Bed Area Extraction Using Stationary Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for extracting the bed area in hospital settings face challenges due to variable bed positions and the inclusion of unrelated line segments, leading to high calculation burdens and inaccurate extractions.
Innovation Solution
A bed area extraction method that detects stationary linear edges and selects bed area candidates based on edge lengths and angles, employing a hierarchical process to filter out non-relevant edges and form candidate shapes, reducing the number of combinations and eliminating the need for pattern matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pattern matching based on bed shape is used to extract bed area, then bed area can be extracted, but calculation burden increases and accuracy decreases due to variable bed positions
Solution Approach 1:
The patent segments the bed area extraction process into multiple hierarchical stages: (1) detecting all linear edges in the image, (2) filtering to retain only stationary linear edges across multiple images, (3) selecting candidate edges based on length criteria, (4) forming U-shaped combinations from selected candidates, and (5) determining the final bed area from valid U-shapes. This segmentation reduces calculation burden by progressively eliminating non-relevant edges before performing complex shape analysis.
Solution Approach 2:
The patent performs preliminary filtering of linear edges by detecting which edges remain stationary across multiple captured images before proceeding to bed area extraction. This preliminary action eliminates edges from moving objects (patients, caregivers, equipment) before the main extraction process, significantly reducing the number of edges that need to be processed in subsequent steps and improving both efficiency and accuracy.
2Productivity
If all line segments are classified and averaged to extract bed area, then extraction is performed, but unrelated line segments reduce accuracy and increase computation
Solution Approach 1:
The patent performs preliminary filtering by capturing multiple images and identifying only those linear edges that remain stationary across all images. This preliminary action occurs before the main extraction process, eliminating edges from moving objects (patients, caregivers, equipment) and retaining only edges from fixed structures like the bed frame, thereby improving both efficiency and accuracy.
Solution Approach 2:
The patent extracts and isolates only the relevant stationary linear edges from the complete set of detected edges by comparing multiple images. This extraction process separates the bed-related edges from unrelated moving edges, allowing the algorithm to focus computation only on the relevant subset of edges that actually define the bed area.
Data Source
AI summary
A method of extracting a bed area of a bed performed by a computer, includes detecting linear edges which remain stationary among plural images; and selecting a bed area candidate based on lengths of the linear edges and angles between the linear edges which cross to each other in a U shape formed as combinations of the detected linear edges.