Pine pest detection using remote sensing satellite images combined with a multi-scale attention-UNet model

Ye, Wujian; Lao, Junming; Liu, Yijun; Chang, Chin‐Chen; Zhang, Ziwen; Li, Hui; Zhou, Huihui

doi:10.1016/j.ecoinf.2022.101906

Cited by 25 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method can significantly improve detection accuracy. Ye et al (2022) designed an end-to-end automatic pest detection framework based on a multi-scale attention-UNet ( MA-UNet ) model and monophasic images. Compared with the traditional model, the proposed model achieves a much better recall rate of 57.38% in detecting pest infested forest areas, while the recall rates of the Support Vector Machine ( SVM ), UNet, attention-UNet, and MedT models are 14.38, 49.33, 48.02, and 33.64%, respectively.…”

Section: Experimental Design and Results Analysismentioning

confidence: 99%

An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration neural network

Yu,

Wan,

Yousaf

et al. 2024

Poultry Science

View full text Add to dashboard Cite

Section: Experimental Design and Results Analysismentioning

confidence: 99%

An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration neural network

Yu,

Wan,

Yousaf

et al. 2024

Poultry Science

View full text Add to dashboard Cite

“…However, there are large differences between different regions, different types of forest (e.g., temperate forest and tropical forest), plant formations (e.g., grass, shrubs, mediate trees and upper trees) and disturbance categories (e.g., wildfires, forest insects and diseases), making this method difficult to transfer and apply. Among the image pattern recognition methods, traditional machine learning algorithms such as random forests [15][16][17] and support vector machines [18,19], as well as deep learning algorithms like the UNet model [20], are often utilized to monitor forest disturbance areas. But such methods have high requirements on data quality and availability, and the accuracy is often limited by ground survey data [16].…”

Section: Introductionmentioning

confidence: 99%

MART3D: A Multilayer Heterogeneous 3D Radiative Transfer Framework for Characterizing Forest Disturbances

Ouyang,

Qi,

Wang

et al. 2024

Forests

View full text Add to dashboard Cite

The utilization of radiative transfer models for interpreting remotely sensed data to evaluate forest disturbances is a cost-effective approach. However, the current radiative transfer modeling approaches are either too abstract (e.g., 1D models) or too complex (detailed 3D models). This study introduces a novel multilayer heterogeneous 3D radiative transfer framework with medium complexity, termed MART3D, for characterizing forest disturbances. MART3D generates 3D canopy structures accounting for the within-crown clumping by clustering leaves, which is modeled as a turbid medium, around branches, applicable for forests of medium complexity, such as temperate forests. It then automatically generates a multilayer forest with grass, shrub and several layers of trees using statistical parameters, such as the leaf area index and fraction of canopy cover. By employing the ray-tracing module within the well-established LargE-Scale remote sensing data and image Simulation model (LESS) as the computation backend, MART3D achieves a high accuracy (RMSE = 0.0022 and 0.018 for red and Near-Infrared bands) in terms of the bidirectional reflectance factor (BRF) over two RAMI forest scenes, even though the individual structures of MART3D are generated solely from statistical parameters. Furthermore, we demonstrated the versatility and user-friendliness of MART3D by evaluating the band selection strategy for computing the normalized burn ratio (NBR) to assess the composite burn index over a forest fire scene. The proposed MART3D is a flexible and easy-to-use tool for studying the remote sensing response under varying vegetation conditions.

show abstract

“…Automatic image identification technologies based on computer vision are promising in insect detection, as documented in the literature (Ahmad et al, 2022;Júnior and Rieder, 2020;Zacarés et al, 2018). They have been implemented in numerous applications in managing insect disease vectors and controlling pests, such as agricultural and forest pests (Domingues et al, 2022;Duarte et al, 2022;Mendoza et al, 2023), in the classification of parasitised fruit fly pupae (Marinho et al, 2023), the detection of pine pests (Ye et al, 2022), the segmentation of ecological images featuring (Filali et al, 2022), the identification of whitefly (Kamei, 2023), and the automated counting of mosquito eggs (Javed et al, 2023).…”

Section: Introductionmentioning

confidence: 99%

Monitoring the spread of a pathogenic insect on vineyards using UAS

Longhi,

Martino,

Lingua

et al. 2024

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Globalisation has contributed to rapid economic growth but has also exposed vulnerabilities such as the spread of pests in agriculture. An example is the Popillia Japonica Newman beetle, introduced to Italy in 2014, which has caused significant economic losses, mainly affecting vine cultures. Reliable identification of pests is essential for its management, but it is time-consuming and laborious. This has prompted growing interest in image-based methods, supported by computer vision (CV), which can significantly improve efficiency in insect detection. This study aims to evaluate a CV algorithm's effectiveness in identifying adult specimens of Popillia using Near-Infrared sensors on Uncrewed Aerial Systems (UAS). The project, conducted in two vineyards in northern Italy, intends to establish a replicable and standardised data acquisition protocol for future monitoring activities. Insects detected by the CV-based method are validated by manual counting performed by entomologists. In a GIS environment, prescription maps are generated in near real-time to identify where the vineyard is most affected and to guide the drone spraying treatment only on the areas in which the threshold is exceeded. The study demonstrates effective semi-automated monitoring, with a clear correlation between CV-based and manual insect measurements, as indicated by the Pearson correlation coefficients ranging from 0.89 to 0.96. Although the CV-based method may overestimate insect numbers, it provides valuable insights for targeted pest management interventions and damage assessment. The project outcomes offer a promising approach to safeguarding agriculture against invasive species, enhancing regional economic resilience while minimising the spread of insecticide, the required time, and human interaction with harmful substances.

show abstract

Pine pest detection using remote sensing satellite images combined with a multi-scale attention-UNet model

Cited by 25 publications

References 14 publications

An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration neural network

An enhancement algorithm for head characteristics of caged chickens detection based on cyclic consistent migration neural network

MART3D: A Multilayer Heterogeneous 3D Radiative Transfer Framework for Characterizing Forest Disturbances

Monitoring the spread of a pathogenic insect on vineyards using UAS

Contact Info

Product

Resources

About