Human identification using individual dental radiograph records

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

The goal of forensic dentistry is to identify individuals based on their dental characteristics. In this chapter, we present a system for automating that process by identifying people from dental x-ray images. Given a dental record of a postmortem (PM), the proposed system retrieves the best matches from an antemortem (AM) database. The system automatically segments dental x-ray images into individual teeth and extracts representative feature vectors for each tooth, which are later used for retrieval. We developed a new method for teeth segmentation and three different methods for representing and matching teeth. The system relies on the three different types of dental radiographs for identification. The identification procedure is carried out by integrating the results of using an individual’s three available types of dental radiographs. In this chapter, we address the problem of identifying individuals based on more than one type of dental radiographs for the first time, where three available types of dental radiographs are used. The experimental results show that using the three types of dental radiographs enhances the overall human identification procedure. Also, the experimental results of the different modules and the results of fusing the matching techniques are presented. To increase the accuracy of the identification process, the three matching techniques are fused together to improve the overall performance. We introduce some scenarios for fusing the three matchers at the score level as well as at the decision level.

Original languageEnglish (US)
Title of host publicationCase Studies in Intelligent Computing
Subtitle of host publicationAchievements and Trends
PublisherCRC Press
Pages329-362
Number of pages34
ISBN (Electronic)9781482207040
ISBN (Print)9781482207033
DOIs
StatePublished - Jan 1 2014

Fingerprint

Dentistry
X rays

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Nomir, O., & Abdel-Mottaleb, M. (2014). Human identification using individual dental radiograph records. In Case Studies in Intelligent Computing: Achievements and Trends (pp. 329-362). CRC Press. https://doi.org/10.1201/b17333

Human identification using individual dental radiograph records. / Nomir, Omaima; Abdel-Mottaleb, Mohamed.

Case Studies in Intelligent Computing: Achievements and Trends. CRC Press, 2014. p. 329-362.

Research output: Chapter in Book/Report/Conference proceedingChapter

Nomir, O & Abdel-Mottaleb, M 2014, Human identification using individual dental radiograph records. in Case Studies in Intelligent Computing: Achievements and Trends. CRC Press, pp. 329-362. https://doi.org/10.1201/b17333
Nomir O, Abdel-Mottaleb M. Human identification using individual dental radiograph records. In Case Studies in Intelligent Computing: Achievements and Trends. CRC Press. 2014. p. 329-362 https://doi.org/10.1201/b17333
Nomir, Omaima ; Abdel-Mottaleb, Mohamed. / Human identification using individual dental radiograph records. Case Studies in Intelligent Computing: Achievements and Trends. CRC Press, 2014. pp. 329-362
@inbook{990d3163111c446fa26d57c99d5534a6,
title = "Human identification using individual dental radiograph records",
abstract = "The goal of forensic dentistry is to identify individuals based on their dental characteristics. In this chapter, we present a system for automating that process by identifying people from dental x-ray images. Given a dental record of a postmortem (PM), the proposed system retrieves the best matches from an antemortem (AM) database. The system automatically segments dental x-ray images into individual teeth and extracts representative feature vectors for each tooth, which are later used for retrieval. We developed a new method for teeth segmentation and three different methods for representing and matching teeth. The system relies on the three different types of dental radiographs for identification. The identification procedure is carried out by integrating the results of using an individual’s three available types of dental radiographs. In this chapter, we address the problem of identifying individuals based on more than one type of dental radiographs for the first time, where three available types of dental radiographs are used. The experimental results show that using the three types of dental radiographs enhances the overall human identification procedure. Also, the experimental results of the different modules and the results of fusing the matching techniques are presented. To increase the accuracy of the identification process, the three matching techniques are fused together to improve the overall performance. We introduce some scenarios for fusing the three matchers at the score level as well as at the decision level.",
author = "Omaima Nomir and Mohamed Abdel-Mottaleb",
year = "2014",
month = "1",
day = "1",
doi = "10.1201/b17333",
language = "English (US)",
isbn = "9781482207033",
pages = "329--362",
booktitle = "Case Studies in Intelligent Computing",
publisher = "CRC Press",

}

TY - CHAP

T1 - Human identification using individual dental radiograph records

AU - Nomir, Omaima

AU - Abdel-Mottaleb, Mohamed

PY - 2014/1/1

Y1 - 2014/1/1

N2 - The goal of forensic dentistry is to identify individuals based on their dental characteristics. In this chapter, we present a system for automating that process by identifying people from dental x-ray images. Given a dental record of a postmortem (PM), the proposed system retrieves the best matches from an antemortem (AM) database. The system automatically segments dental x-ray images into individual teeth and extracts representative feature vectors for each tooth, which are later used for retrieval. We developed a new method for teeth segmentation and three different methods for representing and matching teeth. The system relies on the three different types of dental radiographs for identification. The identification procedure is carried out by integrating the results of using an individual’s three available types of dental radiographs. In this chapter, we address the problem of identifying individuals based on more than one type of dental radiographs for the first time, where three available types of dental radiographs are used. The experimental results show that using the three types of dental radiographs enhances the overall human identification procedure. Also, the experimental results of the different modules and the results of fusing the matching techniques are presented. To increase the accuracy of the identification process, the three matching techniques are fused together to improve the overall performance. We introduce some scenarios for fusing the three matchers at the score level as well as at the decision level.

AB - The goal of forensic dentistry is to identify individuals based on their dental characteristics. In this chapter, we present a system for automating that process by identifying people from dental x-ray images. Given a dental record of a postmortem (PM), the proposed system retrieves the best matches from an antemortem (AM) database. The system automatically segments dental x-ray images into individual teeth and extracts representative feature vectors for each tooth, which are later used for retrieval. We developed a new method for teeth segmentation and three different methods for representing and matching teeth. The system relies on the three different types of dental radiographs for identification. The identification procedure is carried out by integrating the results of using an individual’s three available types of dental radiographs. In this chapter, we address the problem of identifying individuals based on more than one type of dental radiographs for the first time, where three available types of dental radiographs are used. The experimental results show that using the three types of dental radiographs enhances the overall human identification procedure. Also, the experimental results of the different modules and the results of fusing the matching techniques are presented. To increase the accuracy of the identification process, the three matching techniques are fused together to improve the overall performance. We introduce some scenarios for fusing the three matchers at the score level as well as at the decision level.

UR - http://www.scopus.com/inward/record.url?scp=85054460500&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054460500&partnerID=8YFLogxK

U2 - 10.1201/b17333

DO - 10.1201/b17333

M3 - Chapter

SN - 9781482207033

SP - 329

EP - 362

BT - Case Studies in Intelligent Computing

PB - CRC Press

ER -