Image Segmentation Threshold Value Deciding Method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image segmentation methods based on brightness are prone to errors, particularly when the background and hand images have similar brightness or when environmental interference occurs, leading to incorrect determination of hand size and location.
Innovation Solution
An image segmentation threshold value deciding method that dynamically adjusts threshold values by defining multiple image regions, calculating corresponding threshold values, and selecting the optimal value based on segmentation scores, which can include average, maximum, or minimum brightness, and adjusting ratios, to improve image segmentation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single image segmentation threshold value is used for the entire image, then the device complexity is low, but the segmentation accuracy deteriorates when background and hand have similar brightness
Solution Approach 1:
The patent divides the image into multiple regions (first image region and second image region) and calculates different threshold values for each region. This regional segmentation allows the system to adapt to different brightness conditions in different parts of the image, improving segmentation accuracy when background and hand have similar brightness overall but differ in local regions.
Solution Approach 2:
The patent applies different threshold values to different image regions based on their specific brightness characteristics. By calculating threshold values locally for each region rather than using a global threshold, the system achieves better segmentation accuracy in each specific region while managing complexity through focused local analysis.
2Measurement precision
If multiple image regions are defined and different threshold values are calculated for each region, then the segmentation accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent segments the image into multiple regions and calculates threshold values for each region independently. This approach improves hand location determination accuracy by adapting to local brightness variations, while the segmentation framework provides a structured method to manage the increased processing complexity.
Solution Approach 2:
The patent calculates threshold values for multiple regions rather than the entire image at once. This partial action approach allows the system to focus computational resources on specific regions of interest, improving accuracy where needed while managing overall processing complexity through selective analysis.
3Reliability
If a fixed threshold value is used for image segmentation, then the ease of operation is high, but the reliability of gesture detection deteriorates under varying environmental light conditions
Solution Approach 1:
The patent dynamically calculates threshold values based on the actual brightness distribution in different image regions, rather than using a fixed threshold. This dynamic adaptation to varying environmental light conditions significantly improves gesture detection reliability, while the automated calculation process minimizes the need for manual threshold adjustment.
Solution Approach 2:
The system automatically calculates and adjusts threshold values for each image region based on its brightness characteristics, without requiring manual intervention. This self-adjusting mechanism ensures reliable gesture detection under varying environmental conditions while maintaining ease of operation through automation.
4Measurement precision
If the image segmentation threshold value is adjusted dynamically based on region brightness, then the segmentation accuracy is improved, but the processing time increases
Solution Approach 1:
The patent divides the image into multiple regions and calculates threshold values for each region separately. This segmentation approach improves segmentation accuracy by adapting to local brightness variations, while the modular regional processing can be optimized to reduce overall processing time compared to analyzing the entire image uniformly.
Solution Approach 2:
The patent focuses threshold calculation on specific image regions rather than the entire image. This partial action approach concentrates computational resources on areas where segmentation is needed, improving accuracy while potentially reducing total processing time by avoiding unnecessary calculations in irrelevant regions.
Data Source
AI summary
An image segmentation threshold value determining method, comprising: defining a plurality of image regions of a first sensing image; determining a first, a second part image segmentation threshold values according to a first, second image regions of the image regions; performing a first, a second image segmentation operation to the first sensing image according to the first, the second part image segmentation threshold values to acquire a first, a second segmented images; and selecting one of the first part image segmentation threshold value and the second part image segmentation threshold value as the first image segmentation threshold value according to the first segmented image and the second segmented image.