A new accurate pill recognition system using imprint information

Zhiyuan Chen, Seiichiro Kamata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Great achievements in modern medicine benefit human beings. Also, it has brought about an explosive growth of pharmaceuticals that current in the market. In daily life, pharmaceuticals sometimes confuse people when they are found unlabeled. In this paper, we propose an automatic pill recognition technique to solve this problem. It functions mainly based on the imprint feature of the pills, which is extracted by proposed MSWT (modified stroke width transform) and described by WSC (weighted shape context). Experiments show that our proposed pill recognition method can reach an accurate rate up to 92.03% within top 5 ranks when trying to classify more than 10 thousand query pill images into around 2000 categories.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
PublisherSPIE
Volume9067
ISBN (Print)9780819499967
DOIs
Publication statusPublished - 2013
Event6th International Conference on Machine Vision, ICMV 2013 - London
Duration: 2013 Nov 162013 Nov 17

Other

Other6th International Conference on Machine Vision, ICMV 2013
CityLondon
Period13/11/1613/11/17

Fingerprint

Pharmaceuticals
Drug products
strokes
Stroke
medicine
Pharmaceutical Preparations
Medicine
Classify
Query
Transform
Experiment
Experiments
Life
Context
Human
Market

Keywords

  • Feature extraction
  • Image retrieval
  • Modified Stroke Width Transform (MSWT)
  • Pill recognition
  • Weighted Shape Context (WSC)

ASJC Scopus subject areas

  • Applied Mathematics
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics

Cite this

Chen, Z., & Kamata, S. (2013). A new accurate pill recognition system using imprint information. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 9067). [906711] SPIE. https://doi.org/10.1117/12.2051168

A new accurate pill recognition system using imprint information. / Chen, Zhiyuan; Kamata, Seiichiro.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9067 SPIE, 2013. 906711.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, Z & Kamata, S 2013, A new accurate pill recognition system using imprint information. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 9067, 906711, SPIE, 6th International Conference on Machine Vision, ICMV 2013, London, 13/11/16. https://doi.org/10.1117/12.2051168
Chen Z, Kamata S. A new accurate pill recognition system using imprint information. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9067. SPIE. 2013. 906711 https://doi.org/10.1117/12.2051168
Chen, Zhiyuan ; Kamata, Seiichiro. / A new accurate pill recognition system using imprint information. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9067 SPIE, 2013.
@inproceedings{c22fc0bc45f24a70962608498bdb0527,
title = "A new accurate pill recognition system using imprint information",
abstract = "Great achievements in modern medicine benefit human beings. Also, it has brought about an explosive growth of pharmaceuticals that current in the market. In daily life, pharmaceuticals sometimes confuse people when they are found unlabeled. In this paper, we propose an automatic pill recognition technique to solve this problem. It functions mainly based on the imprint feature of the pills, which is extracted by proposed MSWT (modified stroke width transform) and described by WSC (weighted shape context). Experiments show that our proposed pill recognition method can reach an accurate rate up to 92.03{\%} within top 5 ranks when trying to classify more than 10 thousand query pill images into around 2000 categories.",
keywords = "Feature extraction, Image retrieval, Modified Stroke Width Transform (MSWT), Pill recognition, Weighted Shape Context (WSC)",
author = "Zhiyuan Chen and Seiichiro Kamata",
year = "2013",
doi = "10.1117/12.2051168",
language = "English",
isbn = "9780819499967",
volume = "9067",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",

}

TY - GEN

T1 - A new accurate pill recognition system using imprint information

AU - Chen, Zhiyuan

AU - Kamata, Seiichiro

PY - 2013

Y1 - 2013

N2 - Great achievements in modern medicine benefit human beings. Also, it has brought about an explosive growth of pharmaceuticals that current in the market. In daily life, pharmaceuticals sometimes confuse people when they are found unlabeled. In this paper, we propose an automatic pill recognition technique to solve this problem. It functions mainly based on the imprint feature of the pills, which is extracted by proposed MSWT (modified stroke width transform) and described by WSC (weighted shape context). Experiments show that our proposed pill recognition method can reach an accurate rate up to 92.03% within top 5 ranks when trying to classify more than 10 thousand query pill images into around 2000 categories.

AB - Great achievements in modern medicine benefit human beings. Also, it has brought about an explosive growth of pharmaceuticals that current in the market. In daily life, pharmaceuticals sometimes confuse people when they are found unlabeled. In this paper, we propose an automatic pill recognition technique to solve this problem. It functions mainly based on the imprint feature of the pills, which is extracted by proposed MSWT (modified stroke width transform) and described by WSC (weighted shape context). Experiments show that our proposed pill recognition method can reach an accurate rate up to 92.03% within top 5 ranks when trying to classify more than 10 thousand query pill images into around 2000 categories.

KW - Feature extraction

KW - Image retrieval

KW - Modified Stroke Width Transform (MSWT)

KW - Pill recognition

KW - Weighted Shape Context (WSC)

UR - http://www.scopus.com/inward/record.url?scp=84901315893&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901315893&partnerID=8YFLogxK

U2 - 10.1117/12.2051168

DO - 10.1117/12.2051168

M3 - Conference contribution

AN - SCOPUS:84901315893

SN - 9780819499967

VL - 9067

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - SPIE

ER -