Article
Biodiversity Conservation
Yevgeniya Korol, Watit Khokthong, Delphine C. Zemp, Bambang Irawan, Holger Kreft, Dirk Hoelscher
Summary: In tropical landscapes dominated by oil palm monocultures, scattered trees were found to be abundant, mostly small-statured, suggesting that most of the trees are young and do not reach larger dimensions.
GLOBAL ECOLOGY AND CONSERVATION
(2021)
Article
Geography, Physical
Juepeng Zheng, Haohuan Fu, Weijia Li, Wenzhao Wu, Le Yu, Shuai Yuan, Wai Yuk William Tao, Tan Kian Pang, Kasturi Devi Kanniah
Summary: The rapid expansion of oil palm plantations in tropical developing countries has both positive economic benefit and negative ecological impact. Accurate detection of oil palm trees is crucial for improving plantation planning, yield, and reducing manpower and fertilizer consumption. The use of Unmanned Aerial Vehicles (UAVs) shows promise for monitoring individual oil palms, but challenges remain in achieving accuracy due to class imbalance and similarity. The Multi-class Oil Palm Detection approach (MOPAD) proposed in this paper utilizes advanced features and loss modules to achieve accurate detection and monitoring of individual oil palms, demonstrating excellent potential for efficient management.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Agronomy
Emmanuelle Lamade, Guillaume Tcherkez
Summary: Oil palm, a potassium-demanding tree, requires fertilisation throughout its life cycle. The commonly used index for monitoring potassium fertilization is potassium content in leaflets (%K), but uncertainty remains regarding the most responsive organs and the need for separate response curves due to different conditions. Our study investigated potassium responses of oil palm trees in four different progenies, identifying phosphorus and %K in rachis as potential indicators of yield. By combining mineral contents using multivariate analysis, we successfully predicted fresh fruit bunch yield independent of progeny.
EUROPEAN JOURNAL OF AGRONOMY
(2023)
Article
Environmental Sciences
Rudraksh Kapil, Guillermo Castilla, Seyed Mojtaba Marvasti-Zadeh, Devin Goodsman, Nadir Erbilgin, Nilanjan Ray
Summary: Operational forest monitoring often requires detailed information in the form of an orthomosaic created by stitching overlapping aerial images. However, using only RGB drone sensors for imaging results in reduced contrast and descriptive features. To address this, we propose a thermal orthomosaicking workflow that combines RGB and thermal images for higher quality and better alignment. Our technique achieves improved mutual information and preserves radiometric information from the original thermal imagery, making it suitable for downstream tasks such as tree crown detection. We provide an open-source tool to facilitate usage and further development.
Article
Geochemistry & Geophysics
Juepeng Zheng, Wenzhao Wu, Shuai Yuan, Haohuan Fu, Weijia Li, Le Yu
Summary: Providing accurate and timely oil palm information is crucial for economic development and ecological significance. However, large-scale and cross-regional oil palm tree detection is challenging due to the variety and volume of data, as well as environmental heterogeneity. This study proposes a new multisource domain generalization method that achieves promising performance in unknown target domains.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Agriculture, Multidisciplinary
Thani Jintasuttisak, Eran Edirisinghe, Ali Elbattay
Summary: This study uses the YOLO-V5 convolutional neural network to detect date palm trees in images captured by a drone. The results show that the YOLO-V5m model achieves high accuracy and has the ability to detect and localize date palm trees of different sizes in complex and sparse environments.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Computer Science, Artificial Intelligence
Hery Wibowo, Imas Sukaesih Sitanggang, Mushthofa Mushthofa, Hari Agung Adrianto
Summary: This research uses deep learning models YOLOv3, YOLOv4, and YOLOv5m to detect oil palm trees, achieving high accuracy and efficiency. The method has great potential for commercial applications.
BIG DATA AND COGNITIVE COMPUTING
(2022)
Article
Remote Sensing
Tai Yang Lim, Jiyun Kim, Wheemoon Kim, Wonkyong Song
Summary: Wetlands have significant ecological value and play a crucial role in the environment. Recent advancements in remote exploration technology have enabled a quantitative analysis of wetlands through surveys on the type of cover present. This study establishes an effective method for classifying centimeter-scale images using multispectral and hyperspectral techniques.
Article
Forestry
Laura Somenguem Donfack, Alexander Roell, Florian Ellsaesser, Martin Ehbrecht, Bambang Irawan, Dirk Hoelscher, Alexander Knohl, Holger Kreft, Eduard J. Siahaan, Leti Sundawati, Christian Stiegler, Delphine Clara Zemp
Summary: Agroforestry options, such as mixed-species tree planting and natural regeneration, may help alleviate the negative impact of forest loss on biodiversity and ecosystem functioning in oil palm plantations. The study found that tree species diversity did not have a significant impact on microclimate and land surface temperatures (LST), but humidity was higher in planted tree islands compared to areas with natural regeneration.
FOREST ECOLOGY AND MANAGEMENT
(2021)
Article
Environmental Sciences
Gil Goncalves, Umberto Andriolo
Summary: The use of UAS for marine litter surveys has been proven to be feasible. Multispectral images can be used to classify litter type, and automated detection methods show a reasonable categorization and density assessment. This method can improve litter surveys in the environment.
MARINE POLLUTION BULLETIN
(2022)
Article
Automation & Control Systems
Shi-Jinn Horng, Dinh-Trung Vu, Thi-Van Nguyen, Wanlei Zhou, Chin-Teng Lin
Summary: This article proposes a new low-cost palm vein recognition system for smartphones using RGB images, enhancing accuracy by using the saturation channel and introducing an improved method for region of interest extraction. A lightweight deep learning-based model is also designed, achieving better accuracy with fusion strategy.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Agronomy
Cheah See Siang, Siti Aishah Abd Wahid, Christopher Teh Boon Sung
Summary: This study aimed to investigate the dry-matter production and nutrient demand of tenera oil palm at different ages. Through sampling and measuring the dry weights and nutrient contents of different plant parts, we obtained the nutrient requirements and dry-matter production of oil palm at different stages of growth.
Article
Green & Sustainable Science & Technology
Jake E. Bicknell, Jesse R. O'Hanley, Paul R. Armsworth, Eleanor M. Slade, Nicolas J. Deere, Simon L. Mitchell, David Hemprich-Bennett, Victoria Kemp, Stephen J. Rossiter, Owen T. Lewis, David A. Coomes, Agnes L. Agama, Glen Reynolds, Matthew J. Struebig, Zoe G. Davies
Summary: Agricultural expansion is the main factor leading to ecological degradation in the tropics. This study shows that targeted set-asides, especially alongside rivers, can increase biodiversity and ecosystem services without reducing the net cultivated area.
NATURE SUSTAINABILITY
(2023)
Article
Environmental Sciences
Andres C. Rodriguez, Stefano D'Aronco, Konrad Schindler, Jan D. Wegner
Summary: The study proposes a new active deep learning method to estimate oil palm density from Sentinel-2 satellite images at large scale, generating complete maps for Malaysia and Indonesia. The method was used to compute oil palm density maps for 2017 and 2019, analyzing density variations between different states and comparing them to official estimates.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Forestry
Jean-Baptiste Ndamiyehe Ncutirakiza, Sylvie Gourlet-Fleury, Philippe Lejeune, Xavier Bry, Catherine Trottier, Frederic Mortier, Adeline Fayolle, Francois Muhashy Habiyaremye, Leopold Ndjele Mianda-Bungi, Gauthier Ligot
Summary: This study examines the influence of canopy structure on tropical tree growth using data collected through unmanned aerial vehicles (UAVs) and field measurements. The results show that combining UAV and field data can improve the prediction of tree diameter increment. Diameter at breast height and crown area are complementary predictors, and crown-based competition indices significantly enhance prediction models. The calibrated model at one site can accurately predict growth at another site.
FOREST ECOLOGY AND MANAGEMENT
(2024)