereby facilitating investigation and reducing false alarms.
Implementation Method 1
The input stream of video frames may be generated by one or more long wavelength infrared (LWIR) cameras
Data Source
AI summary
A behavioral recognition system may include both a computer vision engine and a machine learning engine configured to observe and learn patterns of behavior in video data. Certain embodiments may be configured to detect and evaluate the presence of sea-surface oil on the water surrounding an offshore oil platform. The computer vision engine may be configured to segment image data into detected patches or blobs of surface oil (foreground) present in the field of view of an infrared camera (or cameras). A machine learning engine may evaluate the detected patches of surface oil to learn to distinguish between sea-surface oil incident to the operation of an offshore platform and the appearance of surface oil that should be investigated by platform personnel.