Forward-Looking Sonar Image Stitching With Variance-Based Blending

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for stitching forward-looking sonar images fail to consider intra-frame and inter-frame artifacts caused by non-ideal sonar imaging configurations, resulting in overly blurred and information-deficient stitched images, particularly in underwater detection tasks.

Innovation Solution

A method for stitching forward-looking sonar images that includes image registration using phase correlation, information extraction through local and global variance statistics, and image blending with a global variance map to adapt to non-ideal configurations and retain relevant information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If existing stitching methods are applied to forward-looking sonar images, then image stitching can be achieved, but intra-frame and inter-frame artifacts cause excessive blurring and information loss

Engineering Contradiction:

Improveimage informationVSAvoidstitching precision

Core Design Contradiction:

Loss of informationVSManufacturing precision

Solution Approach 1:

The patent applies preliminary action by performing image registration and estimating relative rotational displacements using phase correlation method before the actual stitching process. This preliminary alignment compensates for artifacts in advance, preventing information loss during stitching.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes parameters by introducing a global variance map that dynamically adjusts the contribution of different images based on their local variance statistics. This parameter adjustment optimizes the stitching process to retain important image information while reducing artifacts.

Inventive Principle:

Principle #35Parameter changes

2Area of stationary object

If all images in the sequence are averaged during stitching, then complete coverage is achieved, but data volume and processing complexity increase

Engineering Contradiction:

Improveimage coverageVSAvoiddata volume

Core Design Contradiction:

Area of stationary objectVSQuantity of substance

Solution Approach 1:

The patent applies local quality by calculating local variance statistics for different regions of the image sequence. Images are selectively weighted and combined based on their local information content, ensuring that regions with important features are preserved while reducing redundancy in uniform areas.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent uses partial action by selectively averaging only those images that contribute meaningful information to each region, rather than blindly averaging all images. The global variance map controls which images are partially or fully included in the final stitched result, optimizing both coverage and data efficiency.

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If phase correlation method is used for image registration, then rotational displacement estimation is improved, but processing time increases

Engineering Contradiction:

Improverotational displacement estimationVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by using phase correlation method to estimate relative rotational displacements between adjacent images before cumulative alignment. This preliminary estimation provides accurate initial alignment, reducing the need for iterative refinement and overall processing time.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent transforms the 2D image correlation problem into a 1D phase difference calculation in the frequency domain using Fourier transform. This dimensional transformation simplifies the computational complexity while maintaining high precision in rotational displacement estimation.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

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 method effectively addresses intra-frame and inter-frame artifacts, retains important image information, and reduces data volume by selecting images with high local variance values for weighted averaging, enhancing underwater detection efficiency.

Implementation Method 1

estimating relative rotational displacements of the forward-looking sonar image sequence using a phase correlation method

Methodology Applied

Scientific EffectPhase correlation:

Data Source

PatentUS20260030767A1Method for stitching forward-looking sonar images while retaining information

Publication Date: 2026.01.29 ZHEJIANG UNIV

AI summary

A method for stitching forward-looking sonar images while retaining information is provided. In this application, forward-looking sonar is used as underwater detection equipment and acquired forward-looking sonar images are stitched together. A phase correlation method is employed for estimating displacements between images to determine the position of each single forward-looking sonar image within the stitched image. A method based on local statistics is used for image blending to obtain a stitched forward-looking sonar image that retains information. The method for stitching forward-looking sonar images proposed in this application adapts to intra-frame and inter-frame artifacts caused by non-ideal sonar imaging configurations, overcoming the drawbacks of image quality degradation due to the intra-frame and inter-frame artifacts. The method enhances the amount of information contained in the stitched image, which can assist observers in conducting rapid underwater exploration.