Multi-Sensor Vessel Detection Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting vessels at sea, such as radar, optical imaging, and AIS, are unreliable in adverse weather conditions and low light, and suffer from poor spatial resolution and vulnerability to spoofing.
Innovation Solution
A method and system that fuse data from multiple detectors, including RF antennas, radar, AIS receivers, and optical sensors, to determine a match in object detections, enhancing detection reliability and confidence by combining spatial resolution and weather-resistant data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If optical imaging is used for vessel detection, then visual evaluation of vessels is improved, but detection reliability deteriorates in poor weather conditions and low light
Solution Approach 1:
The patent combines multiple detection methods (optical imaging, radar, RF detection, AIS) into a unified detection system. Each detector type compensates for the weaknesses of others, particularly in adverse weather conditions where optical imaging alone fails. The fusion of detection results from multiple sources improves overall reliability while maintaining visual evaluation capabilities.
2Measurement precision
If radar detection is used for vessel detection, then spatial resolution is improved, but detection reliability deteriorates in adverse weather conditions
Solution Approach 1:
The system merges radar detection with optical imaging and other detection methods. Radar provides spatial resolution and operates in adverse weather, while optical imaging provides visual evaluation when conditions permit. The combination ensures continuous reliable detection across varying weather conditions while maintaining measurement precision.
3Adaptability or versatility
If AIS detection is used for vessel detection, then detection coverage is improved, but reliability deteriorates due to vulnerability to spoofing and hijacking
Solution Approach 1:
The patent uses multiple independent detection systems as intermediaries to verify AIS data. Rather than relying solely on AIS transmissions which can be spoofed, the system cross-checks with optical imaging, radar, and RF detection. This multi-layered verification approach maintains detection coverage while mitigating the reliability issues of AIS vulnerability.
4Illumination intensity
If Low-Light Image Enhancement algorithms are used to brighten images, then feature visibility is improved, but signal-to-noise ratio deteriorates
Solution Approach 1:
The system performs detection in multiple spectral bands and formats before final image enhancement. By gathering data from infrared sensors, radar, and other detectors first, the system has additional information to work with that is not degraded by low-light conditions. This preliminary multi-source data collection allows for more effective enhancement with better preserved signal-to-noise ratio.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method of detecting vessels is described. A vessel is detected if a match exists between an object detection from each of two or more detectors (100), based on data obtained by the two or more detectors (100), which are selected from: an RF antenna (102); a light sensor (104); a radar detector (106); and an Automatic Identification System receiver (108). A method for detecting the presence of vessels at night is also described. The method comprises detecting a plurality of blobs within image data, applying an image mask to the image data to remove regions of the image data that are associated with having static objects and to retain the remaining portions of the image data, identifying one or more clusters of the detected plurality of blobs within the retained image data, and determining the presence of one or more vessels based on the one or more identified clusters.