Automatic paddy planthopper detection and counting using faster R-CNN
Counting planthoppers manually is laborious and yields inconsistent results, particularly when dealing with species with similar features, such as the brown planthopper (Nilaparvata lugens; BPH), whitebacked planthopper (Sogatella furcifera; WBPH), zigzag leafhopper (Maiestas dorsalis; ZIGZAG), and...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Multidisciplinary Digital Publishing Institute
2024
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/114534/ http://psasir.upm.edu.my/id/eprint/114534/1/114534.pdf |
| _version_ | 1848866522119798784 |
|---|---|
| author | Khairunniza-Bejo, Siti Ibrahim, Mohd Firdaus Hanafi, Marsyita Jahari, Mahirah Ahmad Saad, Fathinul Syahir Mhd Bookeri, Mohammad Aufa |
| author_facet | Khairunniza-Bejo, Siti Ibrahim, Mohd Firdaus Hanafi, Marsyita Jahari, Mahirah Ahmad Saad, Fathinul Syahir Mhd Bookeri, Mohammad Aufa |
| author_sort | Khairunniza-Bejo, Siti |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | Counting planthoppers manually is laborious and yields inconsistent results, particularly when dealing with species with similar features, such as the brown planthopper (Nilaparvata lugens; BPH), whitebacked planthopper (Sogatella furcifera; WBPH), zigzag leafhopper (Maiestas dorsalis; ZIGZAG), and green leafhopper (Nephotettix malayanus and Nephotettix virescens; GLH). Most of the available automated counting methods are limited to populations of a small density and often do not consider those with a high density, which require more complex solutions due to overlapping objects. Therefore, this research presents a comprehensive assessment of an object detection algorithm specifically developed to precisely detect and quantify planthoppers. It utilises annotated datasets obtained from sticky light traps, comprising 1654 images across four distinct classes of planthoppers and one class of benign insects. The datasets were subjected to data augmentation and utilised to train four convolutional object detection models based on transfer learning. The results indicated that Faster R-CNN VGG 16 outperformed other models, achieving a mean average precision (mAP) score of 97.69% and exhibiting exceptional accuracy in classifying all planthopper categories. The correctness of the model was verified by entomologists, who confirmed a classification and counting accuracy rate of 98.84%. Nevertheless, the model fails to recognise certain samples because of the high density of the population and the significant overlap among them. This research effectively resolved the issue of low- to medium-density samples by achieving very precise and rapid detection and counting. |
| first_indexed | 2025-11-15T14:21:56Z |
| format | Article |
| id | upm-114534 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T14:21:56Z |
| publishDate | 2024 |
| publisher | Multidisciplinary Digital Publishing Institute |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-1145342025-01-17T07:07:16Z http://psasir.upm.edu.my/id/eprint/114534/ Automatic paddy planthopper detection and counting using faster R-CNN Khairunniza-Bejo, Siti Ibrahim, Mohd Firdaus Hanafi, Marsyita Jahari, Mahirah Ahmad Saad, Fathinul Syahir Mhd Bookeri, Mohammad Aufa Counting planthoppers manually is laborious and yields inconsistent results, particularly when dealing with species with similar features, such as the brown planthopper (Nilaparvata lugens; BPH), whitebacked planthopper (Sogatella furcifera; WBPH), zigzag leafhopper (Maiestas dorsalis; ZIGZAG), and green leafhopper (Nephotettix malayanus and Nephotettix virescens; GLH). Most of the available automated counting methods are limited to populations of a small density and often do not consider those with a high density, which require more complex solutions due to overlapping objects. Therefore, this research presents a comprehensive assessment of an object detection algorithm specifically developed to precisely detect and quantify planthoppers. It utilises annotated datasets obtained from sticky light traps, comprising 1654 images across four distinct classes of planthoppers and one class of benign insects. The datasets were subjected to data augmentation and utilised to train four convolutional object detection models based on transfer learning. The results indicated that Faster R-CNN VGG 16 outperformed other models, achieving a mean average precision (mAP) score of 97.69% and exhibiting exceptional accuracy in classifying all planthopper categories. The correctness of the model was verified by entomologists, who confirmed a classification and counting accuracy rate of 98.84%. Nevertheless, the model fails to recognise certain samples because of the high density of the population and the significant overlap among them. This research effectively resolved the issue of low- to medium-density samples by achieving very precise and rapid detection and counting. Multidisciplinary Digital Publishing Institute 2024-09-10 Article PeerReviewed text en cc_by_4 http://psasir.upm.edu.my/id/eprint/114534/1/114534.pdf Khairunniza-Bejo, Siti and Ibrahim, Mohd Firdaus and Hanafi, Marsyita and Jahari, Mahirah and Ahmad Saad, Fathinul Syahir and Mhd Bookeri, Mohammad Aufa (2024) Automatic paddy planthopper detection and counting using faster R-CNN. Agriculture, 14 (9). art. no. 1567. ISSN 2077-0472; eISSN: 2077-0472 https://www.mdpi.com/2077-0472/14/9/1567 10.3390/agriculture14091567 |
| spellingShingle | Khairunniza-Bejo, Siti Ibrahim, Mohd Firdaus Hanafi, Marsyita Jahari, Mahirah Ahmad Saad, Fathinul Syahir Mhd Bookeri, Mohammad Aufa Automatic paddy planthopper detection and counting using faster R-CNN |
| title | Automatic paddy planthopper detection and counting using faster R-CNN |
| title_full | Automatic paddy planthopper detection and counting using faster R-CNN |
| title_fullStr | Automatic paddy planthopper detection and counting using faster R-CNN |
| title_full_unstemmed | Automatic paddy planthopper detection and counting using faster R-CNN |
| title_short | Automatic paddy planthopper detection and counting using faster R-CNN |
| title_sort | automatic paddy planthopper detection and counting using faster r-cnn |
| url | http://psasir.upm.edu.my/id/eprint/114534/ http://psasir.upm.edu.my/id/eprint/114534/ http://psasir.upm.edu.my/id/eprint/114534/ http://psasir.upm.edu.my/id/eprint/114534/1/114534.pdf |