Deep Learning Based Egg Fertility Detection
This study investigates the implementation of deep learning (DL) approaches to the fertile egg-recognition problem, based on incubator images. In this study, we aimed to classify chicken eggs according to both segmentation and fertility status with a Mask R-CNN-based approach. In this manner, images...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Veterinary Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-7381/9/10/574 |
| _version_ | 1827647726125318144 |
|---|---|
| author | Kerim Kürşat Çevik Hasan Erdinç Koçer Mustafa Boğa |
| author_facet | Kerim Kürşat Çevik Hasan Erdinç Koçer Mustafa Boğa |
| author_sort | Kerim Kürşat Çevik |
| collection | DOAJ |
| description | This study investigates the implementation of deep learning (DL) approaches to the fertile egg-recognition problem, based on incubator images. In this study, we aimed to classify chicken eggs according to both segmentation and fertility status with a Mask R-CNN-based approach. In this manner, images can be handled by a single DL model to successfully perform detection, classification and segmentation of fertile and infertile eggs. Two different test processes were used in this study. In the first test application, a data set containing five fertile eggs was used. In the second, testing was carried out on the data set containing 18 fertile eggs. For evaluating this study, we used AP, one of the most important metrics for evaluating object detection and segmentation models in computer vision. When the results obtained were examined, the optimum threshold value (IoU) value was determined as 0.7. According to the IoU of 0.7, it was observed that all fertile eggs in the incubator were determined correctly on the third day of both test periods. Considering the methods used and the ease of the designed system, it can be said that a very successful system has been designed according to the studies in the literature. In order to increase the segmentation performance, it is necessary to carry out an experimental study to improve the camera and lighting setup prepared for taking the images. |
| first_indexed | 2024-03-09T19:23:34Z |
| format | Article |
| id | doaj.art-06a7fe5f25ab44358e05083b724ef9f0 |
| institution | Directory Open Access Journal |
| issn | 2306-7381 |
| language | English |
| last_indexed | 2024-03-09T19:23:34Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Veterinary Sciences |
| spelling | doaj.art-06a7fe5f25ab44358e05083b724ef9f02023-11-24T03:06:57ZengMDPI AGVeterinary Sciences2306-73812022-10-0191057410.3390/vetsci9100574Deep Learning Based Egg Fertility DetectionKerim Kürşat Çevik0Hasan Erdinç Koçer1Mustafa Boğa2Faculty of Applied Sciences, Akdeniz University, Antalya 07070, TurkeyFaculty of Technology, Selçuk University, Konya 42130, TurkeyBor Vocational School, Niğde Ömer Halisdemir University, Niğde 51700, TurkeyThis study investigates the implementation of deep learning (DL) approaches to the fertile egg-recognition problem, based on incubator images. In this study, we aimed to classify chicken eggs according to both segmentation and fertility status with a Mask R-CNN-based approach. In this manner, images can be handled by a single DL model to successfully perform detection, classification and segmentation of fertile and infertile eggs. Two different test processes were used in this study. In the first test application, a data set containing five fertile eggs was used. In the second, testing was carried out on the data set containing 18 fertile eggs. For evaluating this study, we used AP, one of the most important metrics for evaluating object detection and segmentation models in computer vision. When the results obtained were examined, the optimum threshold value (IoU) value was determined as 0.7. According to the IoU of 0.7, it was observed that all fertile eggs in the incubator were determined correctly on the third day of both test periods. Considering the methods used and the ease of the designed system, it can be said that a very successful system has been designed according to the studies in the literature. In order to increase the segmentation performance, it is necessary to carry out an experimental study to improve the camera and lighting setup prepared for taking the images.https://www.mdpi.com/2306-7381/9/10/574deep learningegg fertilityMask R-CNNincubator images |
| spellingShingle | Kerim Kürşat Çevik Hasan Erdinç Koçer Mustafa Boğa Deep Learning Based Egg Fertility Detection Veterinary Sciences deep learning egg fertility Mask R-CNN incubator images |
| title | Deep Learning Based Egg Fertility Detection |
| title_full | Deep Learning Based Egg Fertility Detection |
| title_fullStr | Deep Learning Based Egg Fertility Detection |
| title_full_unstemmed | Deep Learning Based Egg Fertility Detection |
| title_short | Deep Learning Based Egg Fertility Detection |
| title_sort | deep learning based egg fertility detection |
| topic | deep learning egg fertility Mask R-CNN incubator images |
| url | https://www.mdpi.com/2306-7381/9/10/574 |
| work_keys_str_mv | AT kerimkursatcevik deeplearningbasedeggfertilitydetection AT hasanerdinckocer deeplearningbasedeggfertilitydetection AT mustafaboga deeplearningbasedeggfertilitydetection |