EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection
Automated pig monitoring is important for smart pig farms; thus, several deep-learning-based pig monitoring techniques have been proposed recently. In applying automated pig monitoring techniques to real pig farms, however, practical issues such as detecting pigs from overexposed regions, caused by...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/11/12/5577 |
| _version_ | 1797529905132470272 |
|---|---|
| author | Hanse Ahn Seungwook Son Heegon Kim Sungju Lee Yongwha Chung Daihee Park |
| author_facet | Hanse Ahn Seungwook Son Heegon Kim Sungju Lee Yongwha Chung Daihee Park |
| author_sort | Hanse Ahn |
| collection | DOAJ |
| description | Automated pig monitoring is important for smart pig farms; thus, several deep-learning-based pig monitoring techniques have been proposed recently. In applying automated pig monitoring techniques to real pig farms, however, practical issues such as detecting pigs from overexposed regions, caused by strong sunlight through a window, should be considered. Another practical issue in applying deep-learning-based techniques to a specific pig monitoring application is the annotation cost for pig data. In this study, we propose a method for managing these two practical issues. Using annotated data obtained from training images without overexposed regions, we first generated augmented data to reduce the effect of overexposure. Then, we trained YOLOv4 with both the annotated and augmented data and combined the test results from two YOLOv4 models in a bounding box level to further improve the detection accuracy. We propose accuracy metrics for pig detection in a closed pig pen to evaluate the accuracy of the detection without box-level annotation. Our experimental results with 216,000 “unseen” test data from overexposed regions in the same pig pen show that the proposed ensemble method can significantly improve the detection accuracy of the baseline YOLOv4, from 79.93% to 94.33%, with additional execution time. |
| first_indexed | 2024-03-10T10:21:31Z |
| format | Article |
| id | doaj.art-75bd93f5e4d84c86937f4d5b611c516c |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T10:21:31Z |
| publishDate | 2021-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-75bd93f5e4d84c86937f4d5b611c516c2023-11-22T00:23:16ZengMDPI AGApplied Sciences2076-34172021-06-011112557710.3390/app11125577EnsemblePigDet: Ensemble Deep Learning for Accurate Pig DetectionHanse Ahn0Seungwook Son1Heegon Kim2Sungju Lee3Yongwha Chung4Daihee Park5Department of Computer Convergence Software, Korea University, Sejong 30019, KoreaDepartment of Computer Convergence Software, Korea University, Sejong 30019, KoreaDepartment of Computer Convergence Software, Korea University, Sejong 30019, KoreaDepartment of Software, Sangmyung University, Cheonan 31066, KoreaDepartment of Computer Convergence Software, Korea University, Sejong 30019, KoreaDepartment of Computer Convergence Software, Korea University, Sejong 30019, KoreaAutomated pig monitoring is important for smart pig farms; thus, several deep-learning-based pig monitoring techniques have been proposed recently. In applying automated pig monitoring techniques to real pig farms, however, practical issues such as detecting pigs from overexposed regions, caused by strong sunlight through a window, should be considered. Another practical issue in applying deep-learning-based techniques to a specific pig monitoring application is the annotation cost for pig data. In this study, we propose a method for managing these two practical issues. Using annotated data obtained from training images without overexposed regions, we first generated augmented data to reduce the effect of overexposure. Then, we trained YOLOv4 with both the annotated and augmented data and combined the test results from two YOLOv4 models in a bounding box level to further improve the detection accuracy. We propose accuracy metrics for pig detection in a closed pig pen to evaluate the accuracy of the detection without box-level annotation. Our experimental results with 216,000 “unseen” test data from overexposed regions in the same pig pen show that the proposed ensemble method can significantly improve the detection accuracy of the baseline YOLOv4, from 79.93% to 94.33%, with additional execution time.https://www.mdpi.com/2076-3417/11/12/5577agriculture ITcomputer visionpig detectiondeep learningdata augmentationmodel ensemble |
| spellingShingle | Hanse Ahn Seungwook Son Heegon Kim Sungju Lee Yongwha Chung Daihee Park EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection Applied Sciences agriculture IT computer vision pig detection deep learning data augmentation model ensemble |
| title | EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection |
| title_full | EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection |
| title_fullStr | EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection |
| title_full_unstemmed | EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection |
| title_short | EnsemblePigDet: Ensemble Deep Learning for Accurate Pig Detection |
| title_sort | ensemblepigdet ensemble deep learning for accurate pig detection |
| topic | agriculture IT computer vision pig detection deep learning data augmentation model ensemble |
| url | https://www.mdpi.com/2076-3417/11/12/5577 |
| work_keys_str_mv | AT hanseahn ensemblepigdetensembledeeplearningforaccuratepigdetection AT seungwookson ensemblepigdetensembledeeplearningforaccuratepigdetection AT heegonkim ensemblepigdetensembledeeplearningforaccuratepigdetection AT sungjulee ensemblepigdetensembledeeplearningforaccuratepigdetection AT yongwhachung ensemblepigdetensembledeeplearningforaccuratepigdetection AT daiheepark ensemblepigdetensembledeeplearningforaccuratepigdetection |