Comparison of machine learning methods for citrus greening detection on UAV multispectral images

Lan, Yubin; Huang, Zixiao; Deng, Xiaoling; Zhu, Zihao; Huang, Huasheng; Zhang, Zheng; Lian, Bizhen; Zeng, Guoliang; Tong, Zejing

doi:10.1016/j.compag.2020.105234

Cited by 67 publications

(33 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The classification achieved a disease detection accuracy of 92%. In another study [71], the authors extracted vegetation indices (VIs) from multispectral images to enhance information on plant characteristics. The results showed that the VIs feature compressed with PCA and combined with the value of the original data generated an accuracy of 100% using AdaBoost algorithm.…”

Section: Uav Imagingmentioning

confidence: 99%

Computer Vision, IoT and Data Fusion for Crop Disease Detection Using Machine Learning: A Survey and Ongoing Research

et al. 2021

View full text Add to dashboard Cite

Crop diseases constitute a serious issue in agriculture, affecting both quality and quantity of agriculture production. Disease control has been a research object in many scientific and technologic domains. Technological advances in sensors, data storage, computing resources and artificial intelligence have shown enormous potential to control diseases effectively. A growing body of literature recognizes the importance of using data from different types of sensors and machine learning approaches to build models for detection, prediction, analysis, assessment, etc. However, the increasing number and diversity of research studies requires a literature review for further developments and contributions in this area. This paper reviews state-of-the-art machine learning methods that use different data sources, applied to plant disease detection. It lists traditional and deep learning methods associated with the main data acquisition modalities, namely IoT, ground imaging, unmanned aerial vehicle imaging and satellite imaging. In addition, this study examines the role of data fusion for ongoing research in the context of disease detection. It highlights the advantage of intelligent data fusion techniques, from heterogeneous data sources, to improve plant health status prediction and presents the main challenges facing this field. The study concludes with a discussion of several current issues and research trends.

show abstract

Section: Uav Imagingmentioning

confidence: 99%

Computer Vision, IoT and Data Fusion for Crop Disease Detection Using Machine Learning: A Survey and Ongoing Research

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Considering that the accuracy is acceptable in daily identification tasks, it verifies that the proposed WLTs-SDTL can generate a small efficient network suitable for mobile terminals or edge computing devices with low computational power in the practical application of pest and disease identification. 7 Wireless Communications and Mobile Computing devastating disease in the world citrus production, which seriously restricts the development of the citrus industry [37]. The etiolation of leaves can be used for its in-field identification.…”

Section: 22mentioning

confidence: 99%

A Sparse Deep Transfer Learning Model and Its Application for Smart Agriculture

Chen

Zhang

Chen

et al. 2021

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The introduction of deep transfer learning (DTL) further reduces the requirement of data and expert knowledge in various uses of applications, helping DNN-based models effectively reuse information. However, it often transfers all parameters from the source network that might be useful to the task. The redundant trainable parameters restrict DTL in low-computing-power devices and edge computing, while small effective networks with fewer parameters have difficulty transferring knowledge due to structural differences in design. For the challenge of how to transfer a simplified model from a complex network, in this paper, an algorithm is proposed to realize a sparse DTL, which only transfers and retains the most necessary structure to reduce the parameters of the final model. Sparse transfer hypothesis is introduced, in which a compressing strategy is designed to construct deep sparse networks that distill useful information in the auxiliary domain, improving the transfer efficiency. The proposed method is evaluated on representative datasets and applied for smart agriculture to train deep identification models that can effectively detect new pests using few data samples.

show abstract

“…As technological development has changed satellite technology, imagery has reached a more spatial, temporal, and radiometric resolution. Digital agriculture had profited from this advancement and had exploited satellite imagery in precision agriculture that was more oriented using aerial and handheld remote sensing [4,18].…”

Section: Very High-resolution Satellite Imagery and Aerial Imagerymentioning

confidence: 99%

“…Recent studies have shown the advantage of proximal and handheld remote sensing, including aerial imagery to manage the agricultural sector in a finer centimetric spatial resolution [3,4], advanced techniques have allowed monitoring the variability within the field so the intervention could be specific and punctual. The main application had permitted the detection of agricultural disease and its extent to establish the suitable way of intervention and the adequate amount of pesticide to apply [5,6].…”

Section: Introductionmentioning

confidence: 99%

Optical Satellite Images Services for Precision Agricultural use: A Review

Dakir¹,

Zahra²,

Omar³

2021

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

The recent advances in geoscience technologies and earth observation tools have evolved in recent years in a fast cadence. Since 1980 with the apparition of Landsat mission firstly named the Earth Resources Technology Satellite, with 80 m spatial resolution; the ability to capture finest details was limited. The emergence of new concepts in agriculture like digital agriculture and precision agriculture (PA) was challenging the capacity of satellite images to capture the variation within the field. The domain that was firstly more dedicated to aerial and handheld remote sensing is now accessible to satellite remote sensing thanks to recent advancements that had provided more satisfactory spatial resolution reaching 0.3 m in a daily revisit frequency. Different providers have launched commercial, very high-resolution nanosatellite constellations to respond to more precise needs, and increasing the availability of remotely sensed images expands the horizon of our choices of imagery sources. The present work intends to compare spatial, temporal, spectral, and radiometric resolution, considering the farmer's requirements and precision agriculture application's requirements nowadays available and the future generation satellite missions, and benchmark the major satellite images providers. It exposes the characteristics and future missions to facilitate adequate choice. The criterion of selecting the appropriate sensors among the considerable amount of providers was limited to optical data, with less than 10 m spatial resolution and frequent revisit time. The offer of imagery is majorly commercial missions; the available open and accessible alternative is limited in the sentinel mission with 10 m spatial resolution.

show abstract

Comparison of machine learning methods for citrus greening detection on UAV multispectral images

Cited by 67 publications

References 38 publications

Computer Vision, IoT and Data Fusion for Crop Disease Detection Using Machine Learning: A Survey and Ongoing Research

Computer Vision, IoT and Data Fusion for Crop Disease Detection Using Machine Learning: A Survey and Ongoing Research

A Sparse Deep Transfer Learning Model and Its Application for Smart Agriculture

Optical Satellite Images Services for Precision Agricultural use: A Review

Contact Info

Product

Resources

About