Conveyance Image Positioning Under Variable Illuminance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conveyance apparatuses face challenges in accurately controlling the movement of objects in varying environmental conditions, such as changes in weather, time, date, location, and illuminance, due to failures in feature point matching when environmental conditions differ from those at the time of feature point map generation, leading to increased processing time and effort.
Innovation Solution
A control apparatus that utilizes both classical image processing and a deep learning model to detect feature points, generating and referencing multiple feature point maps to accurately calculate the position of a target object, even under varying conditions, without significantly increasing processing time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple feature point maps are generated based on different illuminances to address environmental variations, then the reliability of position estimation is improved, but the processing time and device complexity increase significantly
Solution Approach 1:
The system performs preliminary action by generating multiple feature point maps in advance under different illuminance conditions and storing them. During runtime, the appropriate pre-generated map is selected based on current environmental conditions, avoiding the need to generate maps in real-time and thus reducing processing time while maintaining reliability
Solution Approach 2:
The invention changes the parameter of illuminance by creating feature point maps under different illuminance levels. The system adapts to environmental variations by selecting the map that matches current lighting conditions, thereby maintaining position estimation reliability without requiring real-time map generation
2Reliability
If multiple feature point maps are generated based on different illuminances to address environmental variations, then the reliability of position estimation is improved, but the effort and complexity of map creation increase
Solution Approach 1:
Multiple feature point maps are created in advance under controlled conditions with different illuminances. This preliminary preparation work is done offline, separating the complex map creation process from real-time operation, thereby reducing the perceived complexity during actual use while improving reliability
3Measurement precision
If feature point matching is performed to estimate object position, then the accuracy of position estimation is improved, but the matching may fail when environmental conditions differ from map generation conditions
Solution Approach 1:
The system addresses environmental variations by changing the parameter of illuminance across multiple feature point maps. When environmental conditions differ from map generation conditions, the system selects the map with matching illuminance characteristics, thereby maintaining both high position estimation accuracy and reliable feature point matching
Data Source
AI summary
A storage device stores a first and a second feature point maps of an object in advance, the first feature point map detected by image processing not based on deep learning, and the second feature point map detected using a deep learning model trained in advance. A processing circuit detects first feature points of the object by the image processing on a captured image. The processing circuit detects second feature points of the object from the captured image using the deep learning model. The processing circuit calculates a position of a target object based on the first feature points and the first feature point map. When the position cannot be calculated based on the first feature points and the first feature point map, the processing circuit calculates the position of the target object based on the first and second feature points, and he first and second feature point maps.