Multi-Input Convolutional Neural Networks for Automatic Pollen Classification
Source abstract
Abstract
Pollen allergies are a cause of much suffering for an increasing number of individuals. Current pollen monitoring techniques are lacking due to their reliance on manual counting and classification of pollen by human technicians. In this study, we present a neural network architecture capable of distinguishing pollen species using data from an automated particle measurement device. This work presents an improvement over the current state of the art in the task of automated pollen classification, using fluorescence spectrum data of aerosol particles. We obtained a relative reduction in the error rate of over 48%, from 27% to 14%, for one of the datasets, with similar improvements for the other analyzed datasets. We also use a novel approach for doing hyperparameter tuning for multiple input networks.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Allergic Rhinitis and Sensitization · Advanced Chemical Sensor Technologies · Insect Pheromone Research and Control
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
CAMPUS Research Institute, University Politehnica of Bucharest, 060042 Bucharest, Romania
National Institute for Research and Development for Optoelectronics INOE 2000, 077125 Măgurele, Romania
References (23)
Show source-provided references
1. Advances in the clinical and mechanism research of pollen induced seasonal allergic asthma
2. The experimental research of Charles H. Blackley10.1007/BF02136147 ↗
3. An automatic volumetric spore trap10.1111/j.1744-7348.1952.tb00904.x ↗
4. Sampling bias and sampling errors in pollen counting in aerobiological monitoring in Italy10.1039/b818162b ↗
5. Clowes, M.B., and Penny, J.P. (1968). The problem of pollen recognition. Problems of Picture Interpretation, CSIRO.
6. Principles and methods for automated palynology10.1111/nph.12848 ↗
7. All-optical automatic pollen identification: Towards an operational system10.1016/j.atmosenv.2016.05.062 ↗
8. Automatic pollen recognition with the Rapid-E particle counter: The first-level procedure, experience and next steps10.5194/amt-12-3435-2019 ↗
9. Boldeanu, M., Cucu, H., Burileanu, C., and Mărmureanu, L. (2021, January 26–28). Automatic Pollen Classification Using Convolutional Neural Networks. Proceedings of the 2021 44th International Conference on Telecommunications and Signal Processing (TSP), Brno, Czech Republic.10.1109/TSP52935.2021.9522626 ↗
10. Boldeanu, M., Marin, C., Ene, D., Mărmureanu, L., Cucu, H., and Burileanu, C.L. (2021, January 13–15). MARS: The First Romanian Pollen Dataset using a Rapid-E Particle Analyzer. Proceedings of the 2021 International Conference on Speech Technology and Human-Computer Dialogue (SpeD), Bucharest, Romania.10.1109/SpeD53181.2021.9587447 ↗
11. Ahmed Raza, S.E. (2017, January 18–21). MIMO-Net: A multi-input multi-output convolutional neural network for cell segmentation in fluorescence microscopy images. Proceedings of the 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017), Melbourne, VIC, Australia.10.1109/ISBI.2017.7950532 ↗
12. Deep learning10.1038/nature14539 ↗
13. Multilayer perceptron, fuzzy sets, and classification10.1109/72.159058 ↗
14. Agarap, A.F. (2018). Deep learning using rectified linear units (relu). arXiv.
15. Ioffe, S. (2015, January 6–11). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. Proceedings of the ICML’15: 32nd International Conference on International Conference on Machine Learning, 2015, Lille, France.
16. Nagi, J. (2011, January 16–18). Max-pooling convolutional neural networks for vision-based hand gesture recognition. Proceedings of the 2011 IEEE International Conference on Signal and Image Processing Applications (ICSIPA), Kuala Lumpur, Malaysia.10.1109/ICSIPA.2011.6144164 ↗
17. Dropout: A simple way to prevent neural networks from overfitting
18. Zhang, Z. (2018, January 3–8). Generalized Cross Entropy Loss for Training Deep Neural Networks with Noisy Labels. Proceedings of the 32nd Conference on Neural Information Processing Systems(NeurIPS 2018), Montréal, QC, Canada.
19. Shekar, B.H. (2019, January 25–28). Grid Search-Based Hyperparameter Tuning and Classification of Microarray Cancer Data. Proceedings of the 2019 Second International Conference on Advanced Computational and Communication Paradigms (ICACCP), Sikkim, India.10.1109/ICACCP.2019.8882943 ↗
20. Random Search for Hyper-Parameter Optimization
21. Hyperparameter Optimization for Machine Learning Models Based on Bayesian Optimization
22. Nogueira, F. (2021, October 01). Bayesian Optimization: Open Source Constrained Global Optimization Tool for Python. Available online: https://www.github.com/fmfn/BayesianOptimization.
23. Pollen grains as airborne allergenic particles10.5586/aa.2015.045 ↗