LOKA WICARA: EVALUASI TEKNIS DETEKSI OBJEK LOKAL BERBASIS YOLOV8 NANO UNTUK PROTOTIPE AAC

husni hidayat; Arif Iman Anshori; Muhammad  Ivan Fauzi; Talita Zada Aqila; Andre Maulana Syafi'i

doi:10.47775/ictech.v21i2.439

Penulis

husni hidayat Institut Teknologi dan Sains Nahdlatul Ulama Pekalongan
Arif Iman Anshori Institut Widya Pratama
Muhammad Ivan Fauzi Institut Widya Pratama
Talita Zada Aqila Institut Widya Pratama
Andre Maulana Syafi'i Institut Widya Pratama

DOI:

https://doi.org/10.47775/ictech.v21i2.439

Kata Kunci:

: AAC, ASD, objek lokal, teknologi asistif, YOLOv8 Nano

Abstrak

Anak dengan Autism Spectrum Disorder (ASD) dapat mengalami hambatan komunikasi fungsional, sedangkan aplikasi Augmentative and Alternative Communication (AAC) perlu menggunakan objek dan pesan yang dekat dengan konteks keseharian pengguna. Penelitian ini menyajikan evaluasi teknis awal Loka Wicara, yaitu prototipe AAC yang menghubungkan deteksi objek lokal dengan rancangan pesan kebutuhan berbahasa Indonesia. Dataset aktual terdiri atas 500 citra dan 520 kotak pembatas untuk lima kelas, yaitu bantal, gayung, gelas cangkir, nasi putih, dan sandal. Data dibagi menjadi 350 citra pelatihan, 100 citra validasi, dan 50 citra pengujian. YOLOv8n dan YOLOv5n-compatible dilatih selama 100 epoch menggunakan tiga random seed yang sama. Pada test set, YOLOv8n memperoleh precision 0,8475 ± 0,0652, recall 0,8267 ± 0,0837, F1-score 0,8328 ± 0,0241, mAP50 0,8768 ± 0,0338, dan mAP50–95 0,6115 ± 0,0358. YOLOv5n menghasilkan mAP50 dan mAP50–95 sedikit lebih tinggi, yaitu 0,8800 ± 0,0240 dan 0,6146 ± 0,0469, serta memiliki parameter dan ukuran model lebih kecil. Benchmark pada GPU Tesla T4 menunjukkan latensi rata-rata 18,21 ms untuk YOLOv8n dan 18,69 ms untuk YOLOv5n. Model YOLOv8n juga berhasil diekspor ke format LiteRT/TFLite. Hasil menunjukkan bahwa kedua model kompetitif; YOLOv8n dipertahankan sebagai kandidat prototipe karena recall dan F1-score sedikit lebih tinggi serta alur ekspornya berhasil, bukan karena unggul pada seluruh metrik. Pengujian Android, validasi pakar, dan UAT diperlukan pada tahap berikutnya

Referensi

Al’azmi, N. U., A. Wijiastuti, and W. Wagino. 2025. “Pengembangan Aplikasi Augmentative and Alternative Communication (AAC) Untuk Meningkatkan Komunikasi Ekspresif Anak Dengan Autisme Berbasis Etnopedagogi.” Jurnal ORTOPEDAGOGIA 11 (2): 397–401. https://doi.org/10.17977/um031v11i22025p397-401.

Bateman, K. J., S. E. Wilson, A. Gauvreau, et al. 2023. “Visual Supports to Increase Conversation Engagement for Preschoolers with Autism Spectrum Disorder during Mealtimes: An Initial Investigation.” Journal of Early Intervention 45 (2): 163–84. https://doi.org/10.1177/10538151221111762.

Chavers Edgar, T., R. W. Schlosser, and R. Koul. 2024. “Effects of an Augmentative and Alternative Communication Intervention Package on Socio-Communicative Behaviors between Minimally Speaking Autistic Children and Their Peers.” American Journal of Speech-Language Pathology 33 (4): 1619–38. https://doi.org/10.1044/2024_AJSLP-23-00313.

Dharma, A. S., C. N. S. Pardosi, and Z. P. Silaen. 2025. “Comparative Performance of YOLOv8 and SSD-MobileNet Algorithms for Road Damage Detection in Mobile Applications.” Sinkron: Jurnal Dan Penelitian Teknik Informatika 9 (3): 1159–69. https://doi.org/10.33395/sinkron.v9i3.15008.

Genc-Tosun, D., O. Kurt, Z. Cevher, and E. V. Gregori. 2023. “Teaching Children with Autism Spectrum Disorder to Answer Questions Using an iPad-Based Speech-Generating Device.” Journal of Autism and Developmental Disorders 53 (9): 3724–39. https://doi.org/10.1007/s10803-022-05683-z.

Griffen, B., C. Holyfield, E. R. Lorah, and N. Caldwell. 2024. “Increasing Linguistic and Prelinguistic Communication for Social Closeness during Naturalistic AAC Instruction with Young Children on the Autism Spectrum.” Augmentative and Alternative Communication 40 (3): 168–81. https://doi.org/10.1080/07434618.2023.2283846.

Holyfield, C., S. MacNeil, N. Caldwell, et al. 2024. “Leveraging Communication Partner Speech to Automate Augmented Input for Children on the Autism Spectrum Who Are Minimally Verbal: Prototype Development and Preliminary Efficacy Investigation.” American Journal of Speech-Language Pathology 33 (3): 1174–92. https://doi.org/10.1044/2023_AJSLP-23-00224.

Jocher, G., A. Chaurasia, and J. Qiu. 2023. Ultralytics YOLO (Version 8.0.0). GitHub, released. https://github.com/ultralytics/ultralytics.

Mittal, P. 2024. “A Comprehensive Survey of Deep Learning-Based Lightweight Object Detection Models for Edge Devices.” Artificial Intelligence Review 57: Article 242. https://doi.org/10.1007/s10462-024-10877-1.

More, S. S., and N. Patil. 2025. “Empowering the Visually Impaired: YOLOv8-Based Object Detection in Android Applications.” Procedia Computer Science 252: 457–69. https://doi.org/10.1016/j.procs.2025.01.005.

Muharib, R., V. L. Walker, and W. Dunn. 2024. “Effects of Interventions Involving Tablet-Based Speech-Generating Devices for Individuals with ASD: A Meta-Analysis.” Journal of Autism and Developmental Disorders 54 (12): 4496–514. https://doi.org/10.1007/s10803-023-06173-6.

Saleh, W. A., and N. M. Mutahara. 2023. “Augmentatif Dan Alternatif Komunikasi Dalam Pengembangan Kemampuan Komunikasi Anak Autis.” SPEED Journal: Journal of Special Education 6 (2): 177–82. https://doi.org/10.31537/speed.v6i2.936.

Sari, B. P., R. W. Sholikah, and R. V. H. Ginardi. 2024. “The Development of Mobile Application for Object Recognition Based on Deep Learning to Assist People with Visually Impaired.” 222–27. https://doi.org/10.1109/ISITDI62380.2024.10796984.

Schlosser, R. W., and R. Koul. 2023. “Advances in Augmentative and Alternative Communication Research for Individuals with Autism Spectrum Disorder: Moving Research and Practice Forward.” Augmentative and Alternative Communication 39 (1): 2–6. https://doi.org/10.1080/07434618.2023.2181214.

Sirisha, U., S. Phani Praveen, P. Naga Srinivasu, P. Barsocchi, and A. K. Bhoi. 2023. “Statistical Analysis of Design Aspects of Various YOLO-Based Deep Learning Models for Object Detection.” International Journal of Computational Intelligence Systems 16: Article 126. https://doi.org/10.1007/s44196-023-00302-w.

Terven, J., D. M. Cordova-Esparza, and J. A. Romero-Gonzalez. 2023. “A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS.” Machine Learning and Knowledge Extraction 5 (4): 1680–716. https://doi.org/10.3390/make5040083.

World Health Organization. 2025. “Autism.” September 17. https://www.who.int/news-room/fact-sheets/detail/autism-spectrum-disorders.

Yau, S. H., K. Choo, J. Tan, O. Monson, and S. Bovell. 2024. “Comparing and Contrasting Barriers in Augmentative Alternative Communication Use in Nonspeaking Autism and Complex Communication Needs: Multi-Stakeholder Perspectives.” Frontiers in Psychiatry 15: Article 1393017. https://doi.org/10.3389/fpsyt.2024.1393017.

Zervogianni, V., S. Fletcher-Watson, G. Herrera, et al. 2024. “A User-Based Information Rating Scale to Evaluate the Design of Technology-Based Supports for Autism.” Universal Access in the Information Society 23: 1783–94. https://doi.org/10.1007/s10209-023-00997-y.