Vehicle Navigation Device Using Image Processing for Driving Side Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for determining the side of the road on which traffic drives, such as GPS-based systems, require direct input or reliance on driver knowledge, and are not effective in automatically adapting to changes across international borders without additional processing overhead.
Innovation Solution
A device and method using image processing and dictionary information to analyze road signs, extracting features based on pixel values and spatial relationships, allowing determination of the side of the road traffic drives on, without relying on GPS or direct driver input, by utilizing pictogram extraction and color classification techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If GPS-based systems are used to determine the side of the road, then the system can provide navigation information, but the system requires direct input or driver knowledge and cannot automatically adapt to changes across international borders
Solution Approach 1:
The system automatically determines the driving side by processing images of road signs and traffic scenes captured by the vehicle's camera. The processing unit autonomously extracts features from road signs, compares them against a dictionary of known road sign patterns, and determines the driving side without requiring driver input or manual GPS configuration, enabling the system to self-adapt when crossing international borders
Solution Approach 2:
The patent replaces GPS-based determination with an image processing system that uses computer vision to analyze road signs and traffic scenes. The processing unit extracts visual features from captured images, such as the position and orientation of road signs relative to the lane, and uses pattern recognition to automatically determine whether traffic drives on the left or right side of the road
2Measurement precision
If image processing techniques are used to analyze road signs, then accurate determination of driving side can be achieved, but additional processing overhead is introduced
Solution Approach 1:
The processing unit extracts only the essential features from road sign images that are necessary for determining the driving side, such as the position of the sign relative to the lane, the orientation of the sign, and key visual patterns. By extracting only these critical features rather than processing the entire image in detail, the system achieves accurate determination while minimizing processing overhead
Solution Approach 2:
The system pre-processes images by identifying and isolating road signs before detailed analysis. The processing unit first detects the presence of road signs in the captured image, extracts the region of interest containing the road sign, and then performs feature extraction and comparison. This preliminary action reduces the amount of data that requires intensive processing, thereby reducing overall processing overhead
Data Source
Figure 1~3e
Figure 4
Figure 5
AI summary
The present invention relates to a device (10) for determining vehicle navigation information. It is described to provide (210) an image relating to a scene external to a vehicle. Dictionary information is provided (220), wherein the dictionary information is generated on the basis of a plurality of images, and wherein at least some of the image data of the plurality of images is image data associated with at least one road sign. It is determined (230) whether traffic, in a location where the image was acquired, drives on the right or left of roads in the direction of travel on the basis of image processing of the image and the dictionary information. The image processing comprises: determining (232) an areal position in the image, where image data at the areal position is associated with a road sign of the at least one sign; and determining (234) a spatial relationship between a first feature extracted from the image data at the areal position and a second feature extracted from the image data at the areal position, wherein the first feature is extracted on the basis of at least one pixel value associated with a first colour and the second feature is extracted on the basis of at least one pixel value associated with a second colour. Driving information relating to whether traffic drives on the right or left is then output (240).