Article
Engineering, Chemical
Zhicheng Liu, Long Wang, Zhiyuan Liu, Xufeng Wang, Can Hu, Jianfei Xing
Summary: This paper proposes an image-detection method for cotton seed damage based on an improved YOLOv5 algorithm. Experimental results show that the algorithm achieves high accuracy and recall rate in detecting appearance-based damage of cotton seeds.
Article
Computer Science, Artificial Intelligence
Dongfang Li, Boliao Li, Shuo Kang, Huaiqu Feng, Sifang Long, Jun Wang
Summary: Crop row detection is crucial for visual navigation of agricultural machinery. In this study, a compact and efficient deep learning-based network named E2CropDet is proposed, which models each crop row as an independent object, enabling an end-to-end detection process with no post-processing. The network utilizes generic feature extractors and line-shaped proposals for detection, achieving remarkable results and a detection speed of 166 frames per second.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Optics
Tianyu Wang, Mandar M. Sohoni, Logan G. Wright, Martin M. Stein, Shi-Yuan Ma, Tatsuhiro Onodera, Maxwell G. Anderson, Peter L. McMahon
Summary: A nonlinear optical neural network image sensor based on an image intensifier enables efficient all-optical image encoding for various machine-vision tasks. The nonlinear ONN encoder outperforms linear optical encoders in machine-vision benchmarks, flow-cytometry image classification, and object identification in a 3D printed real scene. This concept allows for a significant reduction in camera resolution and electronic post-processing complexity, and enables image-sensing applications with fewer pixels, photons, higher throughput, and lower latency.
Article
Agronomy
Sheng Jiang, Ziyi Liu, Jiajun Hua, Zhenyu Zhang, Shuai Zhao, Fangnan Xie, Jiangbo Ao, Yechen Wei, Jingye Lu, Zhen Li, Shilei Lyu
Summary: This study introduces a real-time detection and maturity classification method for loofah, which includes a one-stage instance segmentation model called LuffaInst and a machine learning-based maturity classification model. Experimental results show that LuffaInst has lower parameter weights and higher accuracy than other prevalent instance segmentation models. A random forest model relying on color and texture features is also developed for three maturity classifications of loofah fruit instances. The research results have important implications for loofah fruit maturity detection.
Article
Agriculture, Multidisciplinary
Mailson Freire de Oliveira, Adao Felipe dos Santos, Elizabeth Haruna Kazama, Glauco de Souza Rolim, Rouverson Pereira da Silva
Summary: Optimizing the application of phytosanitary products through the use of artificial intelligence and remote sensing techniques can lead to more sustainable agricultural practices. By utilizing remote sensing data and artificial neural networks (MLP), it is possible to estimate coffee tree volume accurately and apply pesticides at a variable rate.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Chemistry, Analytical
Xuelin Zhang, Donghao Zhang, Alexander Leye, Adrian Scott, Luke Visser, Zongyuan Ge, Paul Bonnington
Summary: This paper focuses on improving the performance of scientific instrumentation that uses glass spray chambers for sample introduction, by detecting incidents using deep convolutional models. The indicators of poor quality sample introduction include the formation of liquid beads and flooding in the spray chamber. The proposed frameworks for detecting these incidents leverage modern deep learning architectures and expert knowledge, achieving high accuracy and real-time implementation.
Article
Engineering, Electrical & Electronic
Du-Ming Tsai, Shu-Kai S. Fan, Yi-Hsiang Chou
Summary: This study proposes a deep learning scheme for automatic defect detection in material surfaces, using CycleGAN to automatically generate defect annotations without manual work, resulting in high accuracy and efficiency in training.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Review
Engineering, Marine
Hanchi Liu, Xin Ma, Yining Yu, Liang Wang, Lin Hao
Summary: Automated monitoring and analysis of fish's growth status and behaviors using machine vision and deep learning techniques have become important in aquaculture management. DL-based object detection techniques have been extensively applied to classify and locate fish of interest in images.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Chemistry, Analytical
Yuhang Sun, Mengxuan Li, Ruiwen Dong, Weiyu Chen, Dong Jiang
Summary: A bolt loosening detection method based on YOLOv5 is proposed, achieving high accuracy and robustness in detecting the bolt loosening by identifying the rotation of the nut. The experimental results demonstrate the effectiveness of the method in detecting the loosening angle of the bolted connection, especially for tiny angles of loosening.
Article
Agriculture, Multidisciplinary
RajinderKumar M. Math, Nagaraj Dharwadkar
Summary: Crop protection aims to develop an agriculture system resilient to threats that cause sub-optimal crop growth. This research focuses on developing a deep convolutional neural network model to accurately identify and classify grape diseases based on RGB leaf images. The model achieved an accuracy of 99.34% and demonstrated high precision, recall, and F1 score.
JOURNAL OF PLANT DISEASES AND PROTECTION
(2022)
Article
Food Science & Technology
Tao Li, Jinjie Tong, Muhua Liu, Mingyin Yao, Zhifeng Xiao, Chengjie Li
Summary: In this study, an automatic approach based on machine vision technology is proposed for corn image acquisition, impurity classification and recognition, and impurities content detection. The MSRCR algorithm is used to enhance the image, HSV color space parameter threshold is set for image segmentation, and a comprehensive evaluation index is adopted for quantitatively evaluating the test results. The online detection results show the effectiveness of the proposed algorithm in identifying impurities in corn images and monitoring impurities content in the corn deep-bed drying process.
Article
Computer Science, Information Systems
Ronghui Zhang, Xiaojun Jing, Sheng Wu, Chunxiao Jiang, Junsheng Mu, F. Richard Yu
Summary: The article clarifies the motivation and mechanism of DL-aided WSSs for human detection, exploring the application of DL in WSS and future research directions, emphasizing the potential of combining information from different sensors to enhance the overall performance of practical human detection systems.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Ecology
Juergen Niedballa, Jan Axtner, Timm Fabian Doebert, Andrew Tilker, An Nguyen, Seth T. Wong, Christian Fiderer, Marco Heurich, Andreas Wilting
Summary: This paper introduces an R package imageseg for image segmentation using convolutional neural networks, which can be used to evaluate forest structural metrics with high accuracy and generalization ability across different forest types and biomes.
METHODS IN ECOLOGY AND EVOLUTION
(2022)
Article
Biochemical Research Methods
Xihuizi Liang, Bingqi Chen, Chaojie Wei, Xiongchu Zhang
Summary: This study proposed an algorithm that combined edge detection and OTSU to effectively extract navigation lines for cotton crops grown in wide and narrow rows. The accuracy of route detection reached 99.2%, 98.1%, and 98.4% for cotton, corn, and soybean at the seedling stage, respectively. The algorithm can adapt to different shadow interference and the randomness of crop row growth.
Review
Chemistry, Multidisciplinary
Zhe Chen, Jisheng Lu, Haiyan Wang
Summary: Analysis of pig posture is important for improving welfare and yield in different conditions. Detection of postures can help assess psychological and physiological conditions, predict abnormal behavior, and evaluate farming conditions. Deep learning-based methods show superiority in performance and feasibility compared to traditional methods.
APPLIED SCIENCES-BASEL
(2023)
Article
Agriculture, Multidisciplinary
Md Sultan Mahmud, Azlan Zahid, Long He, Daeun Choi, Grzegorz Krawczyk, Heping Zhu, Paul Heinemann
Summary: An unmanned ground-based canopy density measurement system was developed to support precision spraying in apple orchards. The system utilized a LiDAR sensor to measure tree canopy density and generate a density map. Field evaluation results showed that the system could help reduce excessive pesticide use in orchards by accurately measuring canopy density.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Plant Sciences
Phillip L. Martin, Teresa Krawczyk, Kristen Pierce, Catherine Thomas, Fatemeh Khodadadi, Srdan G. Acimovic, Kari A. Peter
Summary: This study aimed to investigate the causes of increased bitter rot of apple reported by apple growers in the Mid-Atlantic region of the U.S.A. The results showed that fungicide resistance to commonly used single-mode-of-action (single-MoA) fungicides was unlikely to be the main cause of the increase. The study provides guidelines for selecting appropriate single-MoA fungicides for controlling bitter rot in the region.
Article
Plant Sciences
Fatemeh Khodadadi, Phillip L. Martin, Daniel J. Donahue, Kari A. Peter, Srdan G. Acimovic
Summary: Apple orchards in the Mid-Atlantic region of the United States have experienced severe premature defoliation due to apple blotch disease caused by Diplocarpon coronariae. The study investigated the spore dispersal patterns, fungicide efficacy, and host defense-related gene expression to develop effective management practices for apple blotch disease.
Article
Horticulture
Azlan Zahid, Md Sultan Mahmud, Long He, James Schupp, Daeun Choi, Paul Heinemann
Summary: In this study, the cutting torque required for pruning apple trees was measured and it was found that the branch diameter is the most important factor influencing the torque requirement. The results of this study are important for selecting appropriate cutting mechanisms for the future development of a robotic pruning system.
Article
Green & Sustainable Science & Technology
Muhammad Imran, Azlan Zahid, Salma Mouneer, Orhan Ozcatalbas, Shamsheer Ul Haq, Pomi Shahbaz, Muhammad Muzammil, Muhammad Ramiz Murtaza
Summary: This study aims to analyze the relationship between household dynamics and patterns of energy use. The findings reveal that biomass energy accounts for the majority of household energy consumption, and biomass consumption increases with household size but decreases with income level.
Review
Chemistry, Analytical
Mike O. Ojo, Azlan Zahid
Summary: This article presents a systematic review of the application of deep learning (DL) in controlled environment agriculture (CEA). The review provides an overview of DL applications in different CEA facilities and analyzes commonly used DL models, evaluation parameters, and optimizers. The study found that most research focuses on DL applications in greenhouses, particularly in yield estimation and growth monitoring. The review also discusses current challenges and future research directions in this field.
Article
Plant Sciences
Phillip L. Martin, Kari A. Peter
Summary: Bitter rot is a major disease of apple fruit caused by various species in the Colletotrichum gloeosporioides and C. acutatum species complexes. The timing of infection is unclear due to the hemibiotrophic lifestyle of the causal species. Spore dispersal of C. fioriniae was quantified throughout three growing seasons, showing higher quantities in summer and early fall. Late-season-inoculated fruit had more bitter rot, correlated with optimal temperature and moisture for infection. Management should focus on preventing initial biotrophic infections during favorable weather conditions.
Review
Chemistry, Analytical
Md Sultan Mahmud, Azlan Zahid, Anup Kumar Das
Summary: The ornamental crop industry is important for the US economy, but faces challenges caused by rising labor and agricultural costs. Sensing and automation technologies have been introduced to reduce labor requirements and improve efficiency. This article reviews current and prospective technologies, such as sensors, computer vision, AI, ML, IoT, and robotics, for ornamental crop production.
Article
Agriculture, Multidisciplinary
Md Sultan Mahmud, Long He, Azlan Zahid, Paul Heinemann, Daeun Choi, Grzegorz Krawczyk, Heping Zhu
Summary: This study developed a system for automatic detection and segmentation of fire blight infection in apple orchards using image processing and deep learning approaches.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Agronomy
Mike O. Ojo, Azlan Zahid
Summary: The foundation of effectively predicting plant disease in the early stage using deep learning algorithms is ideal for addressing food insecurity, attracting researchers and agricultural specialists to contribute to its effectiveness.
Review
Agricultural Engineering
Jaemyung Shin, Md. Sultan Mahmud, Tanzeel U. U. Rehman, Prabahar Ravichandran, Brandon Heung, Young K. K. Chang
Summary: Introducing machine vision-based automation is essential to meet the food demand of a growing population in agriculture. It can improve productivity and quality by reducing errors and adding flexibility to the work process. Research on advanced machine vision systems is expected to develop overall agricultural management and provide valuable recommendations for farmers.
Article
Agricultural Engineering
Md Sultan Mahmud, Azlan Zahid, Long He, Heping Zhu, Daeun Choi, Grzegorz Krawczy, Paul Heinemann
Summary: An automatic airflow control system was developed to maximize spray droplet coverage on target trees and minimize off-target loss based on orchard tree canopy densities. The system utilized an iris damper and a 3D LiDAR sensor to control the airflow and acquire tree canopy data, respectively. Field experiments were conducted to build models to evaluate the required airflow.
JOURNAL OF THE ASABE
(2022)
Article
Plant Sciences
Phillip L. Martin, William L. King, Terrence H. Bell, Kari A. Peter
Summary: Bitter rot is a disease of apple caused by fungi in the genus Colletotrichum. Orchard floor management can influence fungal succession in apple fruit with bitter rot, but removal of infected fruit from tree canopies should be the primary focus of bitter rot management.
PHYTOBIOMES JOURNAL
(2022)