Classification with reject option using contextual information

Filipe Condessa, Jose Bioucas-Dias, Carlos A. Castro, John A. Ozolek, Jelena Kovacevic

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

Abstract

We propose a new algorithm for classification that merges classification with reject option with classification using contextual information. A reject option is desired in many image-classification applications requiring a robust classifier and when the need for high classification accuracy surpasses the need to classify the entire image. Moreover, our algorithm improves the classifier performance by including local and nonlocal contextual information, at the expense of rejecting a fraction of the samples. As a probabilistic model, we adopt a multinomial logistic regression. We use a discriminative random model for the description of the problem; we introduce reject option into the classification problem through association potential, and contextual information through interaction potential. We validate the method on the images of H&E-stained teratoma tissues and show the increase in the classifier performance when rejecting part of the assigned class labels.

Original languageEnglish (US)
Title of host publicationISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging
Subtitle of host publicationFrom Nano to Macro
Pages1340-1343
Number of pages4
DOIs
StatePublished - Aug 22 2013
Event2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013 - San Francisco, CA, United States
Duration: Apr 7 2013Apr 11 2013

Other

Other2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013
CountryUnited States
CitySan Francisco, CA
Period4/7/134/11/13

Fingerprint

Classifiers
Image classification
Teratoma
Statistical Models
Logistics
Labels
Tissue
Logistic Models

Keywords

  • discriminative random fields
  • image classification
  • reject option

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Condessa, F., Bioucas-Dias, J., Castro, C. A., Ozolek, J. A., & Kovacevic, J. (2013). Classification with reject option using contextual information. In ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro (pp. 1340-1343). [6556780] https://doi.org/10.1109/ISBI.2013.6556780

Classification with reject option using contextual information. / Condessa, Filipe; Bioucas-Dias, Jose; Castro, Carlos A.; Ozolek, John A.; Kovacevic, Jelena.

ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro. 2013. p. 1340-1343 6556780.

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

Condessa, F, Bioucas-Dias, J, Castro, CA, Ozolek, JA & Kovacevic, J 2013, Classification with reject option using contextual information. in ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro., 6556780, pp. 1340-1343, 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2013, San Francisco, CA, United States, 4/7/13. https://doi.org/10.1109/ISBI.2013.6556780
Condessa F, Bioucas-Dias J, Castro CA, Ozolek JA, Kovacevic J. Classification with reject option using contextual information. In ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro. 2013. p. 1340-1343. 6556780 https://doi.org/10.1109/ISBI.2013.6556780
Condessa, Filipe ; Bioucas-Dias, Jose ; Castro, Carlos A. ; Ozolek, John A. ; Kovacevic, Jelena. / Classification with reject option using contextual information. ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging: From Nano to Macro. 2013. pp. 1340-1343
@inproceedings{dd60f02be50340b990f93e2a8172bbaa,
title = "Classification with reject option using contextual information",
abstract = "We propose a new algorithm for classification that merges classification with reject option with classification using contextual information. A reject option is desired in many image-classification applications requiring a robust classifier and when the need for high classification accuracy surpasses the need to classify the entire image. Moreover, our algorithm improves the classifier performance by including local and nonlocal contextual information, at the expense of rejecting a fraction of the samples. As a probabilistic model, we adopt a multinomial logistic regression. We use a discriminative random model for the description of the problem; we introduce reject option into the classification problem through association potential, and contextual information through interaction potential. We validate the method on the images of H&E-stained teratoma tissues and show the increase in the classifier performance when rejecting part of the assigned class labels.",
keywords = "discriminative random fields, image classification, reject option",
author = "Filipe Condessa and Jose Bioucas-Dias and Castro, {Carlos A.} and Ozolek, {John A.} and Jelena Kovacevic",
year = "2013",
month = "8",
day = "22",
doi = "10.1109/ISBI.2013.6556780",
language = "English (US)",
isbn = "9781467364546",
pages = "1340--1343",
booktitle = "ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Classification with reject option using contextual information

AU - Condessa, Filipe

AU - Bioucas-Dias, Jose

AU - Castro, Carlos A.

AU - Ozolek, John A.

AU - Kovacevic, Jelena

PY - 2013/8/22

Y1 - 2013/8/22

N2 - We propose a new algorithm for classification that merges classification with reject option with classification using contextual information. A reject option is desired in many image-classification applications requiring a robust classifier and when the need for high classification accuracy surpasses the need to classify the entire image. Moreover, our algorithm improves the classifier performance by including local and nonlocal contextual information, at the expense of rejecting a fraction of the samples. As a probabilistic model, we adopt a multinomial logistic regression. We use a discriminative random model for the description of the problem; we introduce reject option into the classification problem through association potential, and contextual information through interaction potential. We validate the method on the images of H&E-stained teratoma tissues and show the increase in the classifier performance when rejecting part of the assigned class labels.

AB - We propose a new algorithm for classification that merges classification with reject option with classification using contextual information. A reject option is desired in many image-classification applications requiring a robust classifier and when the need for high classification accuracy surpasses the need to classify the entire image. Moreover, our algorithm improves the classifier performance by including local and nonlocal contextual information, at the expense of rejecting a fraction of the samples. As a probabilistic model, we adopt a multinomial logistic regression. We use a discriminative random model for the description of the problem; we introduce reject option into the classification problem through association potential, and contextual information through interaction potential. We validate the method on the images of H&E-stained teratoma tissues and show the increase in the classifier performance when rejecting part of the assigned class labels.

KW - discriminative random fields

KW - image classification

KW - reject option

UR - http://www.scopus.com/inward/record.url?scp=84881642953&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881642953&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2013.6556780

DO - 10.1109/ISBI.2013.6556780

M3 - Conference contribution

SN - 9781467364546

SP - 1340

EP - 1343

BT - ISBI 2013 - 2013 IEEE 10th International Symposium on Biomedical Imaging

ER -