Deep learning method for collembola identification using single species and community combinations images

Abstract

Deep learning methods are increasingly vital for species classification across animal species. However, the arthropod classification and detection of deep learning remain underexplored, especially soil arthropods. In this study, we collected 5020 images of collembolan individual as the origin dataset, spanning 6 families, 10 genera, and 51 species in China. By employing the cut-paste method, we created community image datasets of collembolan with diversity gradient to train and evaluate YOLOv8 and Faster R-CNN models for collembolan identification using deep learning methods. Our classification model achieved 83.05% precision and detection models exceeded 97%. YOLOv8 models were the most effective, with a higher top-1 precision all over 80% at the family level, compared to the Faster R-CNN model, which had a minimum precision of 60%. Using community datasets of genera belonging to Isotomidae or Entomobryidae family to train YOLOv8 models, we observed precision ranging from 18.51% to 99.17%, reflecting the changes of detection performance based on Collembola diversity. The YOLOv8 models excelled in identifying genera within Isotomidae family compared to Entomobryidae family. Our study pioneers the application of the YOLOv8 deep learning model for rapid detection and classification of Collembola, while proposing an adaptive training strategy specifically designed for diversity gradient samples. Our research demonstrates the potential of deep learning for rapid, accurate Collembola identification and extends to broader ecological assessments.

Graphical abstract

Keywords

Collembola diversity / Collembola identification / deep learning / image identification / object detection

Highlight

	● YOLOv8 was better than Faster R-CNN in Collembola detection, attaining 97% precision.
	● Precision varied widely (18.51%–99.17%) with Collembola diversity at genus-level.
	● Our study shows deep learning’s potential for rapid and precise ecology evaluation.

References

[1] Anslan, S., Tedersoo, L., 2015. Performance of cytochrome c oxidase subunit I (COI), ribosomal DNA Large Subunit (LSU) and Internal Transcribed Spacer 2 (ITS2) in DNA barcoding of Collembola. European Journal of Soil Biology69, 1–7.

[2] Butt, M., Awang Iskandar, D.N.F., Khan, M.A., Latif, G., Bashar, A., 2025. MEDCnet: a memory efficient approach for processing high-resolution fundus images for diabetic retinopathy classification using CNN. International Journal of Imaging Systems and Technology35, e70063.

[3] Carriero, A., Luijken, K., De Hond, A., Moons, K.G.M., van Calster, B., van Smeden, M., 2025. The harms of class imbalance corrections for machine learning based prediction models: a simulation study. Statistics in Medicine44, e10320.

[4] Chen, Q., Wang, P.S., Cheng, A.D., Wang, W.G., Zhang, Y.F., Cheng, J., 2020. Robust one-stage object detection with location-aware classifiers. Pattern Recognition105, 107334.

[5] Dwibedi, D., Misra, I., Hebert, M., 2017. Cut, paste and learn: surprisingly easy synthesis for instance detection. In: Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV). Venice: IEEE, 1310–1319.

[6] Edwards, C.A., 1991. The assessment of populations of soil-inhabiting invertebrates. Agriculture, Ecosystems & Environment34, 145–176.

[7] Eisenhauer, N., Antunes, P.M., Bennett, A.E., Birkhofer, K., Bissett, A., Bowker, M.A., Caruso, T., Chen, B.D., Coleman, D.C., Boer, W.D., Ruiter, P.D., Deluca, T.H., Frati, F., Griffiths, B.S., Hart, M.M., Hättenschwiler, S., Haimi, J., Heethoff, M., Kaneko, N., Kelly, L.C., Leinaas, H.P., Lindo, Z., Macdonald, C., Rillig, M.C., Ruess, L., Scheu, S., Schmidt, O., Seastedt, T.R., Straalen, N. M.V., Tiunov, A.V., Zimmer, M., Powell, J.R., 2017. Priorities for research in soil ecology. Pedobiologia63, 1–7.

[8] Fan, J.Y., Lee, J., Jung, I., Lee, Y., 2021. Improvement of object detection based on Faster R-CNN and YOLO. In: Proceedings of the 2021 36th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC). Jeju: IEEE, 1–4.

[9] Feng, J., Qiao, Z.H., Yan, Q.B., Yao, H.F., Wang, B., Sun, X., 2024. Effects of urbanization and greenspace types on community structure and functional traits of soil Collembola. Acta Ecologica Sinica44, 2582–2596.

[10] Ferreira, A.C., Silva, L.R., Renna, F., Brandl, H.B., Renoult, J.P., Farine, D.R., Covas, R., Doutrelant, C., 2020. Deep learning-based methods for individual recognition in small birds. Methods in Ecology and Evolution11, 1072–1085.

[11] Ghiasi, G., Cui, Y., Srinivas, A., Qian, R., Lin, T.Y., Cubuk, E.D., Le, Q.V., Zoph, B., 2021. Simple copy-paste is a strong data augmentation method for instance segmentation. In: Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2917–2927.

[12] Girshick, R., Donahue, J., Darrell, T., Malik, J., 2014. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 580–587.

[13] Islam, S.B., Valles, D., Hibbitts, T.J., Ryberg, W.A., Walkup, D.K., Forstner, M.R.J., 2023. Animal species recognition with deep convolutional neural networks from ecological camera trap images. Animals13, 1526.

[14] Joulin, A., van der Maaten, L., Jabri, A., Vasilache, N., 2016. Learning visual features from large weakly supervised data. In: Proceedings of the 14th European Conference on Computer Vision (ECCV). Amsterdam: Springer, 67–84.

[15] Kleemann, J., Steinhoff-Knopp, B., Eisenhauer, N., Ristok, C., Xylander, W.E.R., Burkhard, B., 2025. The unexplored links between soil, soil biodiversity, and soil-related ecosystem services. Soil Organisms97, 5–25.

[16] Krizhevsky, A., Sutskever, I., Hinton, G.E., 2017. ImageNet classification with deep convolutional neural networks. Communications of the ACM,60, 84–90.

[17] Lawal, O.M., Tan, Y., Liu, C.L., 2025. MADNet: marine animal detection network using the YOLO platform. PLoS One20, e0322799.

[18] LeCun, Y., Bengio, Y., Hinton, G., 2015. Deep learning. Nature521, 436–444.

[19] Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L., 2014. Microsoft COCO: common objects in context. In: Proceedings of the 13th European Conference on Computer Vision (ECCV). Zurich: Springer, 740–755.

[20] Liu, S.Y., Li, Z., Ke, X., Sun, L.N., Wu, L.H., Zhao, J.J., 2022. Community characteristics of soil collembola around a typical mercury-thallium mining area in Guizhou Province. Biodiversity Science30, 22265.

[21] Lu, J.Z., Yang, J.B., Bluhm, C., Foltran, E., Rivera Pérez, C.A., Glatthorn, J., Ammer, C., Lamersdorf, N., Polle, A., Berg, M., Potapov, A.M., Scheu, S., 2025. Mixed forests with native species mitigate impacts of introduced Douglas fir on soil decomposers (Collembola). Ecological Applications35, e70034.

[22] Maaß, S., Daphi, D., Lehmann, A., Rillig, M.C., 2017. Transport of microplastics by two collembolan species. Environmental Pollution225, 456–459.

[23] Müller, J., Mitesser, O., Schaefer, H.M., Seibold, S., Busse, A., Kriegel, P., Rabl, D., Gelis, R., Arteaga, A., Freile, J., Leite, G.A., De Melo, T.N., Lebien, J., Campos-Cerqueira, M., Blüthgen, N., Tremlett, C.J., Böttger, D., Feldhaar, H., Grella, N., Falconí-López, A., Donoso, D.A., Moriniere, J., Buřivalová, Z., 2023. Author correction: soundscapes and deep learning enable tracking biodiversity recovery in tropical forests. Nature Communications14, 7014.

[24] Nodi, S.S., Paul, M., Robinson, N., Wang, L., Rehman, S.U., Kabir, M.A., 2025. Munsell soil colour prediction from the soil and soil colour book using patching method and deep learning techniques. Sensors25, 287.

[25] Oriol, T., Pasquet, J., Cortet, J., 2024. Automatic identification of Collembola with deep learning techniques. Ecological Informatics81, 102606.

[26] Philipp, L., Sünnemann, M., Schädler, M., Blagodatskaya, E., Tarkka, M., Eisenhauer, N., Reitz, T., 2025. Soil depth shapes the microbial response to land use and climate change in agroecosystems. Applied Soil Ecology209, 106025.

[27] Porco, D., Skarżyński, D., Decaëns, T., Hebert, P.D.N., Deharveng, L., 2014. Barcoding the Collembola of Churchill: a molecular taxonomic reassessment of species diversity in a sub-Arctic area. Molecular Ecology Resources14, 249–261.

[28] Qiao, Z.H., Liu, D., Gong, X., Schädler, M., Zhang, S.C., Yan, Q.B., Liu, X.Y., Xie, Z.J., Chang, L., Wu, D.H., Scheu, S., Sun, X., 2025. Land-use change reshapes communities and guild structure of Collembola across a wide geographic range of the temperate zone. Applied Soil Ecology209, 106036.

[29] Qiao, Z.H., Wang, B., Yao, H.F., Li, Z.P., Scheu, S., Zhu, Y.G., Sun, X., 2022. Urbanization and greenspace type as determinants of species and functional composition of collembolan communities. Geoderma428, 116175.

[30] Redmon, J., Divvala, S., Girshick, R., Farhadi, A., 2016. You only look once: Unified, real-time object detection. In: Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE779–788.

[31] Refaeilzadeh, P., Tang, L., Liu, H., 2016. Cross-validation. In: Liu, L., Özsu, M.T., eds. New York: Springer1–7.

[32] Ren, S.Q., He, K.M., Girshick, R., Sun, J., 2017. Faster R-CNN: Towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence39, 1137–1149.

[33] Schneider, S., Taylor, G.W., Kremer, S.C., Burgess, P., McGroarty, J., Mitsui, K., Zhuang, A., Dewaard, J.R., Fryxell, J.M., 2022. Bulk arthropod abundance, biomass and diversity estimation using deep learning for computer vision. Methods in Ecology and Evolution13, 346–357.

[34] Shabrina, N.H., Lika, R.A., Indarti, S., 2023. Deep learning models for automatic identification of plant-parasitic nematode. Artificial Intelligence in Agriculture7, 1–12.

[35] Singh, B., Davis, L.S., 2018. An analysis of scale invariance in object detection – SNIP. In: Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City: IEEE, 3578–3587.

[36] Sinnott, R.O., Yang, D.H., Ding, X.Y., Ye, Z.Y., 2020. Poisonous spider recognition through deep learning. In: Proceedings of the Australasian Computer Science Week Multiconference. Melbourne: ACM14.

[37] Sys, S., Weißbach, S., Jakob, L., Gerber, S., Schneider, C., 2022. CollembolAI, a macrophotography and computer vision workflow to digitize and characterize samples of soil invertebrate communities preserved in fluid. Methods in Ecology and Evolution13, 2729–2742.

[38] Teixeira, A.C., Ribeiro, J., Morais, R., Sousa, J.J., Cunha, A., 2023. A systematic review on automatic insect detection using deep learning. Agriculture13, 713.

[39] Turnbull, M.S., Stebaeva, S., 2019. Collembola of Canada. Zookeys187–195.

[40] Twardowski, J.P., Hurej, M., Gruss, I., 2016. Diversity and abundance of springtails (Hexapoda: Collembola) in soil under 90-year potato monoculture in relation to crop rotation. Archives of Agronomy and Soil Science62, 1158–1168.

[41] Varghese, R., Sambath, M., 2024. YOLOv8: A novel object detection algorithm with enhanced performance and robustness. In: Proceedings of 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). Chennai: IEEE1–6.

[42] Villon, S., Mouillot, D., Chaumont, M., Darling, E.S., Subsol, G., Claverie, T., Villéger, S., 2018. A Deep learning method for accurate and fast identification of coral reef fishes in underwater images. Ecological Informatics48, 238–244.

[43] Voulodimos, A., Doulamis, N., Doulamis, A., Protopapadakis, E., 2018. Deep learning for computer vision: A brief review. Computational Intelligence and Neuroscience2018, 7068349.

[44] Wu, D.H., Du, E.Z., Eisenhauer, N., Mathieu, J., Chu, C.J., 2025. Global engineering effects of soil invertebrates on ecosystem functions. Nature640, 120–129.

[45] Xiao, Y.Z., Tian, Z.Q., Yu, J.C., Zhang, Y.S., Liu, S., Du, S.Y., Lan, X.G., 2020. A review of object detection based on deep learning. Multimedia Tools and Applications79, 23729–23791.

[46] Zavalloni, M., Targetti, S., Viaggi, D., 2025. Technological innovations for biodiversity monitoring and the design of agri-environmental schemes. Biological Conservation305, 111069.

[47] Zhang, L.Y., Xing, Z.Q., Wang, X.K., 2023. Background instance-based copy-paste data augmentation for object detection. Electronics12, 3781.

RIGHTS & PERMISSIONS

Higher Education Press