Article
Agronomy
Xin Yang, Shichen Gao, Qian Sun, Xiaohe Gu, Tianen Chen, Jingping Zhou, Yuchun Pan
Summary: This study proposes a method for classifying the extent of maize lodging based on deep learning algorithms and UAV images. By analyzing the characteristic variation in RGB and multispectral images, training and comparison of VGG-16, Inception-V3, and ResNet-50 algorithms were conducted. The results show that multispectral images outperform RGB images in classifying different lodging extents.
Article
Agronomy
Sailesh Sigdel, Amitava Chatterjee, Marisol Berti, Abbey Wick, Caley Gasch
Summary: The study indicates that interseeding cover crops in sugar beet production systems can protect the soil from erosion without negatively impacting sugar beet yield or quality. Cover crop biomass accumulation varied depending on rainfall distribution, with early interseeding producing more biomass than late interseeding. Different cover crop species and planting times affected the growth and cover provided by the cover crops.
FIELD CROPS RESEARCH
(2021)
Article
Plant Sciences
Abel Barreto, Facundo Ramon Ispizua Yamati, Mark Varrelmann, Stefan Paulus, Anne-Katrin Mahlein
Summary: A pipeline based on machine learning methods was established for image data analysis and extraction of disease-relevant parameters from UAV data, providing a more precise and nondestructive assessment of plant diseases.
Article
Soil Science
Olga Fishkis, Heinz-Josef Koch
Summary: This study investigated the environmental risks of mechanical weed control as an alternative method in row crops in Europe for the first time. The results showed that mechanical weed control did not affect the abundance of earthworms but significantly reduced runoff and soil erosion. It was more effective in crusted soils and had no effect in non-crusted soils.
SOIL & TILLAGE RESEARCH
(2023)
Article
Computer Science, Artificial Intelligence
Afshin Azizi, Zhao Zhang, Zhaoyu Rui, Yunxi Li, C. Igathinathane, Paulo Flores, Jithin Mathew, Alireza Pourreza, Xiongzhe Han, Man Zhang
Summary: This study proposes a multi-task approach involving aerial images to detect and classify wheat lodging. By combining convolutional neural networks and temporal sequences in a single model, spatiotemporal information from wheat image datasets is utilized. The proposed models achieve more accurate detection despite the limited and imbalanced dataset. The developed methodology paves the way for comprehensive and automatic wheat lodging detection and can be adapted for similar crops.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Interdisciplinary Applications
M. Janani, R. Jebakumar
Summary: In agriculture, it is important to identify nutrient deficiencies or excess in plant leaves to improve crop yield. Existing detection mechanisms for this purpose may be inefficient and expensive. The Nutrient Range Analysis Based on Greenness (NRAG) system is proposed, which uses image processing techniques and a Convolution Neural Network (CNN) based Holding Vector Network (HVN) model to accurately detect and classify nitrogen nutrient levels in groundnut leaves. The developed model achieves high accuracy and can be adopted by farmers in real time.
ADVANCES IN ENGINEERING SOFTWARE
(2023)
Article
Multidisciplinary Sciences
Masanori Onishi, Takeshi Ise
Summary: The study developed a machine vision system for tree identification and mapping using RGB images captured by a UAV and CNN. The system successfully classified seven tree classes with over 90% accuracy, and the Guided Grad-CAM analysis showed that the CNN classified trees based on their shapes and leaf contrasts, enhancing the system's potential for cost-effective classification of individual trees with similar colors.
SCIENTIFIC REPORTS
(2021)
Article
Chemistry, Multidisciplinary
Florin Dumitrescu, Costin-Anton Boiangiu, Mihai-Lucian Voncila
Summary: This paper proposes a fast and reliable object detection algorithm that improves the accuracy by adding a region of interest classification and regression branch. It can be used in an image-based people tracking system with higher inference speed.
APPLIED SCIENCES-BASEL
(2022)
Article
Agriculture, Multidisciplinary
P. E. N. G. Huan, L. I. U. Hui, G. A. O. Li, J. I. A. N. G. Ru, L. Guang-kuo, G. A. O. Hai-feng, W. U. Wei, W. A. N. G. Jun, Z. H. A. N. G. Yu, H. U. A. N. G. Wen-kun, K. O. N. G. Ling-an, P. E. N. G. De-liang
Summary: The sugar beet cyst nematode population collected from sugar beet fields in Xinjiang was confirmed to be H. schachtii based on morphological and molecular characterization.
JOURNAL OF INTEGRATIVE AGRICULTURE
(2022)
Article
Multidisciplinary Sciences
Sheng Zhang, Guang Lin, Samy Tindel
Summary: This study introduces a proper notion of two-dimensional signature for images and demonstrates its excellent accuracy in texture classification. The low-dimensional feature set based on signatures captures essential features of two-dimensional objects such as images.
PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES
(2022)
Article
Multidisciplinary Sciences
Meng Xia, Meenal K. Kheterpal, Samantha C. Wong, Christine Park, William Ratliff, Lawrence Carin, Ricardo Henao
Summary: In this paper, a machine-learning-based approach for lesion identification and malignancy prediction from clinical dermatological images is proposed. The method can handle images acquired via smartphone or dermoscopy capture, and supports both focal and wide-field images. The proposed two-stage approach shows superior performance compared to alternative model architectures. The combination of images and clinical data does not significantly improve over the image-only model. The proposed framework offers comparable performance in terms of malignancy classification to board certified dermatologists.
SCIENTIFIC REPORTS
(2022)
Article
Agronomy
Yunling Liu, Guoli Zhang, Ke Shao, Shunfu Xiao, Qing Wang, Jinyu Zhu, Ruili Wang, Lei Meng, Yuntao Ma
Summary: This paper proposes a method to segment individual leaves of sugar beet plants by reconstructing 3D point clouds and using the multiscale tensor voting method and surface boundary filter. The results show that this method can effectively segment the leaves of field grown sugar beet plants and calculate leaf length and area.
Article
Agronomy
Mario Serouart, Simon Madec, Etienne David, Kaaviya Velumani, Raul Lopez Lozano, Marie Weiss, Frederic Baret
Summary: This study presents a method for pixel segmentation of high-resolution RGB images into vegetation classes. The method achieves accurate segmentation of the images into background, green, and senescent vegetation classes. However, some confusion is observed between the background and senescent vegetation, especially in dark and bright regions. The study also finds that the SVM method provides more precise delineation of the green and senescent patches compared to the convolutional nature of U-net.
Article
Environmental Sciences
Ehsan Khoramshahi, Roope Naesi, Stefan Rua, Raquel A. A. Oliveira, Axel Paivansalo, Oiva Niemelainen, Markku Niskanen, Eija Honkavaara
Summary: This article explores the use of drone techniques to identify alien barleys in oat fields. By employing a machine learning approach and drone images, the study successfully detects and localizes barley plants, providing a useful method for modern grain production industries.
Article
Multidisciplinary Sciences
Yung-Jhe Yan, Weng-Keong Wong, Chih-Jung Chen, Chi-Cho Huang, Jen-Tzung Chien, Mang Ou-Yang
Summary: This study proposes a method to extract signature bands from deep learning models of multispectral data converted from hyperspectral data, and applies these bands to predict the sugar content of Syzygium Samarangense. The results show that using only six signature bands can achieve better prediction accuracy than using more bands.
SCIENTIFIC REPORTS
(2023)
Article
Zoology
Alan Storelli, Andreas Keiser, Sebastian Kiewnick, Matthias Daub, Anne-Katrin Mahlein, Werner Beyer, Mario Schumann
Summary: A new screening mechanism for soil inoculation of Ditylenchus dipsaci in sugar beet was established, determining optimal inoculation time points, inoculum levels, and plant positioning. The study showed that soil inoculation resulted in higher penetration rates compared to leaf axil inoculation, with a recommended inoculum level of 1000 nematodes per pot at plant emergence. Rearing the nematodes for 35 days on carrot discs produced an infective inoculum containing up to 50% eggs, aiding in breeding for resistance against D. dipsaci.
Article
Agronomy
Maximilian M. Muellender, Anne-Katrin Mahlein, Gerd Stammler, Mark Varrelmann
Summary: This study identified mutations associated with reduced sensitivity in C. beticola isolates collected from different European countries. The experiment showed that some mutations were related to lower EC50 values, indicating a possible correlation between target site mutations and reduced sensitivity.
PEST MANAGEMENT SCIENCE
(2021)
Article
Microbiology
Elias Alisaac, Anna Rathgeb, Petr Karlovsky, Anne-Katrin Mahlein
Summary: The study found that F. graminearum grows downward within infected wheat spikes and DON accumulation is largely confined to the colonized tissue. Additionally, F. graminearum was able to infect wheat kernels and cause mycotoxin contamination even when inoculated 25 days after anthesis.
Article
Plant Sciences
David Bohnenkamp, Jan Behmann, Stefan Paulus, Ulrike Steiner, Anne-Katrin Mahlein
Summary: This study established a hyperspectral library of wheat foliar diseases, detected important turning points using spectral changes, and achieved high accuracy in disease detection and identification through machine learning methods.
Article
Agronomy
Alan Storelli, Sebastian Kiewnick, Matthias Daub, Anne-Katrin Mahlein, Mario Schumann, Werner Beyer, Andreas Keiser
Summary: In European sugar beet production, different populations of Ditylenchus dipsaci show varying levels of damage, with the Seeland population having the highest reproduction rate on sugar beets. The reproduction rate of D. dipsaci at 60 dpi is negatively correlated with the fresh biomass of sugar beets. These findings can guide breeding programs for sugar beets resistance.
EUROPEAN JOURNAL OF PLANT PATHOLOGY
(2021)
Article
Agriculture, Multidisciplinary
Alan Storelli, Alexandra Minder, Andreas Keiser, Sebastian Kiewnick, Matthias Daub, Anne-Katrin Mahlein, Mario Schumann, Werner Beyer
Summary: This study conducted three consecutive in vivo bioassays to find potentially resistant sugar beet lines restricting reproduction and penetration of D. dipsaci. Although no resistance was found among the genotypes, a high variation of the penetration rate by D. dipsaci was observed. The study provides a basis for the development of resistant sugar beet cultivars to D. dipsaci, but further large-scale screenings are needed to confirm the observed variations among genotypes.
JOURNAL OF PLANT DISEASES AND PROTECTION
(2021)
Article
Agronomy
Nelia Nause, Joern Strassemeyer, Anne-Katrin Mahlein, Nicol Stockfisch
Summary: The analysis of pesticide use in sugar beet cultivation in Germany from 2010 to 2015 showed that environmental risks were relatively low and can be further reduced by combining field-specific measures and technical options. Certain combinations of active ingredients, application dates, and field-specific environmental conditions were found to provoke higher risks, especially with herbicide use.
PEST MANAGEMENT SCIENCE
(2021)
Article
Agronomy
Anna Brugger, Patrick Schramowski, Stefan Paulus, Ulrike Steiner, Kristian Kersting, Anne-Katrin Mahlein
Summary: Recent studies have expanded the use of hyperspectral sensors to the ultraviolet range for monitoring plant stress processes. This study investigated plant-pathogen interactions in barley using hyperspectral imaging, revealing the potential for differentiation between barley genotypes inoculated with powdery mildew in the UV range. Deep learning classification showed high performance in distinguishing inoculated and noninoculated samples.
Article
Environmental Sciences
Helen Thompson, Sarah Vaughan, Anne-Katrin Mahlein, Erwin Ladewig, Christine Kenter
Summary: This study analyzed the residues of neonicotinoids in soil and succeeding crops, finding that residue levels decreased over time and with lower application frequency. Residues in pollen and nectar were detected to be lower than reported adverse effect concentrations in studies with honeybee and bumble bee individuals, indicating low risk to pollinators.
INTEGRATED ENVIRONMENTAL ASSESSMENT AND MANAGEMENT
(2022)
Article
Agriculture, Multidisciplinary
Stefan Thomas, Jan Behmann, Uwe Rascher, Anne-Katrin Mahlein
Summary: Previous studies on the performance of transmission and reflection datasets for disease detection yielded inconsistent results, with reflection imaging showing superior detection capabilities compared to transmission imaging. The disparity in results may be linked to the different interactions between the pathogens and host plants, as well as the way light interacts with plant tissue. The study offers new insights into the nature of transmission-based hyperspectral imaging and its application range.
JOURNAL OF PLANT DISEASES AND PROTECTION
(2022)
Editorial Material
Agriculture, Multidisciplinary
Anne-Katrin Mahlein, Rene Hans-Juergen Heim, Anna Brugger, Kaitlin Gold, Yang Li, Ali Kashif Bashir, Stefan Paulus, Matheus Thomas Kuska
JOURNAL OF PLANT DISEASES AND PROTECTION
(2022)
Article
Plant Sciences
Abel Barreto, Facundo Ramon Ispizua Yamati, Mark Varrelmann, Stefan Paulus, Anne-Katrin Mahlein
Summary: A pipeline based on machine learning methods was established for image data analysis and extraction of disease-relevant parameters from UAV data, providing a more precise and nondestructive assessment of plant diseases.
Article
Plant Sciences
Anna Brugger, Facundo Ispizua Yamati, Abel Barreto, Stefan Paulus, Patrick Schramowsk, Kristian Kersting, Ulrike Steiner, Susanne Neugart, Anne-Katrin Mahlein
Summary: Fungal infections trigger changes in plant metabolites, which can be detected using destructive or nondestructive methods. This study compared the effects of two diseases, Cercospora leaf spot disease (CLS) and sugar beet rust (BR), on plant metabolites using both destructive analyses and nondestructive hyperspectral measurements in the UV range. The results showed distinct reflectance patterns and spectral changes in response to the two diseases, allowing for differentiation and recognition using machine learning algorithms. The study highlights the utility of nondestructive UV-range hyperspectral imaging in investigating plant diseases.
Article
Biology
Maurice Guender, Facundo R. Ispizua Yamati, Jana Kierdorf, Ribana Roscher, Anne-Katrin Mahlein, Christian Bauckhage
Summary: In this work, a hands-on workflow for the automatized temporal and spatial identification and individualization of crop images from UAVs is presented, improving the analysis and interpretation of UAV data in agriculture significantly. The results show that the approach has similar accuracy to more complex deep learning-based recognition techniques and can automate the processing of large datasets.
Review
Plant Sciences
Clive H. Bock, Sarah J. Pethybridge, Jayme G. A. Barbedo, Paul D. Esker, Anne-Katrin Mahlein, Emerson M. Del Ponte
Summary: Phytopathometry is a critical branch of plant pathology that focuses on estimating and measuring the amount of plant disease, playing a crucial role in analyzing yield loss, breeding disease-resistant plants, evaluating disease control methods, understanding pathogen ecology, and more. It is essential for a unified cross-discipline approach to research and application of tools in phytopathometry.
TROPICAL PLANT PATHOLOGY
(2022)