Olli Koskela

Mr. Olli Koskela

orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
12 July 2022 in Energies
Verified
Journal Article
Open access
PDF file
IMIivo Metsä-Eerola
Iivo Metsä-Eerola
Jukka PulkkinenJukka Pulkkinen
Jukka Pulkkinen
ONOlli Niemitalo
Olli Niemitalo
OKOlli Koskela
Olli Koskela
Optimizing the heating, ventilation, and air conditioning (HVAC) system to minimize district heating usage in large groups of managed buildings is of the utmost important, and it requires a machine learning (ML) model to predict the energy consumption. An industrial use case to reach large building groups is restricted to using normal operational data in the modeling, and this is one reason for the low utilization of ML in HVAC optimization. We present a methodology to select the best-fitting ML model on the basis of both Bayesian optimization of black-box models for defining hyperparameters and a fivefold cross-validation for the assessment of each model’s predictive performance. The methodology was tested in one case study using normal operational data, and the model was applied to analyze the energy savings in two different practical scenarios. The software for the modeling is published on GitHub. The results were promising in terms of predicting the energy consumption, and one of the scenarios also showed energy saving potential. According to our research, the GitHub software for the modeling is a good candidate for predicting the energy consumption in large building groups, but further research is needed to explore its scalability for several buildings.

17 January 2018 in Measurement Science and Technology
Verified
Journal Article
Open access
MLMari Lehti-Polojärvi
Mari Lehti-Polojärvi
OKOlli Koskela
Olli Koskela
ASAku Seppänen
Aku Seppänen
EFEdite Figueiras
Edite Figueiras
JHJari Hyttinen
Jari Hyttinen

26 September 2019 in Scientific Reports
Verified
Journal Article
Open access
PDF file
OKOlli Koskela
Olli Koskela
TMToni Montonen
Toni Montonen
BBBirhanu Belay
Birhanu Belay
EFEdite Figueiras
Edite Figueiras
SPSampsa Pursiainen
Sampsa Pursiainen
JHJari Hyttinen
Jari Hyttinen
This study focuses on improving the reconstruction process of the brightfield optical projection tomography (OPT). OPT is often described as the optical equivalent of X-ray computed tomography, but based on visible light. The detection optics used to collect light in OPT focus on a certain distance and induce blurring in those features out of focus. However, the conventionally used inverse Radon transform assumes an absolute focus throughout the propagation axis. In this study, we model the focusing properties of the detection by coupling Gaussian beam model (GBM) with the Radon transform. The GBM enables the construction of a projection operator that includes modeling of the blurring caused by the light beam. We also introduce the concept of a stretched GBM (SGBM) in which the Gaussian beam is scaled in order to avoid the modeling errors related to the determination of the focal plane. Furthermore, a thresholding approach is used to compress memory usage. We tested the GBM and SGBM approaches using simulated and experimental data in mono- and multifocal modes. When compared with the traditionally used filtered backprojection algorithm, the iteratively computed reconstructions, including the Gaussian models GBM and SGBM, provided smoother images with higher contrast.

16 June 2022 in Agricultural and Food Science
Verified
Journal Article
Open access
PDF file
OKOlli Koskela
Olli Koskela
LPLeonardo Santiago Benitez Pereira
Leonardo Santiago Benitez Pereira
IPIlpo Pölönen
Ilpo Pölönen
IAIlmo Aronen
Ilmo Aronen
IKIivari Kunttu
Iivari Kunttu
Production animals enjoying good health and well-being are more productive and have a higher output quality. Several technical solutions have been used to monitor the animals’ welfare: those based on computer vision provide cost-efficient and scalable options. In this work, we performed a continuous two-month image acquisition of cows in front of an automatic milking station and divided the data into ten different classes related to the most important activities appearing in the images. The data consisted of almost 19 hours of video, equivalent to more than 1.7 million still images. Based on these imaged, we then implemented a convolutional neural network classifier to recognize the cows' behavior. The network was tested using cross-validation methodology and achieved an 86% precision rate and 85% recall rate in the classification. The data and the Python program code used in this study are made available. An image data set that directly resembles the harsh conditions inside a barn and can be used for deep learning purposes has not been previously made available.

1 February 2019 in Physics in Medicine & Biology
Verified
Journal Article
Open access
PDF file
VKVille Oskari Koljonen
Ville Oskari Koljonen
OKOlli Koskela
Olli Koskela
TMToni Montonen
Toni Montonen
ARAtena Rezaei
Atena Rezaei
BBBirhanu Belay
Birhanu Belay
SPSampsa Pursiainen
Sampsa Pursiainen
Solving the fluorophore distribution in a tomographic setting has been difficult because of the lack of physically meaningful and computationally applicable propagation models. This study concentrates on the direct modelling of fluorescence signals in optical projection tomography (OPT), and on the corresponding inverse problem. The reconstruction problem is solved using emission projections corresponding to a series of rotational imaging positions of the sample. Similarly to the bright field OPT bearing resemblance with the transmission x-ray computed tomography, the fluorescent mode OPT is analogous to x-ray fluorescence tomography (XFCT). As an improved direct model for the fluorescent OPT, we derive a weighted Radon transform based on the XFCT literature. Moreover, we propose a simple and fast iteration scheme for the slice-wise reconstruction of the sample. The developed methods are applied in both numerical experiments and inversion of fluorescent OPT data from a zebrafish embryo. The results demonstrate the importance of propagation modelling and our analysis provides a flexible modelling framework for fluorescent OPT that can easily be modified to adapt to different imaging setups.

2 March 2021 in Technology Innovation Management Review
Verified
Journal Article
Open access
ONOlli Niemitalo
Olli Niemitalo
EKEero Koskinen
Eero Koskinen
JHJari Hyväluoma
Jari Hyväluoma
ELEsa Lientola
Esa Lientola
HLHenrik Lindberg
Henrik Lindberg
IKIivari Kunttu
Iivari Kunttu
IntroductionThe development of digitalization and measurement technologies in recent decades has enabled digital devices and sensors to produce huge amounts of data that has great potential in optimizing processes related to production chains or service production. In the field of bioeconomy, the main production processes are related to food and biomass production.