SavedAdd your journal

Journals / Natural and Applied Sciences Journal / 2026 / Cilt: 9 Sayı: 1

Comparative Evaluation of YOLO Architectures with Optimized Preprocessing for Multi-Class Microalgae Detection

Çok Sınıflı Mikroalg Tespiti için Optimize Edilmiş Ön İşleme ile YOLO Mimarilerinin Karşılaştırmalı Değerlendirmesi

Görkem Duman(İZMİR DEMOKRASİ ÜNİVERSİTESİ)0009-0002-7661-3762
Başak Esin Köktürk Güzel(İZMİR DEMOKRASİ ÜNİVERSİTESİ)0000-0002-9429-1149
Pages
27–38
Publication Date
2026-06-30
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

YOLOv9s achieved the best overall performance with the highest precision (0.867), F1-score (0.824), and mAP50 (0.889).

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Accurate detection and counting of microalgae species are essential for biomass monitoring, contamination control, and smart cultivation management in biotechnology applications.
📊
Key Findings & Data:In the first stage, five preprocessing pipelines were comparatively evaluated using YOLOv8s to determine an effective image enhancement strategy; LAB color space conversion combined with CLAHE and sharpening visibly improved cell boundary definition.
💡
Conclusion & Impact:YOLOv9s achieved the best overall performance with the highest precision (0.867), F1-score (0.824), and mAP50 (0.889).
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Mikroalg tespiti · Derin öğrenme · YOLO

Abstract

Accurate detection and counting of microalgae species are essential for biomass monitoring, contamination control, and smart cultivation management in biotechnology applications. However, automated analysis of microscopy images remains challenging because of dense cell distributions, small object sizes, low-contrast boundaries, and inter-class visual similarity. This study presents a two-stage deep learning framework for multi-class microalgae detection using recent YOLO architectures. In the first stage, five preprocessing pipelines were comparatively evaluated using YOLOv8s to determine an effective image enhancement strategy; LAB color space conversion combined with CLAHE and sharpening visibly improved cell boundary definition. In the second stage, the selected preprocessing configuration was fixed and four lightweight detectors (YOLOv8s, YOLOv9s, YOLOv10s, and YOLOv11s) were benchmarked under identical training conditions. YOLOv9s achieved the best overall performance with the highest precision (0.867), F1-score (0.824), and mAP50 (0.889). The findings indicate that detector architecture and preprocessing strategy jointly influence microscopic algae detection performance, and that model recency alone does not guarantee superior results.

Özet

Biyoteknoloji uygulamalarında biyokütle izleme, kontaminasyon kontrolü ve akıllı kültür yönetimi açısından mikroalg türlerinin doğru tespiti ve sayımı büyük önem taşımaktadır. Ancak yoğun hücre dağılımları, küçük nesne boyutları, düşük kontrastlı sınırlar ve sınıflar arası görsel benzerlik nedeniyle mikroskopi görüntülerinin otomatik analizi zorlu olmaya devam etmektedir. Bu çalışmada, güncel YOLO mimarileri kullanılarak çok sınıflı mikroalg tespiti için iki aşamalı bir derin öğrenme çerçevesi sunulmaktadır. Birinci aşamada, etkili bir görüntü iyileştirme stratejisi belirlemek amacıyla beş farklı ön işleme hattı YOLOv8s kullanılarak karşılaştırmalı olarak değerlendirilmiş; LAB renk uzayı dönüşümü ile CLAHE ve keskinleştirme uygulaması hücre sınırlarını belirgin biçimde iyileştirmiştir. İkinci aşamada seçilen ön işleme yapılandırması sabit tutulmuş ve dört hafif dedektör (YOLOv8s, YOLOv9s, YOLOv10s ve YOLOv11s) aynı eğitim koşulları altında kıyaslanmıştır. YOLOv9s, en yüksek kesinlik (0.867), F1 skoru (0.824) ve mAP50 (0.889) değerleriyle genel en iyi performansı elde etmiştir. Bulgular, dedektör mimarisi ile ön işleme stratejisinin birlikte belirleyici olduğunu ve daha yeni mimari nesillerin her zaman üstün sonuç vermediğini ortaya koymaktadır.

🔖

Cite This Article

Duman, G., Güzel, B. E. K. (2026). Comparative Evaluation of YOLO Architectures with Optimized Preprocessing for Multi-Class Microalgae Detection. Natural and Applied Sciences Journal, Cilt: 9 Sayı: 1, 27–38. https://doi.org/10.38061/idunas.1941287

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (18)

  1. Borowitzka, M. A. (2013). High-value products from microalgae—their development and commercialisation. Journal of applied phycology, 25(3), 743-756.
  2. Khan, M. I., Shin, J. H., & Kim, J. D. (2018). The promising future of microalgae: current status, challenges, and optimization of a sustainable and renewable industry for biofuels, feed, and other products. Microbial cell factories, 17(1), 36.
  3. Hlavova, M., Turoczy, Z., & Bisova, K. (2015). Improving microalgae for biotechnology—From genetics to synthetic biology. Biotechnology advances, 33(6), 1194-1203.
  4. Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 779-788).
  5. Gonzalez, R.C. and Woods, R.E. (2018) Digital Image Processing. 4th Edition, Pearson Education, New York, 1022 p.
  6. Yan, H., Peng, X., Chen, C., Xia, A., Huang, Y., Zhu, X.,. & Liao, Q. (2023). YOLOx model-based object detection for microalgal bioprocess. Algal Research, 74, 103178.
  7. Liu, D., Wang, P., Cheng, Y., & Bi, H. (2022). An improved algae-YOLO model based on deep learning for object detection of ocean microalgae considering aquacultural lightweight deployment. Frontiers in Marine Science, 9, 1070638.
  8. Duan, Z., Xie, T., Wang, L., Chen, Y., & Wu, J. (2024). Microalgae detection based on improved YOLOv5. IET image processing, 18(10), 2602-2613.
  9. Park, J., Baek, J., You, K., Nam, S.W., & Kim, J. (2021). Microalgae detection using a deep learning object detection algorithm, YOLOv3. J. Korean Soc. Water Environ., 37, 275–285.
  10. Abdullah, Ali, S., Khan, Z., Hussain, A., Athar, A., & Kim, H. C. (2022). Computer vision based deep learning approach for the detection and classification of algae species using microscopic images. Water, 14(14), 2219.
  11. Ding, G., Shi, Y., Liu, Z., Wang, Y., Yao, Z., Zhou, D.,. & Li, Y. (2025). Fe-YOLO: An efficient deep learning model based on feature-enhanced YOLOv7 for microalgae identification and detection. Biomimetics, 10(1), 62.
  12. Yoshimi, Y., Mine, Y., Ito, S., Takeda, S., Okazaki, S., Nakamoto, T.,. & Tanimoto, K. (2024). Image preprocessing with contrast-limited adaptive histogram equalization improves the segmentation performance of deep learning for the articular disk of the temporomandibular joint on magnetic resonance images. Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology, 138(1), 128-141.
  13. Wang, T. S., Kim, G. T., Kim, M., & Jang, J. (2023). Contrast enhancement-based preprocessing process to improve deep learning object task performance and results. Applied Sciences, 13(19), 10760.
  14. Sun, M., Jiang, J., Yang, Z., Kong, S., Qi, J., Shang, J., Luo, S., Sun, W., Wang, T., Wang, Y., Wang, Q., Dai, T., Chen, T., Zhang, J., Zhang, X., He, Y., Fu, P., Guan, Q., Zhou, S., … Yan, H. (2025). VisAlgae 2023: A Dataset and Challenge for Algae Detection in Microscopy Images. arXiv preprint arXiv:2505.20687.
  15. Wang, C. Y., Yeh, I. H., & Mark Liao, H. Y. (2025). Yolov9: Learning what you want to learn using programmable gradient information. In European conference on computer vision (pp. 1-21). Cham: Springer Nature Switzerland.
  16. Yan, H., Peng, X., Chen, C., Xia, A., Huang, Y., Zhu, X.,. & Liao, Q. (2023). YOLOx model-based object detection for microalgal bioprocess. Algal Research, 74, 103178.
  17. Ding, G., Shi, Y., Liu, Z., Wang, Y., Yao, Z., Zhou, D.,. & Li, Y. (2025). Fe-YOLO: An efficient deep learning model based on feature-enhanced YOLOv7 for microalgae identification and detection. Biomimetics, 10(1), 62.
  18. Yoshimi, Y., Mine, Y., Ito, S., Takeda, S., Okazaki, S., Nakamoto, T.,. & Tanimoto, K. (2024). Image preprocessing with contrast-limited adaptive histogram equalization improves the segmentation performance of deep learning for the articular disk of the temporomandibular joint on magnetic resonance images. Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology, 138(1), 128-141.