Leena Matikainen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Leena Matikainen has a total of 71 co-authors. Here are the top co-authors by number of shared publications:
- Harri Kaartinen
Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute FGI, The National Land Survey of Finland, Vuorimiehentie 5, FI-02150 Espoo, Finland
6 shared publications - Josef Taher
Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (FGI), National Land Survey of Finland, Espoo, Finland
6 shared publications - Matti Lehtomäki
Finnish Geodetic Institute, Department of Remote Sensing and Photogrammetry, P.O. BOX 15, 02431 Masala, Finland
5 shared publications - Antero Kukko
Finnish Geospatial Research Institute, National Land Survey of Finland, 02150 Espoo, Finland
4 shared publications - Juha Hyyppä
Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute, Vuorimiehentie 5, 02150 Espoo, Finland
4 shared publications - Matti Hyyppä
Finnish Geospatial Research Institute, National Land Survey of Finland
4 shared publications
Latest Publications
- Open Access
- 12 September 2026
- Published by Elsevier in ISPRS Open Journal of Photogrammetry and Remote Sensing
- Vol. 22
Optimizing bilberry yield models: Comparison of yield estimation strategies and inter-annual transferability
- 12 September 2026
- Published by Elsevier in ISPRS Open Journal of Photogrammetry and Remote Sensing
- Vol. 22
Open AccessOptimizing bilberry yield models: Comparison of yield estimation strategies and inter-annual transferability
- Open Access
- 5 August 2026
- Published by Elsevier in ISPRS Open Journal of Photogrammetry and Remote Sensing
- Vol. 21
Multispectral airborne laser scanning dataset for tree species classification: MS-ALS-SPECIES
- 5 August 2026
- Published by Elsevier in ISPRS Open Journal of Photogrammetry and Remote Sensing
- Vol. 21
Open AccessMultispectral airborne laser scanning dataset for tree species classification: MS-ALS-SPECIES
- Open Access
- 24 July 2026
- Published by Elsevier in Science of Remote Sensing
- Vol. 14
Riverine land cover mapping through semantic segmentation of multispectral point clouds
- 24 July 2026
- Published by Elsevier in Science of Remote Sensing
- Vol. 14
Open AccessRiverine land cover mapping through semantic segmentation of multispectral point clouds
- Open Access
- 7 June 2026
- Published by Elsevier in Trees, Forests and People
- Vol. 26
A novel approach for bilberry yield mapping integrating airborne laser scanning, map databases, and automatically interpreted terrestrial images
- 7 June 2026
- Published by Elsevier in Trees, Forests and People
- Vol. 26
Open AccessA novel approach for bilberry yield mapping integrating airborne laser scanning, map databases, and automatically interpreted terrestrial images
- Open Access
Quantifying Bilberry Counts and Densities: A Comparative Assessment of Segmentation and Object Detection Models from Drone and Camera Imagery
Nordic forest management is increasingly emphasizing multi-functional goals, expanding beyond timber production towards non-wood forest products such as wild berries. Wild berry yield maps are based on sample plot data combined with meteorological, remote sensing, and geoinformation data. Automating sample plot data processing is crucial, as manual collection is labor-intensive, time-consuming, and complicated by short berry seasons and fluctuating yields. This study compares two methods for automatic bilberry detection and counting: a deep learning detector YOLO and a machine learning model using the segment anything model (SAM) followed by a random forest classification (SAM-RF). Both system camera and drone imagery were evaluated as input data. YOLOv8 clearly outperformed SAM–RF in berry detection, achieving an of 0.98 and an RMSE of 3.8 berries when evaluated against annotated system camera images, compared to an of 0.80 for SAM–RF. System camera imagery consistently produced higher accuracy than drone imagery due to higher image clarity and more optimal viewing angles, with YOLOv8 achieving an of 0.95 against field counts, compared to 0.81 for drone images. The results also indicate that the primary error source in berry counting arises from the fact that many berries are not visible in the captured images. The results from the data analysis support the use of the developed technologies in yield modeling and even in implementing future ‘follow-me’ drone berry assistants.Open AccessQuantifying Bilberry Counts and Densities: A Comparative Assessment of Segmentation and Object Detection Models from Drone and Camera Imagery
Nordic forest management is increasingly emphasizing multi-functional goals, expanding beyond timber production towards non-wood forest products such as wild berries. Wild berry yield maps are based on sample plot data combined with meteorological, remote sensing, and geoinformation data. Automating sample plot data processing is crucial, as manual collection is labor-intensive, time-consuming, and complicated by short berry seasons and fluctuating yields. This study compares two methods for automatic bilberry detection and counting: a deep learning detector YOLO and a machine learning model using the segment anything model (SAM) followed by a random forest classification (SAM-RF). Both system camera and drone imagery were evaluated as input data. YOLOv8 clearly outperformed SAM–RF in berry detection, achieving an of 0.98 and an RMSE of 3.8 berries when evaluated against annotated system camera images, compared to an of 0.80 for SAM–RF. System camera imagery consistently produced higher accuracy than drone imagery due to higher image clarity and more optimal viewing angles, with YOLOv8 achieving an of 0.95 against field counts, compared to 0.81 for drone images. The results also indicate that the primary error source in berry counting arises from the fact that many berries are not visible in the captured images. The results from the data analysis support the use of the developed technologies in yield modeling and even in implementing future ‘follow-me’ drone berry assistants. - Top Cited PapersOpen Access
- 27 January 2026
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 233, 278-309
Multispectral airborne laser scanning for tree species classification: A benchmark of machine learning and deep learning algorithms
- 27 January 2026
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 233, 278-309
Top Cited PapersOpen AccessMultispectral airborne laser scanning for tree species classification: A benchmark of machine learning and deep learning algorithms
- Open Access
Predicting Growth of Individual Trees Directly and Indirectly Using 20-Year Bitemporal Airborne Laser Scanning Point Cloud Data
Reviewing forest carbon sinks is of the utmost importance in efforts to control climate change. This study focuses on reporting the 20-year boreal forest growth values acquired with airborne laser scanning (ALS). The growth was examined on the Kalkkinen research site in southern Finland as a continuation of several earlier growth studies performed in the same area. The data for the study were gathered with three totally different airborne laser scanning systems, namely using Toposys-I Falcon in June 2000 and Riegl VUX-1HA and miniVUX-3UAV in June 2021 with approximate point densities of 11, 1360, and 460 points/m2, respectively. The ALS point cloud was preprocessed to identify individual trees, from each of which different features were extracted either for direct or indirect growth measurement. In the direct method, the growth value is predicted based on differences of features, whereas in the indirect method, the growth value is obtained by subtracting the results of two independent predictions of different years. The growth in individual tree attributes, such as growth in height, diameter at breast height (DBH), and stem volume, were calculated for direct estimation. Field reference campaigns were performed in the summer of 2001 and in November 2021 to validate the obtained growth values. The study showed that long-term series growth of height, DBH, and stem volume are possible to record with a high-to-moderate coefficient of determination () of 0.90, 0.48, and 0.45 in the best-case scenarios. The respective root-mean-squared errors (RMSE) values were 0.98 m, 0.02 m, and 0.17 m3, and the biases were −0.06 m, 0.00 m, and 0.17 m3. The direct method produced better metrics in terms of RMSE-% and bias, but the indirect method produced better best-fit lines. Additionally, the mean growth values for height, diameter, and stem volume intervals were compared, and they are presumed to be usable even for forest modelling.Open AccessPredicting Growth of Individual Trees Directly and Indirectly Using 20-Year Bitemporal Airborne Laser Scanning Point Cloud Data
Reviewing forest carbon sinks is of the utmost importance in efforts to control climate change. This study focuses on reporting the 20-year boreal forest growth values acquired with airborne laser scanning (ALS). The growth was examined on the Kalkkinen research site in southern Finland as a continuation of several earlier growth studies performed in the same area. The data for the study were gathered with three totally different airborne laser scanning systems, namely using Toposys-I Falcon in June 2000 and Riegl VUX-1HA and miniVUX-3UAV in June 2021 with approximate point densities of 11, 1360, and 460 points/m2, respectively. The ALS point cloud was preprocessed to identify individual trees, from each of which different features were extracted either for direct or indirect growth measurement. In the direct method, the growth value is predicted based on differences of features, whereas in the indirect method, the growth value is obtained by subtracting the results of two independent predictions of different years. The growth in individual tree attributes, such as growth in height, diameter at breast height (DBH), and stem volume, were calculated for direct estimation. Field reference campaigns were performed in the summer of 2001 and in November 2021 to validate the obtained growth values. The study showed that long-term series growth of height, DBH, and stem volume are possible to record with a high-to-moderate coefficient of determination () of 0.90, 0.48, and 0.45 in the best-case scenarios. The respective root-mean-squared errors (RMSE) values were 0.98 m, 0.02 m, and 0.17 m3, and the biases were −0.06 m, 0.00 m, and 0.17 m3. The direct method produced better metrics in terms of RMSE-% and bias, but the indirect method produced better best-fit lines. Additionally, the mean growth values for height, diameter, and stem volume intervals were compared, and they are presumed to be usable even for forest modelling. - Open Access
- 1 October 2019
- Published by SPIE-Intl Soc Optical Eng in Journal of Applied Remote Sensing
- Vol. 13 (04) , 044504-35
Toward utilizing multitemporal multispectral airborne laser scanning, Sentinel-2, and mobile laser scanning in map updating
Nationwide topographic databases presenting, for example, buildings, roads, water areas, ground elevation, and some land use information provide the basic material for map production and numerous applications exploiting geospatial data. Currently, there are increasing demands for up-to-date and detailed data. Keeping the databases up to date and further developing their contents and maintenance processes thus becomes more and more important (see, e.g., Refs. 1 and 2). In Finland, as in many other countries, aerial images covering the entire country are acquired regularly at a few years’ intervals and used as the main remotely sensed data source in map updating. Data from single-channel airborne laser scanning (ALS) are also used, primarily to create high-quality digital terrain models (DTMs). The first scanning of the entire country with a minimum density of is planned to be finished in 2019. Multitemporal ALS datasets are not yet commonly available, and the intensity information of the ALS is not generally in active use. Some objects in the National Land Survey of Finland (NLS) Topographic Database3 are currently updated continuously (e.g., roads) or yearly (e.g., buildings) using information obtained, for example, from municipalities. More extensive updating of the database is done at a few years’ intervals following aerial image acquisitions. This updating process is mainly based on visual interpretation of the images.- 1 October 2019
- Published by SPIE-Intl Soc Optical Eng in Journal of Applied Remote Sensing
- Vol. 13 (04) , 044504-35
Open AccessToward utilizing multitemporal multispectral airborne laser scanning, Sentinel-2, and mobile laser scanning in map updating
Nationwide topographic databases presenting, for example, buildings, roads, water areas, ground elevation, and some land use information provide the basic material for map production and numerous applications exploiting geospatial data. Currently, there are increasing demands for up-to-date and detailed data. Keeping the databases up to date and further developing their contents and maintenance processes thus becomes more and more important (see, e.g., Refs. 1 and 2). In Finland, as in many other countries, aerial images covering the entire country are acquired regularly at a few years’ intervals and used as the main remotely sensed data source in map updating. Data from single-channel airborne laser scanning (ALS) are also used, primarily to create high-quality digital terrain models (DTMs). The first scanning of the entire country with a minimum density of is planned to be finished in 2019. Multitemporal ALS datasets are not yet commonly available, and the intensity information of the ALS is not generally in active use. Some objects in the National Land Survey of Finland (NLS) Topographic Database3 are currently updated continuously (e.g., roads) or yearly (e.g., buildings) using information obtained, for example, from municipalities. More extensive updating of the database is done at a few years’ intervals following aerial image acquisitions. This updating process is mainly based on visual interpretation of the images.