Image Watermark Extraction Using Positioning Block Alignment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods face difficulties in extracting watermarks from images due to deformation caused by acquisition angles, leading to reduced success rates.

Innovation Solution

A method involving a first identification model to identify positioning blocks, followed by affine transformation to align the image, and a second identification model to extract information blocks, utilizing encoder-decoder structures with self-attention models to enhance watermark extraction from deformed images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a trained machine learning model is used to directly extract watermarks from acquired images, then the extraction process is simple, but the extraction success rate is reduced due to image deformation from acquisition angles

Engineering Contradiction:

Improveextraction process simplicityVSAvoidextraction success rate

Core Design Contradiction:

Ease of operationVSReliability

Solution Approach 1:

The patent applies preliminary action by performing image deformation correction before watermark extraction. The system first identifies positioning blocks in the deformed image, then corrects the deformation based on the identified block positions, and only after this preliminary correction does it extract the watermark using the machine learning model. This ensures the watermark extraction is performed on a corrected, non-deformed image, significantly improving extraction success rate while maintaining operational simplicity.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If image deformation correction is performed before watermark extraction, then extraction success rate is improved, but the processing complexity increases

Engineering Contradiction:

Improveextraction success rateVSAvoidprocessing complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the watermark into distinct positioning blocks and information blocks. The positioning blocks are used to determine deformation parameters, while the information blocks contain the actual watermark data. This segmentation allows the system to separately handle deformation correction (using positioning blocks) and watermark extraction (using information blocks), improving extraction success rate while managing processing complexity through modular operations.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent uses positioning blocks as an intermediary element between the deformed image and the watermark extraction process. These positioning blocks serve as mediators that carry geometric information about the deformation, enabling the system to correct the image deformation without directly processing the entire image complexity. This intermediary approach improves extraction success rate while reducing overall processing complexity.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If positioning block identification is performed on deformed images, then the method handles real-world acquisition conditions, but the position relationship between blocks becomes inconsistent

Engineering Contradiction:

Improvehandling of acquisition conditionsVSAvoidposition relationship consistency

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The patent applies parameter changes by calculating deformation parameters based on the positions of positioning blocks in the deformed image. These parameters (such as affine transformation matrices) are then used to correct the image deformation. By changing the geometric parameters of the image based on the identified positioning block locations, the system adapts to real-world acquisition conditions while restoring consistent position relationships between blocks, enabling both adaptability and precision.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260024159A1Watermark Extraction Method, Model Training Method, Watermark Addition Method, and Electronic Apparatus

Publication Date: 2026.01.22 BOE TECHNOLOGY GROUP CO LTD

AI summary

The present disclosure provides a watermark extraction method, a model training method, a watermark addition method and an electronic device. The watermark extraction method includes the following steps acquiring an image to be processed embedded with a watermark, wherein the watermark includes a plurality of positioning blocks and a plurality of information blocks; identifying the image to be processed by a first identification model to identify at least part of positioning blocks among the plurality of the positioning blocks; deforming the image to be processed according to a current position relationship of the at least part of positioning blocks and a reference position relationship of the at least part of positioning blocks; extracting the plurality of information blocks from the deformed image to be processed; and identifying the plurality of the information blocks by a second identification model to obtain the watermark carried by the plurality of information blocks.