(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Deep learning based butterfly species identification system through wings pattern

Deep learning based butterfly species identification system through wings pattern / Tan Yan Jia
Rama-rama selalu dianggap sebagai penunjuk-biologi bagi hutan atau ekologi oleh pakar entomologi. Daripada pemerhatian ke atas spesis rama-rama tertentu, keadaan ekologi akan dinilai dan kerosakan ekosistem dapat dikesan secara awal supaya pendekatan yang sesuai boleh diambil untuk mengurangkan impak dan meneutralisasikan punca kerosakan itu. Aplikasi pengenalan spesis secara automatik dapat membantu pakar entomologi untuk meninjau suatu kawasan dengan adanya Pembelajaran Mesin (ML). Namun, ini bukan sesuatu yang mudah kerana imej yang ditangkap adalah sangat berbeza dari segi posisi. Dilengkapi dengan perkakasan yang canggih, idea Rangkaian Konvolusi Neural (CNN) yang dicadangkan pada suatu ketika dahulu akhirnya boleh direalisasikan. Dalam pertandingan tahunan yang dianjurkan oleh ImageNet iaitu ImageNet Large Scale Visual Recognition Competition (ILSVRC), kebanyakan pemenang pada beberapa tahun ini mengintegrasikan idea CNN dalam model mereka. Dalam projek ini, salah satu model CNN yang terkenal, Inception V3, digunakan sebagai teras bagi sistem pengenalan spesis rama-rama. Optimasi dilakukan secara berperingkat dengan menjalankan eksperimen terhadap parameter untuk latihan rangkaian supaya nilai optimum boleh dicapai. Dengan penggunaan transfer learning, rekod terbaik ralat ialah sebanyak 14.3% bagi sepuluh spesis yang merangkumi dua pasangan spesis dengan ciri-ciri yang hampir sama. Masa pemprosesan bagi satu imej adalah kurang daripada 0.15 saat untuk penggunaan Unit Pemprosesan Grafik (GPU), tiga kali lebih cepat daripada penggunaan Unit Pemprosesan Pusat (CPU). _______________________________________________________________________________________________________ For entomologists, butterflies act as a bio-indicator for the health of a forest or an ecosystem. From the observation of butterflies, the condition of the forest can be determined and counter-measure can be taken earlier to neutralize or minimize the threat. The automated identification application, is helpful in assisting entomologist to survey an area with the implementation of Machine Learning (ML). It was once a difficult challenge due to uncertainties in the captured images such as obstructions and high variations of pose. With the hardware getting better and powerful, the concept of Convolutional Neural Network (CNN) that was once an idea can be realised. It is evident in the object recognition challenge organised by ImageNet, ImageNet Large Scale Visual Recognition Challenge (ILSVRC), where CNN produced state-of-the-art result. In this project, Inception V3 is used as the core engine to power the species recognition system that aims to help the entomologists to identify the butterfly species. Optimisation is performed by experimenting in stages with several training parameters to obtain the best value for this unique purpose. Using transfer learning, the best result recorded is 14.3% error with capability of identifying ten species, which includes two pairs of butterfly species with similar features. With Graphics Processing Unit (GPU), the processing time is less than 0.15 seconds per image which is three times faster than using a Central Processing Unit (CPU).
Contributor(s):
Tan Yan Jin - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875007234
Barcode : 00003107114
Language:
English
Subject Keywords:
butterflies; bio-indicator; ecosystem
First presented to the public:
6/1/2017
Original Publication Date:
4/16/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 73
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-04-16 16:33:36.248
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Deep learning based butterfly species identification system through wings pattern1 2018-04-16 16:33:36.248