Endoscope Focus Control via Weighted Region Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Endoscope systems with high-pixel image sensors experience shallow depth of field, leading to difficulties in focusing on tissues during procedures due to obstacles like treatment tools, which have higher contrast and are inadvertently brought into focus, requiring cumbersome user intervention to adjust autofocus settings.
Innovation Solution
A focus control device that sets regions in an image, determines the direction of a target focusing position relative to a reference, and performs weighted comparisons between near and far area information to control the in-focus object plane position, reducing the impact of high-contrast obstacles and allowing automatic focus adjustment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a treatment tool with higher contrast than tissue is present in the field of view, then the treatment tool is brought into focus by conventional autofocus, but the tissue fails to be brought into focus
Solution Approach 1:
The image is divided into multiple regions, and focus direction is determined independently for each region. The processor segments the image into first and second regions, evaluates focus direction for each segment separately, and then integrates the results. This allows the system to identify that the treatment tool in one region should not dictate the focus direction for the entire image, thereby preventing high-contrast obstacles from incorrectly controlling autofocus.
Solution Approach 2:
Different regions of the image are assigned different weights based on their importance. The processor determines a weight for each region and performs weighted integration of focus direction information. Regions containing the treatment tool can be assigned lower weights, while regions containing the tissue of interest receive higher weights, allowing local quality assessment to override the global autofocus decision that would otherwise be dominated by high-contrast obstacles.
2Ease of operation
If conventional autofocus is used to bring high-contrast treatment tools into focus, then the treatment tool becomes visible, but the tissue remains out of focus requiring user intervention
Solution Approach 1:
The autofocus system performs self-correction by automatically identifying and excluding regions containing treatment tools from controlling the focus direction. The processor autonomously evaluates focus direction for each region, determines appropriate weights, and integrates the results without user input. This self-service capability allows the system to automatically prioritize tissue over treatment tools, eliminating the need for user intervention to correct focus errors caused by high-contrast obstacles.
3Measurement precision
If the depth of field is shallow due to using an image sensor with a large number of pixels, then the imaging resolution is improved, but the range of in-focus objects is reduced making autofocus more difficult
Solution Approach 1:
The system dynamically adjusts the focus direction for each region based on real-time image analysis. Rather than using a fixed focus plane, the processor continuously evaluates the focus direction (NEAR or FAR) for each region and adjusts the autofocus accordingly. This dynamic adaptation allows the system to maintain high imaging resolution while compensating for the limited depth of field by actively selecting the appropriate focus direction for different regions containing tissue versus treatment tools.
Data Source
AI summary
A focus control device includes a processor including hardware. The processor sets a plurality of regions, each including a plurality of pixels, to an image acquired by an imaging section, obtains a direction determination result for each region in some or all of the plurality of regions set, by determining whether a target focusing position that is a target of an in-focus object plane position is on a NEAR side or a FAR side relative to a reference position, determines an in-focus direction by performing weighted comparison between NEAR area information and FAR area information, based on the direction determination result and weight information, and controls the in-focus object plane position based on the in-focus direction.