Map Generation Using Street View Image Recognition for Isolation Belt Marking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic maps primarily focus on roads and directions, failing to provide refined services necessary for ensuring safety and accuracy, as they do not effectively mark isolation belts such as green belts and fences, which are crucial for navigation.
Innovation Solution
A method and apparatus that utilize a street view image input into a recognition model to obtain and correct data on candidate isolation belts, incorporating their information and probability to accurately mark isolation belts on a base map, enhancing navigation performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If existing electronic maps focus only on roads and directions, then the map structure remains simple, but the navigation service refinement and safety information completeness deteriorate
Solution Approach 1:
The patent segments the map data into multiple layers: basic road information layer and isolated belt information layer. The isolated belts are further segmented into different types (green belts, fences, etc.) and marked with distinct icons. This segmentation allows the map to maintain a simple overall structure while incorporating detailed navigation information without overwhelming complexity.
Solution Approach 2:
The patent adds a new dimension of information by incorporating isolated belt data alongside traditional road and direction information. This creates a multi-dimensional map representation that includes both conventional navigation elements and safety-critical isolated belt markers, enabling more comprehensive navigation services without fundamentally restructuring the entire map system.
2Device complexity
If isolation belts are not marked on the map, then the map maintains simplicity, but the safety information and navigation accuracy deteriorate
Solution Approach 1:
The patent extracts isolated belt information from street view images using image recognition technology. The extraction process identifies and separates isolated belt features (green belts, fences, etc.) from the complex street scene, then marks them on the map with specific icons. This extraction ensures safety information is captured and displayed without requiring the entire map to become overly complex.
Solution Approach 2:
The patent uses different colored icons to represent different types of isolated belts on the map. This visual differentiation allows users to quickly identify various isolated belt types (green belts, fences, etc.) without increasing textual or structural complexity. The color-coding system efficiently communicates safety information through intuitive visual cues.
3Measurement precision
If manual marking of isolation belts is used, then marking accuracy may be high, but the processing time and labor cost increase
Solution Approach 1:
The patent replaces the manual mechanical process of marking isolated belts with an automated image recognition system. The system processes street view images through neural networks to automatically identify and mark isolated belts on the map. This substitution dramatically reduces processing time and labor requirements while maintaining high accuracy through the sophisticated recognition algorithms.
Solution Approach 2:
The patent enables the map system to automatically identify and mark isolated belts without human intervention. The image recognition model processes street view images and autonomously extracts isolated belt information, then integrates it into the map data structure. This self-service capability eliminates the need for manual marking operations while maintaining high precision through automated detection algorithms.
Data Source
AI summary
The present disclosure discloses a method and an apparatus for generating a map, relating to a field of intelligent transportation technologies in the field of computer technologies. The method includes the following. A street view image corresponding to a target road is obtained. The street view image is input into a preset deep learning model to obtain data of a candidate isolation belt corresponding to the street view image. The data of the candidate isolation belt includes information of the candidate isolation belt and a probability corresponding to the information of the candidate isolation belt. The information of the candidate isolation belt is corrected based on a preset correction strategy and the probability to obtain information of a target isolation belt. An icon corresponding to the information of the target isolation belt is added on a base map including the target road to generate a map.