Oto-BaC„: an automated Artificial Intelligence (AI) detector and counter for bagworm (Lepidoptera: Psychidae) census

Ishak Aris, . and Abdul Rashid Mohamed Shariff, . and Izhal Abdul Halin, . and Ramle Moslim, . and Mohd Najib Ahmad, . Oto-BaC„: an automated Artificial Intelligence (AI) detector and counter for bagworm (Lepidoptera: Psychidae) census. pp. 1-16. ISSN 2735-1084

The bagworm species of Metisa plana is one of the major species of leaf-eating insect pest that attack oil palm in Peninsular Malaysia. Without any treatment this situation may cause 43 yield loss from a moderate attack. In 2020 the economic loss due to the bagworm attack was recorded at around RM 180 million. Based on this scenario it is necessary to closely monitor the bagworm outbreak at the infested areas. Accuracy and precise data collection is debatable due to human errors such as miscounting cheating and creating data. The objective of this technology is to design and develop a specific machine vision that incorporates image processing algorithm according to its functional modes. The device Automated Bagworm Counter or Oto-BaC„ us the first in the world to be developed. The software functions based on a graphic processing unit computation and used TensorFlow/Teano library set up for the trained dataset. The technology is based on the developed deep learning with Faster Regions with Convolutional Neural Networks technique towards real time object detection. The Oto-BaC„ uses an ordinary camera. By using self-developed Deep Learning algorithms a motion-tracking and false color analysis are applied to detect and count number of living and dead larvae and pupae population per frond respectively corresponding to three major groups or sizes classification. Initially in the first trial the Oto-BaC„ has resulted in low percentages of detection accuracy for the living and dead G1 larvae (47.0 71.7) G2 larvae (39.1 50.0) and G3 pupae (30.1 20.9). After some improvements on training dataset the percentages increased in the next field trial amount of 40.5 and 7.0 for the living and dead G1 larvae 40.1 and 29.2 for the living and dead G2 larvae and 47.7 and 54.6 for the living and dead pupae. Furthermore the development of the ground-based device is the pioneer in the oil palm industry in which it reduces human error when conducting census while promoting precision agriculture practice.

Full text available from: Official URL: http://journals.hh-publisher.com/index.php/AAFRJ/a...

Abstract

The bagworm species of Metisa plana is one of the major species of leaf-eating insect pest that attack oil palm in Peninsular Malaysia. Without any treatment this situation may cause 43 yield loss from a moderate attack. In 2020 the economic loss due to the bagworm attack was recorded at around RM 180 million. Based on this scenario it is necessary to closely monitor the bagworm outbreak at the infested areas. Accuracy and precise data collection is debatable due to human errors such as miscounting cheating and creating data. The objective of this technology is to design and develop a specific machine vision that incorporates image processing algorithm according to its functional modes. The device Automated Bagworm Counter or Oto-BaC„ us the first in the world to be developed. The software functions based on a graphic processing unit computation and used TensorFlow/Teano library set up for the trained dataset. The technology is based on the developed deep learning with Faster Regions with Convolutional Neural Networks technique towards real time object detection. The Oto-BaC„ uses an ordinary camera. By using self-developed Deep Learning algorithms a motion-tracking and false color analysis are applied to detect and count number of living and dead larvae and pupae population per frond respectively corresponding to three major groups or sizes classification. Initially in the first trial the Oto-BaC„ has resulted in low percentages of detection accuracy for the living and dead G1 larvae (47.0 71.7) G2 larvae (39.1 50.0) and G3 pupae (30.1 20.9). After some improvements on training dataset the percentages increased in the next field trial amount of 40.5 and 7.0 for the living and dead G1 larvae 40.1 and 29.2 for the living and dead G2 larvae and 47.7 and 54.6 for the living and dead pupae. Furthermore the development of the ground-based device is the pioneer in the oil palm industry in which it reduces human error when conducting census while promoting precision agriculture practice.

[error in script]

Item Type:	Article
AGROVOC Term:	Oil palm
AGROVOC Term:	Plantations
AGROVOC Term:	Pest management
AGROVOC Term:	Pest control
AGROVOC Term:	Pest surveys
AGROVOC Term:	Psychidae
AGROVOC Term:	Lepidoptera
AGROVOC Term:	Leaf eating insects
AGROVOC Term:	Application of technology
AGROVOC Term:	Ai (artificial intelligence)
Depositing User:	Mr. AFANDI ABDUL MALEK
Last Modified:	24 Apr 2025 00:55
URI:	http://webagris.upm.edu.my/id/eprint/10121

Actions (login required)

View Item