Related Experiment Video
Updated: Sep 27, 2025
12:08
From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
Published on: August 13, 2014
24.7K
A High-Precision Method for Segmentation and Recognition of Shopping Mall Plans
Ming Su1, Wei Shi1, Dangjun Zhao1
1School of Aeronautics and Astronautics, Central South University, Changsha 410083, China.
Sensors (Basel, Switzerland)
|April 12, 2022
Summary
This study presents a novel method for segmenting and recognizing rooms in shopping mall floor plans. The approach achieves high accuracy in identifying room locations and semantics, crucial for applications like robot navigation.
Area of Science:
- Computer Vision
- Spatial Data Analysis
- Architectural Informatics
Background:
- Limited research exists on segmenting and recognizing shopping mall floor plans compared to architectural plans.
- Accurate spatial and semantic information from mall plans is essential for various applications.
Purpose of the Study:
- To develop and evaluate a method for accurate segmentation and recognition of shopping mall floor plans.
- To extract room location and semantic information for enhanced spatial analysis and applications.
Main Methods:
- A two-stage region growth method for floor plan segmentation.
- Integration of optical character recognition (OCR) for room number identification.
- Matching OCR results with catalog data to determine room names and semantics.
Main Results:
- Achieved 92.54% accuracy in room segmentation.
- Achieved 90.56% accuracy in room recognition.
- Overall detection accuracy reached 83.81% on a dataset of 1340 rooms.
Conclusions:
- The proposed method effectively segments and recognizes rooms in shopping mall floor plans.
- The system successfully extracts crucial spatial and semantic information.
- This technique has potential applications in indoor navigation, building analysis, and 3D reconstruction.
