USV Shoreline Segmentation Using Visible, Thermal, and Radar Fusion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing shoreline recognition and segmentation systems for unmanned surface vessels (USVs) are inadequate for dynamic and complex aquatic environments, as they rely on single sensors, require high lighting conditions, and lack self-control and depth feature extraction capabilities.

Innovation Solution

A method and device for shoreline segmentation in complex environments using an unmanned surface vessel, which involves obtaining visible light, thermal infrared, and radar echo images, fusing them to create all-weather two-dimensional image information, and extracting multi-feature point cloud datasets for shoreline segmentation, while employing a deep reinforcement learning-based path planning model for effective navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple sensors (visible light, thermal infrared, radar) are used for shoreline segmentation, then measurement precision and reliability are improved, but device complexity increases

Engineering Contradiction:

Improveshoreline segmentation precisionVSAvoidmulti-sensor system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system segments the shoreline recognition task into three parallel sensor streams: visible light imaging for optical features, thermal infrared imaging for temperature contrast features, and radar imaging for surface texture features. Each sensor processes independent feature extraction, and the results are fused to achieve comprehensive shoreline segmentation with improved precision while managing system complexity through modular architecture.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent merges data from multiple heterogeneous sensors (visible light camera, thermal infrared camera, radar) into a unified shoreline recognition system. The fusion of multi-source imaging data compensates for individual sensor limitations and achieves all-weather, all-condition shoreline segmentation capability, resolving the contradiction between improved measurement precision and increased device complexity.

Inventive Principle:

Principle #5Merging (Combining)

2Ease of operation

If visible light images are used for shoreline recognition, then ease of operation is improved, but reliability deteriorates due to lighting condition dependencies

Engineering Contradiction:

Improveimage acquisition simplicityVSAvoidshoreline recognition reliability

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The system employs a composite sensing approach by combining visible light imaging with thermal infrared imaging and radar imaging. This multi-modal composite system ensures reliable shoreline recognition across all lighting conditions: visible light provides ease of operation during daytime, while thermal infrared and radar ensure reliability during nighttime or poor lighting conditions, eliminating the reliability deterioration caused by lighting dependencies.

Inventive Principle:

Principle #40Composite materials

3Measurement precision

If high-performance models are applied for shoreline segmentation, then measurement precision is improved, but productivity deteriorates due to computational power limitations

Engineering Contradiction:

Improveshoreline segmentation accuracyVSAvoidreal-time processing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system segments the computational workload by processing different sensor data streams independently and in parallel. Visible light, thermal infrared, and radar data are processed separately through dedicated feature extraction pipelines, then fused to produce the final shoreline segmentation. This segmented processing approach maintains high measurement precision while improving productivity by distributing computational tasks and enabling real-time operation on USV platforms with limited computational power.

Inventive Principle:

Principle #1Segmentation

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 solution enables robust, real-time shoreline segmentation and navigation in complex environments, overcoming limitations of single-sensor reliance and high lighting requirements, and ensuring effective self-control and depth feature extraction for USVs.

Implementation Method 1

obtaining a visible light image, a thermal infrared image and a raw radar echo image of a shoreline

Methodology Applied

Scientific EffectThermal radiation: Thermal Radiation

Implementation Method 2

obtaining an echo image of the shoreline based on the raw radar echo image

Methodology Applied

Scientific EffectRadar echo: Radar

Data Source

PatentUS20250139930A1Method and device for shoreline segmentation in complex environments based on the perspective of an unmanned surface vessel (USV)

Publication Date: 2025.05.01 HUNAN UNIV

AI summary

A method for shoreline segmentation in complex environments based on the perspective of an unmanned surface vessel is provided. A visible light image, a thermal infrared image and a raw radar echo image of a shoreline are obtained. The visible light image and the thermal infrared image are subjected to fusion and feasible region segmentation to obtain an all-weather two-dimensional image information of the shoreline, and an echo image including tiny features is obtained based on the raw radar echo image. An extraction region is constrained and shoreline features are enhanced based on the all-weather two-dimensional image information and the echo image to obtain a multi-feature point cloud dataset for shoreline segmentation.