Search the digital library catalog

Query:	search in
	search in
	search in
	search in
Work type:	* old and bologna study programme
Language:
Search in:
Options:	Show only hits with full text
	Reset

1 - 3 / 3

Long-term temperature prediction with hybrid autoencoder algorithms
Jorge Pérez-Aracil, Dušan Fister, C. M. Marina, César Peláez-Rodriguez, L. Cornejo-Bueno, P. A. Gutiérrez, Matteo Giuliani, A. Castelleti, Sancho Salcedo-Sanz, 2024, original scientific article

Abstract: This paper proposes two hybrid approaches based on Autoencoders (AEs) for long-term temperature prediction. The first algorithm comprises an AE trained to learn temperature patterns, which is then linked to a second AE, used to detect possible anomalies and provide a final temperature prediction. The second proposed approach involves training an AE and then using the resulting latent space as input of a neural network, which will provide the final prediction output. Both approaches are tested in long-term air temperature prediction in European cities: seven European locations where major heat waves occurred have been considered. The longterm temperature prediction for the entire year of the heatwave events has been analysed. Results show that the proposed approaches can obtain accurate long-term (up to 4 weeks) temperature prediction, improving Persistence and Climatology in the benchmark models compared. In heatwave periods, where the persistence of the temperature is extremely high, our approach beat the persistence operator in three locations and works similarly in the rest of the cases, showing the potential of this AE-based method for long-term temperature prediction.
Keywords: autoencoder, temperature prediction, hybrid models, heatwave
Published in DKUM: 29.01.2025; Views: 0; Downloads: 2
Full text (1,82 MB)

Background purification framework with extended morphological attribute profile for hyperspectral anomaly detection
Ju Huang, Kang Liu, Mingliang Xu, Matjaž Perc, Xuelong Li, 2021, original scientific article

Abstract: Hyperspectral anomaly detection has attracted extensive interests for its wide use in military and civilian fields, and three main categories of detection methods have been developed successively over past few decades, including statistical model-based, representation-based, and deep-learning-based methods. Most of these algorithms are essentially trying to construct proper background profiles, which describe the characteristics of background and then identify the pixels that do not conform to the profiles as anomalies. Apparently, the crucial issue is how to build an accurate background profile; however, the background profiles constructed by existing methods are not accurate enough. In this article, a novel and universal background purification framework with extended morphological attribute profiles is proposed. It explores the spatial characteristic of image and removes suspect anomaly pixels from the image to obtain a purified background. Moreover, three detectors with this framework covering different categories are also developed. The experiments implemented on four real hyperspectral images demonstrate that the background purification framework is effective, universal, and suitable. Furthermore, compared with other popular algorithms, the detectors with the framework perform well in terms of accuracy and efficiency.
Keywords: detectors, anomaly detection, image reconstruction, hyperspectral imaging, training, optics, dictionaries, background purification, extended attribute profile, sparse representation, stacked autoencoder
Published in DKUM: 19.08.2024; Views: 92; Downloads: 9
Full text (5,36 MB)
This document has many files! More...

NiaNetAD: Autoencoder architecture search for tabular anomaly detection powered by HPC
Sašo Pavlič, Sašo Karakatič, Iztok Fister, 2023, published scientific conference contribution

Keywords: predictive maintenance, autoencoder, anomaly detection, nature-inspired algorithms, optimization, high-performance computing, unsupervised learning
Published in DKUM: 26.01.2024; Views: 447; Downloads: 0

Search done in 0.07 sec.