| LDR | | 00000nam u2200205 4500 |
| 001 | | 000000433514 |
| 005 | | 20200225142846 |
| 008 | | 200131s2019 ||||||||||||||||| ||eng d |
| 020 | |
▼a 9781085727778 |
| 035 | |
▼a (MiAaPQ)AAI22615310 |
| 040 | |
▼a MiAaPQ
▼c MiAaPQ
▼d 247004 |
| 082 | 0 |
▼a 004 |
| 100 | 1 |
▼a Araujo Fernandes, Arthur Francisco. |
| 245 | 10 |
▼a Computer Vision Applied to Automated Measurement of Biometric Traits and Prediction of Body Weight and Carcass Traits in Live Animals. |
| 260 | |
▼a [S.l.]:
▼b The University of Wisconsin - Madison.,
▼c 2019. |
| 260 | 1 |
▼a Ann Arbor:
▼b ProQuest Dissertations & Theses,
▼c 2019. |
| 300 | |
▼a 118 p. |
| 500 | |
▼a Source: Dissertations Abstracts International, Volume: 81-03, Section: B. |
| 500 | |
▼a Advisor: Rosa, Guilherme J. M. |
| 502 | 1 |
▼a Thesis (Ph.D.)--The University of Wisconsin - Madison, 2019. |
| 506 | |
▼a This item must not be sold to any third party vendors. |
| 506 | |
▼a This item must not be added to any third party search indexes. |
| 520 | |
▼a This dissertation deals with applications of computer vision in animal sciences. The first chapter provides an introduction to computer vision and literature review of applications in animal sciences. The second chapter focuses on digital image analysis for Nile tilapia in which images are used for extraction of body measurements and prediction of body and carcass weight, as well as carcass yield. The best segmentation network achieved results for intersection over union of 99, 90 and 64% for background, fish body and fin areas, respectively. Also, a predictive model including body area as covariate achieved R2 of 0.95 and 0.94 for body and carcass weights, respectively. However, none of the evaluated models gave satisfactory predictions of carcass yield. The third chapter aims at the use of 3D images for measurement of pig biometric traits and body weight. In this chapter, classical image processing strategies were used for the development of a system for classification of a pig image, extraction of variables, and prediction of body weight. A linear model using the truncated median estimate of selected variables achieved an R2 of 0.94 for prediction of body weight. This result was similar to previous reports in which a human evaluator manually selected images from animals to be used. In conclusion, it was possible to achieve satisfactory results for prediction of pig body weight automatically using 3D sensors. In the fourth chapter, several modeling approaches were evaluated for prediction of body weight, lean meat, and back fat in pigs. Again, 3D images were processed for extraction of image variables that were used as input for either linear regression and machine learning models. In addition, a deep learning approach using the raw image as input was evaluated. Over the evaluated models, the deep learning approach presented the best predictive performance for body weight and muscle depth, while for back fat, there was no clear advantage compared to linear regression. In conclusion, it was possible to predict body weight, muscle depth, and back fat using 3D images of the pigs. Nonetheless, additional research should be performed, towards the improvement of predictions of back fat. |
| 590 | |
▼a School code: 0262. |
| 650 | 4 |
▼a Animal sciences. |
| 650 | 4 |
▼a Computer science. |
| 690 | |
▼a 0475 |
| 690 | |
▼a 0984 |
| 710 | 20 |
▼a The University of Wisconsin - Madison.
▼b Animal Sciences. |
| 773 | 0 |
▼t Dissertations Abstracts International
▼g 81-03B. |
| 773 | |
▼t Dissertation Abstract International |
| 790 | |
▼a 0262 |
| 791 | |
▼a Ph.D. |
| 792 | |
▼a 2019 |
| 793 | |
▼a English |
| 856 | 40 |
▼u http://www.riss.kr/pdu/ddodLink.do?id=T15493289
▼n KERIS
▼z 이 자료의 원문은 한국교육학술정보원에서 제공합니다. |
| 980 | |
▼a 202002
▼f 2020 |
| 990 | |
▼a ***1008102 |
| 991 | |
▼a E-BOOK |