Facility Management via Text Recognition for Road Map Noise Removal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for creating high-precision road maps struggle with accurately identifying and managing facility status, and removing noise such as dynamic objects, while being cost-effective and efficient.
Innovation Solution
A method for facility management through text recognition, which involves detecting facilities from images captured by vehicle-mounted cameras, recognizing text on these facilities, and identifying their types, using a computer program to execute these processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If AI models are used to identify objects in images for determining facility status, then identification accuracy is improved, but device cost increases due to requirement of expensive GPUs
Solution Approach 1:
The patent replaces expensive AI models requiring GPUs with a lightweight text recognition system using OCR technology. This substitutes costly computational resources with affordable, simple text processing algorithms that can run on standard processors, dramatically reducing device cost while maintaining facility identification capability through text-based recognition rather than complex image analysis
Solution Approach 2:
The patent substitutes the mechanical/computational system of AI-based object identification with a text-based recognition system. Instead of using complex neural networks and GPU-based image processing, the system uses OCR (optical character recognition) to read and interpret text on facilities, replacing expensive computational mechanics with simpler text processing mechanics
2Loss of information
If high-precision road maps include all captured objects, then map completeness is improved, but noise from dynamic objects increases
Solution Approach 1:
The patent extracts and removes dynamic objects from the captured image data before generating the high-precision road map. By identifying and excluding moving objects such as vehicles and pedestrians from the final map data, the system maintains completeness of static facility information while eliminating noise from dynamic elements that would otherwise contaminate the map
Solution Approach 2:
The patent introduces an intermediary filtering process between image capture and map generation. This intermediary step analyzes captured images to distinguish between static facilities (to be included in the map) and dynamic objects (to be excluded), acting as a mediator that separates useful information from noise while preserving map completeness
Data Source
AI summary
Proposed is a method for facility management through text recognition, capable of detecting a facility from an image captured by a camera mounted on a vehicle that travels on the road and recognizing text written on the detected facility to identify the type of facility. The method for facility management includes identifying, by a data processing device, an object corresponding to a preset facility on an image captured by a camera, recognizing, by the data processing device, text included in the identified object, and identifying, by the data processing device, a type of facility corresponding to the identified object based on the recognized text. The present method is technology developed with support from the Ministry of Land, Infrastructure and Transport/Korea Agency for Land, Infrastructure and Transport Science and Technology Promotion (task number RS2021-KA160637).