Chrominance-Based Shadow Detection for Accurate Lawn Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing shadow detection methods often misjudge lawn areas in shadow or under strong light as non-lawn areas, leading to inaccurate segmentation.
Innovation Solution
An image shadow detection method that utilizes chrominance components and feature values to segment images, adjusting segmentation thresholds based on peak and valley detection in chrominance histograms, and employing different segmentation rules based on shadow presence to reduce misjudgment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If shadow detection methods based on geometric models or shadow features are used, then shadow regions can be identified, but lawn areas in shadow or under strong light are misjudged as non-lawn areas
Solution Approach 1:
The patent segments the image into different regions (lawn area and non-lawn area) based on chrominance components, and applies different processing strategies for shadow detection in different regions. This regional segmentation allows the system to handle shadow and strong light conditions differently, preventing misjudgment of lawn areas.
Solution Approach 2:
The patent dynamically adjusts the shadow detection threshold based on the chrominance characteristics of the image. By calculating the average chrominance value of the lawn area and comparing it with standard values, the system adapts the detection parameters to different lighting conditions, preventing misjudgment caused by fixed thresholds.
2Device complexity
If fixed segmentation thresholds are used, then segmentation process is simple, but shadow areas are misjudged as non-lawn areas
Solution Approach 1:
The patent transforms the fixed segmentation threshold into a dynamic threshold that adapts to different images. The system calculates the average chrominance value of the lawn area in each image and uses this to determine the segmentation threshold, making the threshold dynamic rather than fixed. This improves segmentation accuracy while maintaining reasonable computational complexity.
Solution Approach 2:
The system performs self-adjustment by automatically calculating the average chrominance value of the lawn area from the input image itself, without requiring external calibration or manual parameter setting. This self-service approach allows the segmentation threshold to adapt to each specific image while keeping the process automated and relatively simple.
Data Source
AI summary
Disclosed is an image shadow detection method, a system, an image segmentation device, and a readable storage medium. The method comprises the following steps: acquiring a chrominance component of an image; segmenting the image by a first segmentation rule based on the chrominance components of the image and the preset chrominance interval and acquiring a first segmentation result, the first segmentation result comparing a first region and a second region; calculating feature values of the first region or the second region in the image; carrying out the comparison processing to judge whether the image has shadow based on the feature value and the preset threshold value. According to the present disclosure, the image is segmented by the first segmentation rule and the first segmentation result is acquired. The first segmentation result is acquired based on the chrominance component and is not affected by shadows. The shadow area is identified by combining the first segmentation result with the judgment of feature values, thus reducing the possibility of shadow misjudgment.