Veli-Pekka Kivinen | University of Helsinki (original) (raw)

Uploads

Papers by Veli-Pekka Kivinen

Research paper thumbnail of EMO: A tool for predicting forest composition based on an efficient forest sampling method

Research paper thumbnail of Assessing log geometry and wood quality in standing timber using terrestrial laser-scanning point clouds

Forestry: An International Journal of Forest Research, 2018

Wood procurement in sawmills could be improved by resolving detailed three-dimensional stem geome... more Wood procurement in sawmills could be improved by resolving detailed three-dimensional stem geometry references from standing timber. This could be achieved, using the increasingly available terrestrial point clouds from various sources. Here, we collected terrestrial laser-scanning (TLS) data from 52 Scots pines (Pinus sylvestris L.) with the purpose of evaluating the accuracy of the log geometry and analysing its relationship with wood quality. For reference, the log-specific top-end diameter, volume, tapering, sweep, basic density and knottiness were measured in a sawmill. We produced stem models from the TLS data and bucked them into logs similar to those measured in the sawmill. In comparison to the sawmill data, the log-specific TLSbased top-end diameter, volume, taper and sweep estimates showed relative mean differences of 1.6, −2.4, −3.0 and 78 per cent, respectively. The correlation coefficients between increasing taper and decreasing wood density and whorl-to-whorl distances were 0.49 and −0.51, respectively. Although the stem-model geometry was resolved from the point clouds with similar accuracy to that at the sawmills, the remaining uncertainty in defining the sweep and linking the wood quality with stem geometry may currently limit the method's feasibilities. Instead of static TLS, mobile platforms would likely be more suitable for operational point cloud data acquisition.

Research paper thumbnail of Logging Trail Segmentation via a Novel U-Net Convolutional Neural Network and High-Density Laser Scanning Data

Remote Sensing, 2022

Logging trails are one of the main components of modern forestry. However, spotting the accurate ... more Logging trails are one of the main components of modern forestry. However, spotting the accurate locations of old logging trails through common approaches is challenging and time consuming. This study was established to develop an approach, using cutting-edge deep-learning convolutional neural networks and high-density laser scanning data, to detect logging trails in different stages of commercial thinning, in Southern Finland. We constructed a U-Net architecture, consisting of encoder and decoder paths with several convolutional layers, pooling and non-linear operations. The canopy height model (CHM), digital surface model (DSM), and digital elevation models (DEMs) were derived from the laser scanning data and were used as image datasets for training the model. The labeled dataset for the logging trails was generated from different references as well. Three forest areas were selected to test the efficiency of the algorithm that was developed for detecting logging trails. We designe...

Research paper thumbnail of Factors affecting the output distribution of logs with modern tree bucking optimization system

Аннотация В данной статье рассматривается вопрос о том, что такое самовольное строительство, что ... more Аннотация В данной статье рассматривается вопрос о том, что такое самовольное строительство, что такое объект самовольного строительства, какой правовой статус имеет такая постройка на территории Российской Федерации. Также в статье говорится о том, в соответствии с какими пунктами гражданского кодекса решаются вопросы, связанные с проблемами незаконной постройки. Рассматривается фактическая ситуация в стране по этому вопросу. Ключевые слова: самовольное строительство, самовольная постройка, объект самовольного строительства.

Research paper thumbnail of Implementing SmartForest forest visualization tool on PC environment

Research paper thumbnail of Wood value optimization including logistic costs, production costs and product incomes: a case study in central Finland

Research paper thumbnail of Introduction to tree bucking control

Research paper thumbnail of Applying fuzzy logic to tree bucking control

Forest Science, 2002

In this article, we introduce a fuzzy logic based control system that calibrates the price list c... more In this article, we introduce a fuzzy logic based control system that calibrates the price list controlling the tree bucking of a modern harvester according to the unique conditions of an individual stand prior to harvesting. The system employs a bucking simulator that predicts an output distribution for any price list generated. A fuzzy controller is used to recalculate the price list according to the difference between the demand distribution and the actual output distribution. The tests with perfect knowledge of the stems showed that the control system clearly works efficiently. However, calibrating the price list prior to harvesting seemed to be very sensitive to the precision of the prior information on the trees. The simulations showed that while no gain can be achieved by preharvest inventorying of the stands and fine-tuning the price lists with this imperfect knowledge, the same or even better bucking results can be obtained by the bucking-to-demand process in which the pric...

Research paper thumbnail of Kahden eri katkontamenetelmän vaikutus mäntysahatavaran laatuun

Metsätieteen aikakauskirja, 1970

Research paper thumbnail of Sumea logiikka ja sen soveltaminen hakkuukoneen apteerauksen ohjauksessa

Metsätieteen aikakauskirja, 1970

Research paper thumbnail of Optimisation of wood raw material conversion

Research paper thumbnail of Optimizing logistics of saw logs for a medium sized sawmill

Research paper thumbnail of A genetic algorithm approach to tree bucking optimization

Forest Science, 2004

... the other string, ie, no mixing between various stands was allowed. Several crossover operato... more ... the other string, ie, no mixing between various stands was allowed. Several crossover operators for real-valued representations have been suggested (Beasley et al. 1993b). In this study, a price matrix pair was combined into ...

Research paper thumbnail of Estimating the accuracy and time consumption of a mobile machine vision application in measuring timber stacks

Computers and Electronics in Agriculture, 2019

Research paper thumbnail of Leimikon ennakkomittauksen ajanmenekki

Metsätieteen aikakauskirja, 1970

Leimikon ennakkomittauksella tarkoitetaan kevyttä otantaan perustuvaa pystymittausta, jonka tavoi... more Leimikon ennakkomittauksella tarkoitetaan kevyttä otantaan perustuvaa pystymittausta, jonka tavoitteena on tuottaa ennakkotietoa jalostukseen tulevan leimikon puulajikohtaisista läpimitta-, pituus- ja laatujakaumista. Tutkimuksen tavoitteena oli selvittää leimikkotekijöiden vaikutus leimikon ennakkomittauksen eri työnosien ajanmenekkiin erilaisissa päätehakkuuoloissa sekä edelleen luoda matemaattiset mallit ennakkomittauksen työnosien ajanmenekille kahdella eri menetelmällä. Tutkimuksessa kehitettäviä ajanmenekkifunktioita käytetään ensisijaisesti leimikon ennakkomittausmenetelmien kehittämistyössä.

Research paper thumbnail of A step towards optimal wood supply chain: A case study on optimal tree bucking in Central Finland

In this paper we estimate how large a net profit can be achieved provided that a forest-level tre... more In this paper we estimate how large a net profit can be achieved provided that a forest-level tree bucking optimum can be achieved within a normal Finnish wood supply chain. We demonstrate a holistic system where all costs and revenues within the wood supply chain are calculated. The forest-level optimum of the complete processing chain is then solved with the GA-based algorithm. We calculate both the maximum and minimum net profit for the forest-level bucking optimization problem and find out what could be the maximum difference between the maximum and minimum net profit. The system is demonstrated and validated with the aid of case study material collected from Central Finland. The GAbased stand level allocation system seems to produce logical solutions. There are no clear differences in logistics costs between the best and the worst feasible solution, but there needs to be an explanation of the differences in processing costs and revenues. It seems that there is a possibility to ...

Research paper thumbnail of On the analysis of cubic smoothing spline-based stem curve prediction for forest harvesters

Canadian Journal of Forest Research, 2006

Research paper thumbnail of Comparison of four measures designed for assessing the fit between the demand and output distributions of logs

Canadian Journal of Forest Research, 2005

ABSTRACT In recent years, as customer-oriented production strategies have gained ground, especial... more ABSTRACT In recent years, as customer-oriented production strategies have gained ground, especially in the sawmill industry, the fit between the demand and the actual output for logs of different sizes and qualities has become an important criterion for evaluating bucking outcomes. In this paper, we present four measures for determining the similarity between the demand and output log distributions: (1) the apportionment degree, (2) the χ2 statistic, (3) Laspeyres' quantity index, and (4) the price-weighted apportionment degree. The potential of each measure for determining similarity was analyzed in two ways. (1) In an experiment involving 10 artificial Norway spruce (Picea abies (L.) Karst.) stands and two demand matrices for spruce logs, each study stand was harvested using a bucking simulator and the resulting log distributions were then compared with the desired distribution in each of the four measures. (2) The advantages and disadvantages of the measures were analyzed in relation to the requirements for an ideal goodness-of-fit measure. Both analyses suggested that not one of the four measures tested is superior, but that all can be used in actual wood procurement.

Research paper thumbnail of Bayesian estimation of diameter distribution during harvesting

Silva Fennica, 2006

2006. Bayesian estimation of diameter distribution during harvesting. Silva Fennica 40(4): 663-671.

Research paper thumbnail of Design and testing of stand-specific bucking instructions for use on modern cut-to-length harvesters

This study addresses three important issues in tree bucking optimization in the context of cut-to... more This study addresses three important issues in tree bucking optimization in the context of cut-to-length harvesting. (1) Would the fit between the log demand and log output distributions be better if the price and/or demand matrices controlling the bucking decisions on modern cut-to-length harvesters were adjusted to the unique conditions of each individual stand? In what ways can we generate stand and product specific price and demand matrices? (3) What alternatives do we have to measure the fit between the log demand and log output distributions, and what would be an ideal goodness-of-fit measure?

Research paper thumbnail of EMO: A tool for predicting forest composition based on an efficient forest sampling method

Research paper thumbnail of Assessing log geometry and wood quality in standing timber using terrestrial laser-scanning point clouds

Forestry: An International Journal of Forest Research, 2018

Wood procurement in sawmills could be improved by resolving detailed three-dimensional stem geome... more Wood procurement in sawmills could be improved by resolving detailed three-dimensional stem geometry references from standing timber. This could be achieved, using the increasingly available terrestrial point clouds from various sources. Here, we collected terrestrial laser-scanning (TLS) data from 52 Scots pines (Pinus sylvestris L.) with the purpose of evaluating the accuracy of the log geometry and analysing its relationship with wood quality. For reference, the log-specific top-end diameter, volume, tapering, sweep, basic density and knottiness were measured in a sawmill. We produced stem models from the TLS data and bucked them into logs similar to those measured in the sawmill. In comparison to the sawmill data, the log-specific TLSbased top-end diameter, volume, taper and sweep estimates showed relative mean differences of 1.6, −2.4, −3.0 and 78 per cent, respectively. The correlation coefficients between increasing taper and decreasing wood density and whorl-to-whorl distances were 0.49 and −0.51, respectively. Although the stem-model geometry was resolved from the point clouds with similar accuracy to that at the sawmills, the remaining uncertainty in defining the sweep and linking the wood quality with stem geometry may currently limit the method's feasibilities. Instead of static TLS, mobile platforms would likely be more suitable for operational point cloud data acquisition.

Research paper thumbnail of Logging Trail Segmentation via a Novel U-Net Convolutional Neural Network and High-Density Laser Scanning Data

Remote Sensing, 2022

Logging trails are one of the main components of modern forestry. However, spotting the accurate ... more Logging trails are one of the main components of modern forestry. However, spotting the accurate locations of old logging trails through common approaches is challenging and time consuming. This study was established to develop an approach, using cutting-edge deep-learning convolutional neural networks and high-density laser scanning data, to detect logging trails in different stages of commercial thinning, in Southern Finland. We constructed a U-Net architecture, consisting of encoder and decoder paths with several convolutional layers, pooling and non-linear operations. The canopy height model (CHM), digital surface model (DSM), and digital elevation models (DEMs) were derived from the laser scanning data and were used as image datasets for training the model. The labeled dataset for the logging trails was generated from different references as well. Three forest areas were selected to test the efficiency of the algorithm that was developed for detecting logging trails. We designe...

Research paper thumbnail of Factors affecting the output distribution of logs with modern tree bucking optimization system

Аннотация В данной статье рассматривается вопрос о том, что такое самовольное строительство, что ... more Аннотация В данной статье рассматривается вопрос о том, что такое самовольное строительство, что такое объект самовольного строительства, какой правовой статус имеет такая постройка на территории Российской Федерации. Также в статье говорится о том, в соответствии с какими пунктами гражданского кодекса решаются вопросы, связанные с проблемами незаконной постройки. Рассматривается фактическая ситуация в стране по этому вопросу. Ключевые слова: самовольное строительство, самовольная постройка, объект самовольного строительства.

Research paper thumbnail of Implementing SmartForest forest visualization tool on PC environment

Research paper thumbnail of Wood value optimization including logistic costs, production costs and product incomes: a case study in central Finland

Research paper thumbnail of Introduction to tree bucking control

Research paper thumbnail of Applying fuzzy logic to tree bucking control

Forest Science, 2002

In this article, we introduce a fuzzy logic based control system that calibrates the price list c... more In this article, we introduce a fuzzy logic based control system that calibrates the price list controlling the tree bucking of a modern harvester according to the unique conditions of an individual stand prior to harvesting. The system employs a bucking simulator that predicts an output distribution for any price list generated. A fuzzy controller is used to recalculate the price list according to the difference between the demand distribution and the actual output distribution. The tests with perfect knowledge of the stems showed that the control system clearly works efficiently. However, calibrating the price list prior to harvesting seemed to be very sensitive to the precision of the prior information on the trees. The simulations showed that while no gain can be achieved by preharvest inventorying of the stands and fine-tuning the price lists with this imperfect knowledge, the same or even better bucking results can be obtained by the bucking-to-demand process in which the pric...

Research paper thumbnail of Kahden eri katkontamenetelmän vaikutus mäntysahatavaran laatuun

Metsätieteen aikakauskirja, 1970

Research paper thumbnail of Sumea logiikka ja sen soveltaminen hakkuukoneen apteerauksen ohjauksessa

Metsätieteen aikakauskirja, 1970

Research paper thumbnail of Optimisation of wood raw material conversion

Research paper thumbnail of Optimizing logistics of saw logs for a medium sized sawmill

Research paper thumbnail of A genetic algorithm approach to tree bucking optimization

Forest Science, 2004

... the other string, ie, no mixing between various stands was allowed. Several crossover operato... more ... the other string, ie, no mixing between various stands was allowed. Several crossover operators for real-valued representations have been suggested (Beasley et al. 1993b). In this study, a price matrix pair was combined into ...

Research paper thumbnail of Estimating the accuracy and time consumption of a mobile machine vision application in measuring timber stacks

Computers and Electronics in Agriculture, 2019

Research paper thumbnail of Leimikon ennakkomittauksen ajanmenekki

Metsätieteen aikakauskirja, 1970

Leimikon ennakkomittauksella tarkoitetaan kevyttä otantaan perustuvaa pystymittausta, jonka tavoi... more Leimikon ennakkomittauksella tarkoitetaan kevyttä otantaan perustuvaa pystymittausta, jonka tavoitteena on tuottaa ennakkotietoa jalostukseen tulevan leimikon puulajikohtaisista läpimitta-, pituus- ja laatujakaumista. Tutkimuksen tavoitteena oli selvittää leimikkotekijöiden vaikutus leimikon ennakkomittauksen eri työnosien ajanmenekkiin erilaisissa päätehakkuuoloissa sekä edelleen luoda matemaattiset mallit ennakkomittauksen työnosien ajanmenekille kahdella eri menetelmällä. Tutkimuksessa kehitettäviä ajanmenekkifunktioita käytetään ensisijaisesti leimikon ennakkomittausmenetelmien kehittämistyössä.

Research paper thumbnail of A step towards optimal wood supply chain: A case study on optimal tree bucking in Central Finland

In this paper we estimate how large a net profit can be achieved provided that a forest-level tre... more In this paper we estimate how large a net profit can be achieved provided that a forest-level tree bucking optimum can be achieved within a normal Finnish wood supply chain. We demonstrate a holistic system where all costs and revenues within the wood supply chain are calculated. The forest-level optimum of the complete processing chain is then solved with the GA-based algorithm. We calculate both the maximum and minimum net profit for the forest-level bucking optimization problem and find out what could be the maximum difference between the maximum and minimum net profit. The system is demonstrated and validated with the aid of case study material collected from Central Finland. The GAbased stand level allocation system seems to produce logical solutions. There are no clear differences in logistics costs between the best and the worst feasible solution, but there needs to be an explanation of the differences in processing costs and revenues. It seems that there is a possibility to ...

Research paper thumbnail of On the analysis of cubic smoothing spline-based stem curve prediction for forest harvesters

Canadian Journal of Forest Research, 2006

Research paper thumbnail of Comparison of four measures designed for assessing the fit between the demand and output distributions of logs

Canadian Journal of Forest Research, 2005

ABSTRACT In recent years, as customer-oriented production strategies have gained ground, especial... more ABSTRACT In recent years, as customer-oriented production strategies have gained ground, especially in the sawmill industry, the fit between the demand and the actual output for logs of different sizes and qualities has become an important criterion for evaluating bucking outcomes. In this paper, we present four measures for determining the similarity between the demand and output log distributions: (1) the apportionment degree, (2) the χ2 statistic, (3) Laspeyres' quantity index, and (4) the price-weighted apportionment degree. The potential of each measure for determining similarity was analyzed in two ways. (1) In an experiment involving 10 artificial Norway spruce (Picea abies (L.) Karst.) stands and two demand matrices for spruce logs, each study stand was harvested using a bucking simulator and the resulting log distributions were then compared with the desired distribution in each of the four measures. (2) The advantages and disadvantages of the measures were analyzed in relation to the requirements for an ideal goodness-of-fit measure. Both analyses suggested that not one of the four measures tested is superior, but that all can be used in actual wood procurement.

Research paper thumbnail of Bayesian estimation of diameter distribution during harvesting

Silva Fennica, 2006

2006. Bayesian estimation of diameter distribution during harvesting. Silva Fennica 40(4): 663-671.

Research paper thumbnail of Design and testing of stand-specific bucking instructions for use on modern cut-to-length harvesters

This study addresses three important issues in tree bucking optimization in the context of cut-to... more This study addresses three important issues in tree bucking optimization in the context of cut-to-length harvesting. (1) Would the fit between the log demand and log output distributions be better if the price and/or demand matrices controlling the bucking decisions on modern cut-to-length harvesters were adjusted to the unique conditions of each individual stand? In what ways can we generate stand and product specific price and demand matrices? (3) What alternatives do we have to measure the fit between the log demand and log output distributions, and what would be an ideal goodness-of-fit measure?