Project description
I will develop machine learning algorithms to conquer challenges typically encountered automated image-based identification in biomonitoring. The considered challenges are i) unbalanced occurrence of different taxa, while most interesting taxa are rare, ii) variations in imaging conditions, such as lighting, which may harm identification accuracy, iii) detection of rare or invasive taxa that are absent in previously collected datasets, and iv) hierarchical nature of the identification task. I will apply state-of-the-art machine learning techniques and propose improvements to them to tackle the challenges.
Publications related to the project
General publications on taxa identification:
J. Raitoharju and K. Meissner, "On Confidences and Their Use in (Semi-)Automatic Multi-Image Taxa Identification," IEEE Symposium Series on Computational Intelligence (SSCI), 2019. IEEEXplore
J. Ärje, C. Melvad, M.R. Jeppesen, S.A. Madsen, J. Raitoharju, M.S. Rasmussen, A. Iosifidis, V. Tirronen, M. Gabbouj, K. Meissner, and T.T. Høye, "Automatic image-based identification and biomass estimation of invertebrates," Methods in Ecology and Evolution, vol. 11, no. 8, 2020. BES
J. Ärje, J. Raitoharju, A. Iosifidis, V. Tirronen, K. Meissner, M. Gabbouj, S. Kiranyaz, and S. Kärkkäinen, "Human experts vs. machines in taxa recognition," Signal Processing: Image Communication, vol. 87, 2020. ScienceDirect, arXiv
T.T. Høye, J. Ärje, K. Bjerge, O.L.P. Hansen, A. Iosifidis, F. Leese, H.M.R. Mann, K. Meissner, C. Melvad, and J. Raitoharju, "Deep learning and computer vision will transform entomology," Proceedings of the National Academy of Sciences, 2021. PNAS, bioRxiv
Publication on imbalanced classification (task i):
M. Impiö, "On imbalanced classification of benthic macroinvertebrates: Metrics and loss-functions," BSc thesis, (supervision J. Raitoharju), Tampere University, 2020. Trepo
Publications on color constancy and attention (task ii):
F. Laakom, J. Raitoharju, A. Iosifidis, J. Nikkanen, and M. Gabbouj, "Color Constancy Convolutional Autoencoder," IEEE Symposium Series on Computational Intelligence (SSCI), 2019. IEEEXplore, arXiv
F. Laakom, J. Raitoharju, A. Iosifidis, U. Tuna, J. Nikkanen, and M. Gabbouj, "Probabilistic Color Constancy," IEEE International Conference on Image Processing (ICIP), 2020. IEEEXplore, arXiv
F. Laakom, N. Passalis, J. Raitoharju, J. Nikkanen, A. Tefas, A. Iosifidis, and M. Gabbouj, "Bag of Color Features for Color Constancy," IEEE Transactions on Image Processing, vol. 29, 2020. IEEEXplore, arXiv
F. Laakom, J. Raitoharju, J. Nikkanen, A. Iosifidis, and M. Gabbouj, "INTEL-TAU: A Color Constancy Dataset," IEEE Access, vol. 9, pp. 39560-39567, 2021. IEEEXplore
F. Laakom, J. Raitoharju, A. Iosifidis, J. Nikkanen, and M. Gabbouj, "Monte Carlo Dropout Ensembles for Robust Illumination Estimation," International Joint Conference of Neural Networks (IJCNN), 2021. arXiv
F. Laakom, J. Raitoharju, A. Iosifidis, J. Nikkanen, and M. Gabbouj, "Robust Channel-wise Illumination Estimation," British Machine Vision Conference (BMVC), 2021. arXiv
F. Laakom, K. Chumachenko, J. Raitoharju, A. Iosifidis, and M. Gabbouj, "Learning to ignore: rethinking attention in CNNs," British Machine Vision Conference (BMVC), 2021. arXiv
Publications on outlier detection (task iii):
F. Sohrab, J. Raitoharju, A. Iosifidis and M. Gabbouj, "Ellipsoidal Subspace Support Vector Data Description," IEEE Access, vol. 8, pp. 122013-122025, 2020. IEEEXplore, arXiv
F. Sohrab and J. Raitoharju, "Boosting Rare Benthic Macroinvertebrates Taxa Identification With One-Class Classification," IEEE Symposium Series on Computational Intelligence (SSCI), 2020. IEEEXplore, arXiv
F. Sohrab, J. Raitoharju, A. Iosifidis, and M. Gabbouj, "Multimodal subspace support vector data description," Pattern Recognition, vol. 110, 2021. ScienceDirect, arXiv
F. Sohrab, A. Iosifidis and M. Gabbouj, and J. Raitoharju "Graph-Embedded Subspace Support Vector Data Description," Pattern Recognition, vol. 133, pp. 122013-122025, 2022. ScienceDirect
J. Raitoharju and A. Iosifidis, "Generalized Reference Kernel for One-Class Classification", International Joint Conference on Neural Networks (IJCNN), 2022. arXiv
Publications on hierarchical networks (task iv):
N. Passalis, J. Raitoharju, A. Tefas and M. Gabbouj, "Adaptive Inference Using Hierarchical Convolutional Bag-of-Features for Low-Power Embedded Platforms," IEEE International Conference on Image Processing (ICIP), 2019. IEEEXplore
N. Passalis, J. Raitoharju, A. Tefas, and M. Gabbouj, "Efficient adaptive inference for deep convolutional neural networks using hierarchical early exits," Pattern Recognition, vol. 105, 2020. ScienceDirect
N. Passalis, J. Raitoharju, M. Gabbouj and A. Tefas, "Efficient Adaptive Inference Leveraging Bag-of-Features-based Early Exits," IEEE International Workshop on Multimedia Signal Processing (MMSP), 2020. IEEEXplore
Publications on multi-label classification (task iv):
L. Xu, J. Raitoharju, A. Iosifidis and M. Gabbouj, "Saliency-Based Multilabel Linear Discriminant Analysis", IEEE Transactions on Cybernetics, vol. 52, pp.10200-10213m 2022. IEEEXplore
More information
Senior Research Scientist Jenni Raitoharju