Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

Abstract

The manual postmortem inspection of broilers and their viscera is becoming a bottleneck as the slaughter rate increases. Computer vision can assist veterinarians during the inspection, but specular highlights can hide crucial details when inspecting for diseases on the viscera set. This study aims to restore details behind these specular highlights by capturing two images of the same viscera using shifting light positions. The dataset consists of images captured in-line at a poultry processing plant. The method achieves an average SSIM score of 0.96 over a test set of 100 image sets. The result is visually pleasing images with correct textural information instead of specular highlights.
Original languageEnglish
Title of host publicationComputer Vision – ACCV 2018 Workshops : 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers
EditorsGustavo Carneiro, Shaodi You
Number of pages16
PublisherSpringer
Publication dateJun 2019
Pages357-372
ISBN (Print)978-3-030-21073-1
ISBN (Electronic)978-3-030-21074-8
DOIs
Publication statusPublished - Jun 2019
EventAsian Conference on Computer Vision - Perth, Australia
Duration: 2 Dec 20186 Dec 2018

Conference

ConferenceAsian Conference on Computer Vision
CountryAustralia
CityPerth
Period02/12/201806/12/2018
SeriesLecture Notes in Computer Science
Volume11367
ISSN0302-9743

Fingerprint

Inspection
Poultry
Computer vision
Processing

Cite this

Jørgensen, A., Pedersen, M., Gade, R., Fagertun, J., & Moeslund, T. B. (2019). Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera. In G. Carneiro, & S. You (Eds.), Computer Vision – ACCV 2018 Workshops: 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers (pp. 357-372). Springer. Lecture Notes in Computer Science, Vol.. 11367 https://doi.org/10.1007/978-3-030-21074-8_30
Jørgensen, Anders ; Pedersen, Malte ; Gade, Rikke ; Fagertun, Jens ; Moeslund, Thomas B. / Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera. Computer Vision – ACCV 2018 Workshops: 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. editor / Gustavo Carneiro ; Shaodi You. Springer, 2019. pp. 357-372 (Lecture Notes in Computer Science, Vol. 11367).
@inproceedings{3dd422f225bf424bbbdaaa42153b141f,
title = "Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera",
abstract = "The manual postmortem inspection of broilers and their viscera is becoming a bottleneck as the slaughter rate increases. Computer vision can assist veterinarians during the inspection, but specular highlights can hide crucial details when inspecting for diseases on the viscera set. This study aims to restore details behind these specular highlights by capturing two images of the same viscera using shifting light positions. The dataset consists of images captured in-line at a poultry processing plant. The method achieves an average SSIM score of 0.96 over a test set of 100 image sets. The result is visually pleasing images with correct textural information instead of specular highlights.",
author = "Anders J{\o}rgensen and Malte Pedersen and Rikke Gade and Jens Fagertun and Moeslund, {Thomas B.}",
year = "2019",
month = "6",
doi = "10.1007/978-3-030-21074-8_30",
language = "English",
isbn = "978-3-030-21073-1",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "357--372",
editor = "Gustavo Carneiro and Shaodi You",
booktitle = "Computer Vision – ACCV 2018 Workshops",
address = "Germany",

}

Jørgensen, A, Pedersen, M, Gade, R, Fagertun, J & Moeslund, TB 2019, Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera. in G Carneiro & S You (eds), Computer Vision – ACCV 2018 Workshops: 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. Springer, Lecture Notes in Computer Science, vol. 11367, pp. 357-372, Asian Conference on Computer Vision, Perth, Australia, 02/12/2018. https://doi.org/10.1007/978-3-030-21074-8_30

Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera. / Jørgensen, Anders; Pedersen, Malte; Gade, Rikke; Fagertun, Jens; Moeslund, Thomas B.

Computer Vision – ACCV 2018 Workshops: 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. ed. / Gustavo Carneiro; Shaodi You. Springer, 2019. p. 357-372 (Lecture Notes in Computer Science, Vol. 11367).

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera

AU - Jørgensen, Anders

AU - Pedersen, Malte

AU - Gade, Rikke

AU - Fagertun, Jens

AU - Moeslund, Thomas B.

PY - 2019/6

Y1 - 2019/6

N2 - The manual postmortem inspection of broilers and their viscera is becoming a bottleneck as the slaughter rate increases. Computer vision can assist veterinarians during the inspection, but specular highlights can hide crucial details when inspecting for diseases on the viscera set. This study aims to restore details behind these specular highlights by capturing two images of the same viscera using shifting light positions. The dataset consists of images captured in-line at a poultry processing plant. The method achieves an average SSIM score of 0.96 over a test set of 100 image sets. The result is visually pleasing images with correct textural information instead of specular highlights.

AB - The manual postmortem inspection of broilers and their viscera is becoming a bottleneck as the slaughter rate increases. Computer vision can assist veterinarians during the inspection, but specular highlights can hide crucial details when inspecting for diseases on the viscera set. This study aims to restore details behind these specular highlights by capturing two images of the same viscera using shifting light positions. The dataset consists of images captured in-line at a poultry processing plant. The method achieves an average SSIM score of 0.96 over a test set of 100 image sets. The result is visually pleasing images with correct textural information instead of specular highlights.

UR - http://www.scopus.com/inward/record.url?scp=85068454241&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-21074-8_30

DO - 10.1007/978-3-030-21074-8_30

M3 - Article in proceeding

SN - 978-3-030-21073-1

T3 - Lecture Notes in Computer Science

SP - 357

EP - 372

BT - Computer Vision – ACCV 2018 Workshops

A2 - Carneiro, Gustavo

A2 - You, Shaodi

PB - Springer

ER -

Jørgensen A, Pedersen M, Gade R, Fagertun J, Moeslund TB. Reaching behind Specular Highlights by Registration of Two Images of Broiler Viscera. In Carneiro G, You S, editors, Computer Vision – ACCV 2018 Workshops: 14th Asian Conference on Computer Vision, 2018, Revised Selected Papers. Springer. 2019. p. 357-372. (Lecture Notes in Computer Science, Vol. 11367). https://doi.org/10.1007/978-3-030-21074-8_30