Geo-tagged Image Establishment Identification via OCR and Mapping Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current techniques lack an effective method for computers to identify establishments, such as buildings and signs, in geo-tagged images, which are prevalent due to the use of digital cameras and positioning devices.
Innovation Solution
A computing environment with an image server and client system that utilizes geo-tagged images, optical character recognition (OCR), and establishment databases to detect and identify known establishments by recognizing text regions, comparing OCR results with nearby establishment information, and selecting representative images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional OCR engines are used to interpret street scene images, then text recognition can be performed, but accuracy is insufficient without mapping information constraints
Solution Approach 1:
The system performs preliminary actions by obtaining mapping information (establishment names, locations, boundaries) before OCR interpretation. This prior knowledge is used to constrain and guide the OCR engine, significantly improving text recognition accuracy without requiring complex real-time processing during image interpretation
Solution Approach 2:
Mapping information serves as an intermediary between the raw image data and the OCR engine. The established boundaries, names, and location data from mapping sources act as constraints that guide the OCR interpretation process, enabling accurate text recognition while maintaining system manageability
2Measurement precision
If mapping information is used as prior constraints for OCR, then text recognition accuracy improves significantly, but the system requires integration of multiple data sources
Solution Approach 1:
The system achieves multi-functionality by using mapping information for multiple purposes: constraining OCR interpretation, identifying establishments in images, and providing contextual information for image corpus analysis. This unified approach to data utilization improves accuracy while managing integration complexity through a single framework
Solution Approach 2:
The mapping information database serves itself by providing constraints and contextual data that enable the system to automatically identify establishments and interpret images without requiring external intervention. The established boundaries and names from mapping sources self-constrain the OCR process, improving accuracy while the system manages its own data integration needs
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Establishments are identified in geo-tagged images. According to one aspect, text regions are located in a geo-tagged image and text strings in the text regions are recognized using Optical Character Recognition (OCR) techniques. Text phrases are extracted from information associated with establishments known to be near the geographic location specified in the geo-tag of the image. The text strings recognized in the image are compared with the phrases for the establishments for approximate matches, and an establishment is selected as the establishment in the image based on the approximate matches. According to another aspect, text strings recognized in a collection of geo-tagged images are compared with phrases for establishments in the geographic area identified by the geo-tags to generate scores for image-establishment pairs. Establishments in each of the large collection of images as well as representative images showing each establishment are identified using the scores.