Peripheral Vision Display Using Skeleton Marker Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems struggle to allow drivers to recognize the type of object in their peripheral vision without turning their eyes, leading to increased risk of collisions due to the difference in visual sensitivity between foveal and peripheral vision.
Innovation Solution
A display control method and apparatus that uses a recognition unit to identify objects in the driver's peripheral vision and generates images based on biological motion patterns, allowing the driver to recognize the type of object through a pattern of markers representing the object's skeleton, which changes over time to indicate movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the driver pays attention to the road in front or vehicle ahead, then the driver can maintain focus on primary driving tasks, but the driver fails to detect pedestrians or vehicles in peripheral vision
Solution Approach 1:
The visual field is segmented into foveal vision (central 2-7 degrees) and peripheral vision regions. The system processes information from these different regions separately, using foveal vision for detailed detection of primary driving targets and peripheral vision for motion-sensitive detection of potential hazards, thereby resolving the contradiction between focused attention and peripheral awareness
Solution Approach 2:
The invention introduces an intermediary processing mechanism that takes input from both foveal and peripheral vision channels and integrates them to provide comprehensive situational awareness. This intermediary system allows the driver to maintain focus on primary tasks while still detecting peripheral objects through the specialized characteristics of each vision type
2Measurement precision
If the driver turns eyes to objects in peripheral vision, then the driver can recognize object details, but the driver loses continuous monitoring of the primary driving path
Solution Approach 1:
The system segments the visual processing tasks by assigning different functions to foveal and peripheral vision. Peripheral vision is used for initial detection and motion sensing of potential hazards, while foveal vision is engaged only when detailed examination is required. This segmentation eliminates the need for continuous eye movements while maintaining comprehensive monitoring
Solution Approach 2:
The system performs preliminary detection using peripheral vision's motion sensitivity to identify potential hazards before they require detailed examination. This preliminary action allows the driver to maintain continuous monitoring of the primary driving path while still being alerted to objects that need further attention
3Loss of information
If static objects are displayed in peripheral vision, then the driver can see object presence, but the driver cannot detect motion or type of object
Solution Approach 1:
The invention transforms static display into dynamic presentation by showing sequential images at different time points. This dynamic display leverages peripheral vision's motion sensitivity to draw driver attention while providing information about object type and motion state, resolving the contradiction between presence detection and detailed recognition
Solution Approach 2:
The system uses periodic display of images at different time points to create motion perception in the peripheral vision. This periodic action allows the driver to detect both the presence and motion characteristics of objects without requiring foveal attention, while still providing information about object type through the pattern of motion
Data Source
Figure 1
Figure 2
Figure 3A~3D
AI summary
A computer performs a process to determine whether an object is a predetermined object and a process to control a display unit to generate a first image based on a result of recognized object at a first timing and generate a second image based on a result of the recognized object at a second timing that is later than the first timing if the predetermined object is determined. The first image is an image formed by a pattern of markers representing a skeleton of the object, and the second image is an image formed by a pattern of markers corresponding to the pattern of markers in the first image, and the position of at least one marker of the pattern of markers in the first image differs from the position of the corresponding marker in the second image.