Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Jurnal Sistem Komput...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Jurnal Sistem Komputer dan Informatika (JSON)
Article . 2025 . Peer-reviewed
License: CC BY SA
Data sources: Crossref
addClaim

Peningkatan Akurasi Klasifikasi Ikan kepe-kepe (Famili Chaetodontidae) dengan EfficientNetV2 dan Bayesian Hyperparameter Tuning

Authors: I Gusti Agung Putu Mahendra; null Muhammad Ikhsan Wibowo; null Zuliar Efendi;

Peningkatan Akurasi Klasifikasi Ikan kepe-kepe (Famili Chaetodontidae) dengan EfficientNetV2 dan Bayesian Hyperparameter Tuning

Abstract

Identifikasi cepat dan akurat spesies Chaetodontidae penting untuk monitoring keanekaragaman hayati laut, namun pendekatan manual tidak skala dan rentan kesalahan pada dataset besar. GAP riset yang kami tangani adalah: (i) ketiadaan kajian yang secara khusus mengombinasikan EfficientNetV2 dengan Bayesian hyperparameter tuning untuk klasifikasi Chaetodontidae, dan (ii) belum adanya evaluasi yang menekankan efisiensi penalaan adaptif beserta dampaknya terhadap performa. Kebaruan (novelty) studi ini ialah perancangan pipeline ringkas-efisien berbasis EfficientNetV2 dengan Bayesian Optimization (10 percobaan) pada learning rate, dropout, dan unfreeze backbone, dipadukan augmentasi kuat (MixUp, CutMix) serta regularisasi (label smoothing, L2). Dataset mencakup 1.427 citra/13 spesies dengan praproses center-crop 80% dan resize 224×224. Konfigurasi terbaik (unfreeze=True, dropout=0,2, LR 3,73×10⁻⁴) mencapai val-accuracy 92,75% dan akurasi uji 97%, dengan precision–recall rata-rata >95%, menunjukkan generalisasi yang baik bahkan pada kelas bermorfologi mirip. Dibanding penalaan manual/grid, pendekatan ini lebih hemat eksperimen sekaligus meningkatkan akurasi. Temuan tersebut menegaskan bahwa integrasi EfficientNetV2 + Bayesian tuning efektif dan siap diadopsi untuk sistem identifikasi–monitoring ikan berbasis citra pada konteks konservasi laut Indonesia.

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
gold