Peningkatan Performa MobilenetV3 dengan Squeeze-and-excitation (Studi Kasus Klasifikasi Kesegaran Ikan Berdasarkan Mata Ikan)

GALIH ASHARI RAKHMAT; MUHAMMAD FIKRI HAEKAL

doi:10.26760/mindjournal.v8i1.27-41

Peningkatan Performa MobilenetV3 dengan Squeeze-and-excitation (Studi Kasus Klasifikasi Kesegaran Ikan Berdasarkan Mata Ikan)

GALIH ASHARI RAKHMAT, MUHAMMAD FIKRI HAEKAL

Sari

Abstrak

Ikan mengandung banyak nutrisi, protein tinggi yang memiliki banyak manfaat untuk tubuh manusia. Hal tersebut, membuat banyak masyarakat mengkonsumsi ikan sebagai makanan sehari â€“ hari. Sehingga diperlukan ketelitian masyarakat ketika membeli ikan menjadi perhatian serius dalam memilih ikan terkait kesegarannya. Penelitian dilakukanÂ dengan mengklasifikasikan 24 kelas kesegaran ikan berdasarkan mata ikan dengan menguji performa arsitektur MobileNetV3 dengan Squeeze-and-excitation. Hasil yang didapatkan model dengan kinerja terbaik diperoleh pada hyperparameter learning rate 0.00001, batch size train 10 val 10, optimizer ADAM, epochs 100 dengan model arsitektur MobileNetV3-Small. berdasarkan hasil evaluasi performa model didapatkan 68% accuracy, 69% precision, 67% recall dan 68% f1-score pada pengujian 876 data uji dengan 24 kelas yaitu terdiri dari tingkat kesegaran dan jenis ikan yang berbeda.

Kata kunci: MobileNetV3, Hyperparameter, Mata Ikan, Klasifikasi

Abstract

Fish contains many nutrients, high protein which has many benefits for the human body. This, makes many people consume fish as daily food. So that people need to be careful when buying fish to be a serious concern in choosing fish related to its freshness. The research was conducted by classifying 24 classes of fish freshness based on fish eye by testing the performance of the MobileNetV3 architecture with Squeeze-and-excitation. The results obtained for the model with the best performance were obtained at hyperparameter learning rate 0.00001, batch size train 10 val 10, ADAM optimizer, epochs 100 with the MobileNetV3-Small architectural model. Based on the results of the model performance evaluation, it obtained 68% accuracy, 69% precision, 67% recall and 68% f1-score on 876 test data with 24 classes consisting of different levels of freshness and types of fish.

Keywords: MobileNetV3, Hyperparameter, Fish Eye, Classification

Teks Lengkap:

PDF

Referensi

Agarwal, M., Gupta, S. K., & Biswas, K. K. (2020). Development of Efficient CNN model for Tomato crop disease identification. Sustainable Computing: Informatics and Systems, 28, 100407.

Chu, X., Zhang, B., & Xu, R. (2020). Moga: Searching beyond mobilenetv3. ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 4042â€“4046.

Diana, F. M. (2013). Omega 3 dan kecerdasan anak. Jurnal Kesehatan Masyarakat Andalas, 7(2), 82â€“88.

Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. ArXiv Preprint ArXiv:1704.04861.

Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., & Vasudevan, V. (2019). Searching for mobilenetv3. Proceedings of the IEEE/CVF International Conference on Computer Vision, 1314â€“1324.

Hu, J., Shen, L., & Sun, G. (2018). Squeeze-and-excitation networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 7132â€“7141.

Kandel, I., & Castelli, M. (2020). The effect of batch size on the generalizability of the convolutional neural networks on a histopathology dataset. ICT Express, 6(4), 312â€“315.

Prasetyo, E., Purbaningtyas, R., Adityo, R. D., Suciati, N., & Fatichah, C. (2022). Combining MobileNetV1 and Depthwise Separable convolution bottleneck with Expansion for classifying the freshness of fish eyes. Information Processing in Agriculture. https://doi.org/10.1016/j.inpa.2022.01.002

Qian, S., Ning, C., & Hu, Y. (2021). MobileNetV3 for Image Classification. 2021 IEEE 2nd International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering, ICBAIE 2021, Icbaie, 490â€“497. https://doi.org/10.1109/ICBAIE52039.2021.9389905

Radiuk, P. M. (2017). Impact of training set batch size on the performance of convolutional neural networks for diverse datasets. Information Technology and Management Science, 20(1), 20â€“24.

Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L. (2018). Mobilenetv2: Inverted residuals and linear bottlenecks. in proc. ieee conference on computer vision and pattern recognition.

Ying, X. (2019). An overview of overfitting and its solutions. Journal of Physics: Conference Series, 1168, 022022.

Zaki, S. Z. M., Zulkifley, M. A., Stofa, M. M., Kamari, N. A. M., & Mohamed, N. A. (2020). Classification of tomato leaf diseases using MobileNet v2. IAES International Journal of Artificial Intelligence, 9(2), 290.

DOI: https://doi.org/10.26760/mindjournal.v8i1.27-41