{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T04:58:51Z","timestamp":1777006731790,"version":"3.51.4"},"reference-count":63,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2026,4,17]],"date-time":"2026-04-17T00:00:00Z","timestamp":1776384000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Odour recognition with low-cost gas sensors is challenging in open-air settings due to the non-specificity of the sensors and environmental variability. This can be mitigated by incorporating additional information into the classification process. This paper investigates odour-image multimodality in two case-studies of increasing complexity: banana ripening in open-air environment and strawberry ripening in a glasshouse environment. Data were collected using custom acquisition platforms equipped with cameras and MOX gas sensors operated with temperature modulation. For the visual modality, image classification (MobileNetV3) and object detection (YoloV5) models are trained. For the odour modality, established classical machine learning methods (Random Forest, XGBoost, SVM and Logistic Regression) and neural networks (1D-CNN, LSTM, MLP, and ELM) are employed. Each modality is analysed independently and together to critically assess scenarios in which combining modalities provides a clear advantage over using either modality alone. Results show that models trained on odour data achieve high accuracy in controlled environments but underperform in more dynamic open-air settings. Image-based models are sensitive to the image quality in all environments; however, they are more robust when deployed in different environments. Lastly, it is demonstrated that decision fusion consistently increases the accuracy, by as much as +12.36% in the banana ripening and +3.63% in the strawberry ripening scenario. Where decision fusion does not improve classification accuracy significantly, it is shown that the multimodal approach can still be leveraged to identify high-confidence predictions by selecting samples where both modalities agree on the label.<\/jats:p>","DOI":"10.3390\/s26082493","type":"journal-article","created":{"date-parts":[[2026,4,17]],"date-time":"2026-04-17T14:24:51Z","timestamp":1776435891000},"page":"2493","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Improving the Robustness of Odour Recognition with Odour-Image Data Fusion in Open-Air Settings"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6099-3321","authenticated-orcid":false,"given":"Fanny","family":"Monori","sequence":"first","affiliation":[{"name":"Department of Electronic Engineering, Royal Holloway, University of London, Egham TW20 0EX, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8280-2722","authenticated-orcid":false,"given":"Alin","family":"Tisan","sequence":"additional","affiliation":[{"name":"Department of Electronic Engineering, Royal Holloway, University of London, Egham TW20 0EX, UK"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,17]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1145\/3545572","article-title":"A Review on Methods and Applications in Multimodal Deep Learning","volume":"19","author":"Jabeen","year":"2023","journal-title":"ACM Trans. Multimed. Comput. Commun. Appl."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1145\/3713070","article-title":"A Survey of Multimodal Learning: Methods, Applications, and Future","volume":"57","author":"Yuan","year":"2025","journal-title":"ACM Comput. Surv."},{"key":"ref_3","unstructured":"Bellot, D., Boyer, A., and Charpillet, F. (2002, January 8\u201311). A new definition of qualified gain in a data fusion process: Application to telemedicine. Proceedings of the Fifth International Conference on Information Fusion (FUSION 2002), Annapolis, MD, USA."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Gadzicki, K., Khamsehashari, R., and Zetzsche, C. (2020, January 6\u20139). Early vs Late Fusion in Multimodal Convolutional Neural Networks. Proceedings of the 2020 IEEE 23rd International Conference on Information Fusion (FUSION 2020), Rustenburg, South Africa.","DOI":"10.23919\/FUSION45008.2020.9190246"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"6247","DOI":"10.1109\/ACCESS.2017.2694050","article-title":"The Fall of One, the Rise of Many: A Survey on Multi-Biometric Fusion Methods","volume":"5","author":"Dinca","year":"2017","journal-title":"IEEE Access"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"171","DOI":"10.1038\/s41746-022-00712-8","article-title":"Multimodal machine learning in precision health: A scoping review","volume":"5","author":"Kline","year":"2022","journal-title":"NPJ Digit. Med."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"2939","DOI":"10.1007\/s00371-021-02166-7","article-title":"A survey on deep multimodal learning for computer vision: Advances, trends, applications, and datasets","volume":"38","author":"Bayoudh","year":"2022","journal-title":"Vis. Comput."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2200213","DOI":"10.1002\/aisy.202200213","article-title":"Multimodal Sensors and ML-Based Data Fusion for Advanced Robots","volume":"4","author":"Duan","year":"2022","journal-title":"Adv. Intell. Syst."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1007\/s00604-006-0623-9","article-title":"Data analysis for electronic nose systems","volume":"156","author":"Scott","year":"2006","journal-title":"Microchim. Acta"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1016\/j.snb.2012.01.074","article-title":"Chemical gas sensor drift compensation using classifier ensembles","volume":"166\u2013167","author":"Vergara","year":"2012","journal-title":"Sens. Actuators B Chem."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"131668","DOI":"10.1016\/j.snb.2022.131668","article-title":"Drift in a popular metal oxide sensor dataset reveals limitations for gas classification benchmarks","volume":"361","author":"Dennler","year":"2022","journal-title":"Sens. Actuators B Chem."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1016\/j.snb.2013.05.027","article-title":"On the performance of gas sensor arrays in open sampling systems using Inhibitory Support Vector Machines","volume":"185","author":"Vergara","year":"2013","journal-title":"Sens. Actuators B Chem."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"9635","DOI":"10.3390\/s120709635","article-title":"A Survey on Gas Sensing Technology","volume":"12","author":"Liu","year":"2012","journal-title":"Sensors"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"134814","DOI":"10.1016\/j.foodchem.2022.134814","article-title":"Recent advances in chromatography-mass spectrometry and electronic nose technology in food flavor analysis and detection","volume":"405","author":"Wei","year":"2023","journal-title":"Food Chem."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Esfahani, S., Tiele, A., Agbroko, S.O., and Covington, J.A. (2020). Development of a Tuneable NDIR Optical Electronic Nose. Sensors, 20.","DOI":"10.3390\/s20236875"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"11202","DOI":"10.1038\/s41598-021-90471-5","article-title":"Supervised binary classification methods for strawberry ripeness discrimination from bioimpedance data","volume":"11","author":"Ibba","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_17","unstructured":"United States Department of Agriculture, Agricultural Marketing Service (2026, March 18). Strawberries Grades and Standards, Available online: https:\/\/www.ams.usda.gov\/grades-standards\/strawberries-grades-and-standards."},{"key":"ref_18","unstructured":"European Commission (2026, March 18). Commission Implementing Regulation (EU) No 543\/2011 Laying Down Detailed Rules for the Application of Council Regulation (EC) No 1234\/2007 in Respect of the Fruit and Vegetables and Processed Fruit and Vegetables Sectors, Annex I, Part B, Part 7 (Strawberries). Official Journal of the European Union, L 157\/1, 15 June 2011. 2011. Available online: https:\/\/eur-lex.europa.eu\/eli\/reg_impl\/2011\/543\/oj."},{"key":"ref_19","unstructured":"European Commission (2026, March 18). Commission Implementing Regulation (EU) No 1333\/2011 of 19 December 2011 Laying Down Marketing Standards for Bananas, Rules on the Verification of Compliance with Those Marketing Standards and Requirements for Notifications in the Banana Sector. Official Journal of the European Union, L 336\/23, 20 Dec 2011. 2011. Available online: https:\/\/eur-lex.europa.eu\/eli\/reg_impl\/2011\/1333\/oj."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1016\/j.jksuci.2018.06.002","article-title":"Fruits and vegetables quality evaluation using computer vision: A review","volume":"33","author":"Bhargava","year":"2021","journal-title":"J. King Saud Univ.-Comput. Inf. Sci."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Yu, K., Zhong, M., Zhu, W., Rashid, A., Han, R., Virk, M.S., Duan, K., Zhao, Y., and Ren, X. (2025). Advances in computer vision and spectroscopy techniques for non-destructive quality assessment of citrus fruits: A comprehensive review. Foods, 14.","DOI":"10.3390\/foods14030386"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Abbas, I., Liu, J., Amin, M., Tariq, A., and Tunio, M.H. (2021). Strawberry Fungal Leaf Scorch Disease Identification in Real-Time Strawberry Field Using Deep Learning Architectures. Plants, 10.","DOI":"10.3390\/plants10122643"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"108304","DOI":"10.1016\/j.compag.2023.108304","article-title":"DSW-YOLO: A detection method for ground-planted strawberry fruits under different occlusion levels","volume":"214","author":"Du","year":"2023","journal-title":"Comput. Electron. Agric."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"43","DOI":"10.1016\/j.compag.2018.01.011","article-title":"A methodology for fresh tomato maturity detection using computer vision","volume":"146","author":"Wan","year":"2018","journal-title":"Comput. Electron. Agric."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"508","DOI":"10.1080\/23335777.2025.2467639","article-title":"A review of fruit ripeness recognition methods based on deep learning","volume":"11","author":"Zhou","year":"2025","journal-title":"Cyber-Phys. Syst."},{"key":"ref_26","first-page":"259","article-title":"Electronic nose for detecting strawberry fruit maturity","volume":"123","author":"Du","year":"2010","journal-title":"Proc. Fla. State Hortic. Sci."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Palumbo, M., Cozzolino, R., Laurino, C., Malorni, L., Picariello, G., Siano, F., Stocchero, M., Cefola, M., Corvino, A., and Romaniello, R. (2022). Rapid and Non-Destructive Techniques for the Discrimination of Ripening Stages in Candonga Strawberries. Foods, 11.","DOI":"10.3390\/foods11111534"},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Chen, L.Y., Wu, C.C., Chou, T.I., Chiu, S.W., and Tang, K.T. (2018). Development of a Dual MOS Electronic Nose\/Camera System for Improving Fruit Ripeness Classification. Sensors, 18.","DOI":"10.3390\/s18103256"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1016\/j.compag.2006.07.002","article-title":"Evaluation of tomato maturity by electronic nose","volume":"54","author":"Hu","year":"2006","journal-title":"Comput. Electron. Agric."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"9272","DOI":"10.1007\/s11694-025-03699-4","article-title":"Advancements in quality assessment of fruits and vegetables: A comprehensive review of E-nose technology","volume":"19","author":"Rahman","year":"2025","journal-title":"J. Food Meas. Charact."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"105810","DOI":"10.1016\/j.engappai.2022.105810","article-title":"Strawberry localization in a ridge planting with an autonomous rover","volume":"119","author":"Mejia","year":"2023","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"24","DOI":"10.1016\/j.ifacol.2022.11.109","article-title":"Real-Time CNN-based Computer Vision System for Open-Field Strawberry Harvesting Robot","volume":"55","author":"Lemsalu","year":"2022","journal-title":"IFAC-PapersOnLine"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"100001","DOI":"10.1016\/j.atech.2021.100001","article-title":"Strawberry Maturity Classification from UAV and Near-Ground Imaging Using Deep Learning","volume":"1","author":"Zhou","year":"2021","journal-title":"Smart Agric. Technol."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"1439","DOI":"10.1109\/JIOT.2023.3288603","article-title":"IoT-Based Strawberry Disease Detection with Wall-Mounted Monitoring Cameras","volume":"11","author":"Lin","year":"2024","journal-title":"IEEE Internet Things J."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"132651","DOI":"10.1016\/j.foodchem.2022.132651","article-title":"Fusion of electronic nose and hyperspectral imaging for mutton freshness detection using input-modified convolution neural network","volume":"385","author":"Liu","year":"2022","journal-title":"Food Chem."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"Kodogiannis, V.S., and Alshejari, A. (2025). Data Fusion of Electronic Nose and Multispectral Imaging for Meat Spoilage Detection Using Machine Learning Techniques. Sensors, 25.","DOI":"10.3390\/s25103198"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"152","DOI":"10.1016\/j.postharvbio.2019.03.017","article-title":"Information fusion of hyperspectral imaging and electronic nose for evaluation of fungal contamination in strawberries during decay","volume":"153","author":"Liu","year":"2019","journal-title":"Postharvest Biol. Technol."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"110034","DOI":"10.1016\/j.microc.2024.110034","article-title":"Fusion of hyperspectral imaging and electronic nose for identification of green vegetable in egg pancakes","volume":"199","author":"Gao","year":"2024","journal-title":"Microchem. J."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"113778","DOI":"10.1016\/j.lwt.2022.113778","article-title":"A decision fusion method based on hyperspectral imaging and electronic nose techniques for moisture content prediction in frozen-thawed pork","volume":"165","author":"Cheng","year":"2022","journal-title":"LWT"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"102454","DOI":"10.1016\/j.fbio.2023.102454","article-title":"Monitoring black tea fermentation quality by intelligent sensors: Comparison of image, e-nose and data fusion","volume":"52","author":"Zhou","year":"2023","journal-title":"Food Biosci."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1016\/j.jfoodeng.2018.07.020","article-title":"Rapid identification of tea quality by E-nose and computer vision combining with a synergetic data fusion strategy","volume":"241","author":"Xu","year":"2019","journal-title":"J. Food Eng."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"e12708","DOI":"10.1111\/jfs.12708","article-title":"Rapid and nondestructive detection of freshness quality of postharvest spinaches based on machine vision and electronic nose","volume":"39","author":"Huang","year":"2019","journal-title":"J. Food Saf."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"e12832","DOI":"10.1111\/jfpe.12832","article-title":"Evaluating quality of tomato during storage using fusion information of computer vision and electronic nose","volume":"41","author":"Huang","year":"2018","journal-title":"J. Food Process Eng."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"7317","DOI":"10.1109\/JSEN.2023.3328915","article-title":"AIoT-Based Meat Quality Monitoring Using Camera and Gas Sensor with Wireless Charging","volume":"24","author":"Kim","year":"2024","journal-title":"IEEE Sens. J."},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Narkhede, P., Walambe, R., Mandaokar, S., Chandel, P., Kotecha, K., and Ghinea, G. (2021). Gas Detection and Identification Using Multimodal Artificial Intelligence Based Sensor Fusion. Appl. Syst. Innov., 4.","DOI":"10.3390\/asi4010003"},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1016\/j.snb.2013.04.056","article-title":"Active temperature modulation of metal-oxide sensors for quantitative analysis of gas mixtures","volume":"185","author":"Gosangi","year":"2013","journal-title":"Sens. Actuators B Chem."},{"key":"ref_47","doi-asserted-by":"crossref","unstructured":"Hasenfratz, D., Saukh, O., and Thiele, L. (2012). On-the-Fly Calibration of Low-Cost Gas Sensors. Proceedings of the Wireless Sensor Networks, Springer.","DOI":"10.1007\/978-3-642-28169-3_15"},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"105103","DOI":"10.1016\/j.compag.2019.105103","article-title":"A framework for predicting soft-fruit yields and phenology using embedded, networked microsensors, coupled weather models and machine-learning techniques","volume":"168","author":"Lee","year":"2020","journal-title":"Comput. Electron. Agric."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"20460","DOI":"10.1109\/JSEN.2023.3302868","article-title":"Recognition Algorithms in E-Nose: A Review","volume":"23","author":"Yang","year":"2023","journal-title":"IEEE Sens. J."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"1790","DOI":"10.1109\/TIM.2014.2367775","article-title":"Domain Adaptation Extreme Learning Machines for Drift Compensation in E-Nose Systems","volume":"64","author":"Zhang","year":"2015","journal-title":"IEEE Trans. Instrum. Meas."},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"134716","DOI":"10.1016\/j.snb.2023.134716","article-title":"Online drift compensation framework based on active learning for gas classification and concentration prediction","volume":"398","author":"Se","year":"2024","journal-title":"Sens. Actuators B Chem."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"110024","DOI":"10.1016\/j.knosys.2022.110024","article-title":"A dual drift compensation framework based on subspace learning and cross-domain adaptive extreme learning machine for gas sensors","volume":"259","author":"Se","year":"2023","journal-title":"Knowl.-Based Syst."},{"key":"ref_53","doi-asserted-by":"crossref","unstructured":"Singh, A.K., and Krishnan, S. (2023). Trends in EEG signal feature extraction applications. Front. Artif. Intell., 5.","DOI":"10.3389\/frai.2022.1072801"},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"22","DOI":"10.1186\/s12938-023-01075-1","article-title":"ECG signal feature extraction trends in methods and applications","volume":"22","author":"Singh","year":"2023","journal-title":"Biomed. Eng. OnLine"},{"key":"ref_55","doi-asserted-by":"crossref","first-page":"111874","DOI":"10.1016\/j.sna.2020.111874","article-title":"An optimized deep convolutional neural network for dendrobium classification based on electronic nose","volume":"307","author":"Wang","year":"2020","journal-title":"Sens. Actuators A Phys."},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"3941","DOI":"10.1007\/s00216-014-7807-7","article-title":"The need for external validation in machine olfaction: Emphasis on health-related applications","volume":"406","author":"Marco","year":"2014","journal-title":"Anal. Bioanal. Chem."},{"key":"ref_57","unstructured":"Masaponto (2026, February 01). Python-ELM. Available online: https:\/\/github.com\/masaponto\/Python-ELM."},{"key":"ref_58","doi-asserted-by":"crossref","unstructured":"Howard, A., Sandler, M., Chu, G., Chen, L.C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., and Vasudevan, V. (2019). Searching for MobileNetV3. arXiv.","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref_59","doi-asserted-by":"crossref","unstructured":"Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and Fei-Fei, L. (2009, January 20\u201325). ImageNet: A large-scale hierarchical image database. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA.","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref_60","unstructured":"PyTorch (2026, February 01). MobileNetV3. Available online: https:\/\/docs.pytorch.org\/vision\/main\/models\/mobilenetv3.html."},{"key":"ref_61","unstructured":"Jocher, G. (2020). Ultralytics YOLOv5, Zenodo."},{"key":"ref_62","unstructured":"Pastell Matti, L.M., and Victor, B. (2022). Strawberry Dataset for Object Detection; Dataset published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), Zenodo."},{"key":"ref_63","doi-asserted-by":"crossref","first-page":"106622","DOI":"10.1016\/j.knosys.2020.106622","article-title":"AutoML: A survey of the state-of-the-art","volume":"212","author":"He","year":"2021","journal-title":"Knowl.-Based Syst."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/8\/2493\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T04:23:53Z","timestamp":1777004633000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/8\/2493"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,17]]},"references-count":63,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["s26082493"],"URL":"https:\/\/doi.org\/10.3390\/s26082493","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,17]]}}}