Automating Information Extraction from 3-D Scan Data.

Abstract

The object of the research reported here was to evaluate the success of ARNScan versions 7.1 and 7.2, newly developed software for extracting body measurements from 3-D scans. Investigators used traditional methods to measure 123 male and female subjects for 21 dimensions associated with the sizing and design of military clothing. The same subjects were then scanned, and the same or comparable measurements extracted from the 3-D images. A comparison of the results obtained by each method yielded a number of differences. A variety of statistical procedures were then undertaken to establish whether these differences were important or significant; that is, were they large enough to place subjects in different sizes? Differences were tested against three standards: (1) acceptable measurer error established in the 1988 U.S. Army survey (ANSUR), (2) acceptable error estimated by three experienced tailors, and (3) garment grade in traditionally sized dress clothing. ARNScan fared well against the garment grade criterion; it did not do well against the ANSUR standard. The comparison produced mixed results with respect to the tailors' standards. ARNScan is still in the development stage and is currently being refined.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 14, 1998
Accession Number
ADA361314

Entities

People

  • Bruce Bradtmiller

Tags

Communities of Interest

  • Air Platforms
  • Cyber
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Algorithms
  • Body Regions
  • Clothing
  • Computer Programs
  • Data Analysis
  • Databases
  • Descriptive Analytics
  • Extraction
  • Information Science
  • Measurement
  • Shoulder
  • Standards
  • Statistical Analysis
  • Statistics
  • Surveys
  • Three Dimensional
  • Upper Extremity

Readers

  • Gender and Food Studies
  • Regression Analysis.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference