Identification of the genus of stingless bee via faster R-CNN

A. Nizam; W. Mohd-Isa; A. Ali

doi:10.1117/12.2521380

22 March 2019 Identification of the genus of stingless bee via faster R-CNN

A. Nizam, W. Mohd-Isa, A. Ali

Proceedings Volume 11049, International Workshop on Advanced Image Technology (IWAIT) 2019; 1104943 (2019) https://doi.org/10.1117/12.2521380
Event: 2019 Joint International Workshop on Advanced Image Technology (IWAIT) and International Forum on Medical Imaging in Asia (IFMIA), 2019, Singapore, Singapore

Abstract

This study presents an interesting approach to identifying the Genus of the Stingless bee aided by machine learning technology. The conventional way of identifying the Genus of the Stingless bee or “Lebah Kelulut” relied on the face-to-face meetings with local bee experts. This particular process is considered to be outdated and time consuming. Thus, the proposed solution incorporated the machine learning tool called the “TensorFlow Object Detection API”. This tool is provided by Google TensorFlow and uses the Faster Region-based Convolutional Neural Network (Faster R-CNN), which incorporates the Region Proposal Network to enhance the current network. The data set used for training and testing consisted of 400 images, which belong to two types of bee species namely, the Heterotrigona Erythrogasta and Heterotrigona Itama. The evaluation of the model produced an accuracy rate of 73.75% for an average computing time per image of 0.65 seconds.

Citation Download Citation

A. Nizam, W. Mohd-Isa, and A. Ali "Identification of the genus of stingless bee via faster R-CNN", Proc. SPIE 11049, International Workshop on Advanced Image Technology (IWAIT) 2019, 1104943 (22 March 2019); https://doi.org/10.1117/12.2521380

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available