Jorma Kilpi

Dr. Jorma Kilpi

VTT Technical Research Centre of Finland Ltd
orcid

Research Focus & Metrics

Academic Career Length

25Years (2002-Present)

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

Representative Publications

View All Publications
1 March 2019 in Remote Sensing of Environment
Verified
Journal Article
Open access
HAHeikki Astola
Heikki Astola
THTuomas Häme
Tuomas Häme
LSLaura Sirro
Laura Sirro
Matthieu MolinierMatthieu Molinier
Matthieu Molinier
JKJorma Kilpi
Jorma Kilpi

1 January 2002
Verified
Conference Paper
Open access
PDF file
JKJorma Kilpi
Jorma Kilpi
INIlkka Norros
Ilkka Norros

6 February 2013 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Closed access
THTuomas Hame
Tuomas Hame
YRYrjö Rauste
Yrjö Rauste
OAOleg Antropov
Oleg Antropov
HAHeikki A. Ahola
Heikki A. Ahola
JKJorma Kilpi
Jorma Kilpi
Performance of the above ground (dry) biomass estimation with the medium resolution optical (ALOS AVNIR) data and radar (ALOS PALSAR) data was evaluated on a tropical forest site in Lao PDR (Laos). The average biomass of ground reference plots was relatively low, 78 t/ha, due to strong anthropogenic influence in most of the study area. The biomass estimates were computed using linear regression analysis and the Probability method that combines unsupervised clustering and fuzzy estimation. The predictions were validated with independent field plot data. With all the methods and data types, the root mean square error (RMSE) ranged from 33.6 t/ha to 40.1 t/ha (44.2% and 52.8% of mean biomass, respectively). The Probability method produced a larger dynamic range to the predictions than the regression models, which saturated at approximately 100 t/ha. Large errors for higher biomass plots increased the RMSE of Probability over the RMSE of the regression models. The bias ranged from -0.8 to 3.9% except with the Probability model for PALSAR data where the bias was 12.5%. Our study showed that PALSAR data were nearly as good for the biomass estimation as the AVNIR data. A combination of mono-temporal ALOS PALSAR and ALOS AVNIR data did not improve biomass estimation over the performance obtained with AVNIR data alone. For the Probability method, ground reference data should be more representative than that available in this study.

14 June 2018 in Remote Sensing
Verified
Journal Article
Open access
PDF file
LSLaura Sirro
Laura Sirro
THTuomas Häme
Tuomas Häme
YRYrjö Rauste
Yrjö Rauste
JKJorma Kilpi
Jorma Kilpi
JHJarno Hämäläinen
Jarno Hämäläinen
FPFernando Paz Pellat
Fernando Paz Pellat
The applicability of optical and synthetic aperture radar (SAR) data for land cover classification to support REDD+ (Reducing Emissions from Deforestation and Forest Degradation) MRV (measuring, reporting and verification) services was tested on a tropical to sub-tropical test site. The 100 km by 100 km test site was situated in the State of Chiapas in Mexico. Land cover classifications were computed using RapidEye and Landsat TM optical satellite images and ALOS PALSAR L-band and Envisat ASAR C-band images. Identical sample plot data from Kompsat-2 imagery of one-metre spatial resolution were used for the accuracy assessment. The overall accuracy for forest and non-forest classification varied between 95% for the RapidEye classification and 74% for the Envisat ASAR classification. For more detailed land cover classification, the accuracies varied between 89% and 70%, respectively. A combination of Landsat TM and ALOS PALSAR data sets provided only 1% improvement in the overall accuracy. The biases were small in most classifications, varying from practically zero for the Landsat TM based classification to a 7% overestimation of forest area in the Envisat ASAR classification. Considering the pros and cons of the data types, we recommend optical data of 10 m spatial resolution as the primary data source for REDD MRV purposes. The results with L-band SAR data were nearly as accurate as the optical data but considering the present maturity of the imaging systems and image analysis methods, the L-band SAR is recommended as a secondary data source. The C-band SAR clearly has poorer potential than the L-band but it is applicable in stratification for a statistical sampling when other image types are unavailable.

22 February 2013 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Closed access
THTuomas Hame
Tuomas Hame
JKJorma Kilpi
Jorma Kilpi
HAHeikki A. Ahola
Heikki A. Ahola
YRYrjö Rauste
Yrjö Rauste
OAOleg Antropov
Oleg Antropov
SBSengthong Bounpone
Sengthong Bounpone
This paper describes an improved concept for the mapping of tropical forest classes with ALOS AVNIR and ALOS PALSAR data. The improvement comes from a combination of a sample of very high resolution (VHR) satellite images with medium resolution wall-to-wall mapping in a statistical sampling framework. The approach developed makes it possible to obtain reliable information on mapping accuracy over the whole area of interest. A simulation study indicated that the sample of VHR images should be collected in a stratified manner using small (25 km) images. The VHR images should cover approximately one percent of the total area of interest, depending on the accuracy requirement. The recommended size of the reference plots (population units) that are selected within the VHR imagery is in the order of 50 m by 50 m. In a systematic selection the plots should be located at a distance of several hundred meters from each other. The forest variables were predicted with an unsupervised fuzzy classification method. The ALOS AVNIR-based forest/non-forest mapping accuracies varied between 68% and 97% of the areas of the VHR images. The corresponding ALOS PALSAR mapping accuracies were poorer. At AVNIR resolution, the area of natural forest was over-estimated, and the degree of disturbance underestimated in humid, heavily disturbed parts of the study area in Laos. The three predictions for the total forest fraction from VHR, AVNIR and PALSAR data over the area that was covered by the VHR images were 55.1%, 53.6%, and 52.8%, respectively.

11 January 2024 in Heliyon
Verified
Journal Article
Open access
HRHannu Rummukainen
Hannu Rummukainen
HHHanna Hörhammer
Hanna Hörhammer
PKPirkko Kuusela
Pirkko Kuusela
JKJorma Kilpi
Jorma Kilpi
JSJari Sirviö
Jari Sirviö
MMMikko Mäkelä
Mikko Mäkelä
Traditional design of experiments and response surface methodology are widely used in engineering and process development. Bayesian optimization is an alternative machine learning approach that adaptively selects successive experimental conditions based on a predefined performance measure. Here we compared the two approaches using simulations and empirical experiments on alkaline wood delignification to identify important benefits and drawbacks of Bayesian optimization in the context of design of experiments. The simulations showed that the selection of initial experiments and measurement noise influenced the convergence of the Bayesian optimization algorithm to known optimal conditions. Both methods, however, showed comparable pilot-scale results on optimal digestion conditions, where high cellulose yields were combined with acceptable kappa numbers and pulp viscosities. Bayesian optimization did not enable a decrease in the number of experiments required for reaching these conditions but provided a more accurate model in the vicinity of the optimum based on additional modelling and cross-validation. These results shed light on the practical differences between the two methodologies for process development and are an important contribution to the chemometrics and machine learning communities.