Hierarchical Structuring Method for Large-Capacity MMS Point Cloud Data

Seok Chan Kang, Jeongwon Lee, Hyun Sang Choi, Jiyeong Lee

Research output: Contribution to journalArticlepeer-review


Ground MMS (Mobile Mapping System) survey is a technology that acquires point cloud data including position (x, y, z) values, color (RGB) values, and intensity values by observing the surroundings while vehicles equipped with various sensors are driving on the road. The ground MMS survey obtains tens of gigabytes of point cloud data depending on the shooting distance, which is stored in a standard format, LAS (LASer) file. When storing large amounts of data as a single file, there is a problem with data inquiry, so it is stored separately in units of 1 gigabyte of less. If the divided and stored data is used as it is, the same task must be repeatedly performed, and finally, the separated data must be integrated again. Therefore, this study proposes a method of hierarchically structuring divided and stored LAS files so that they can be inquired continuously. Point group data structuring was organized hierarchically by applying octree-based spatial division techiniques to process large amounts of data, and file-based structuring was attempted to process ultra-large capacity without limitations on memory. To implement the method, dozens of divided point data were hierarchically structured and indexed to confirm the results of storage and inquiry as integrated data. As a result of applying the structure of seperating and storing files for each block through octree-based hierarchical structure, it is confirmed that tens of gigabytes of data can be viewed at once.

Original languageEnglish
Pages (from-to)109-125
Number of pages17
JournalJournal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
Issue number2
StatePublished - 2023


  • Ground MMS Survey
  • Hierarchical Structuring
  • Octree
  • Point Cloud Data


Dive into the research topics of 'Hierarchical Structuring Method for Large-Capacity MMS Point Cloud Data'. Together they form a unique fingerprint.

Cite this