UAV-based weed detection in Chinese cabbage using deep learning

Ong, Pauline; Teo, Kiat Soon; Kiong, Sia Chee

doi:10.1016/j.atech.2023.100181

Cited by 23 publications

(12 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The use of CNN models is prevalent in crop classification using UAV systems due to the difficulty in capturing temporal images of the same cultivated area, presenting a challenge for other models, caused by operational constraints and the data acquisition process associated with drones. Despite this, in the study [64], the CNN model has been proven to be highly effective in weed detection amongst commercial crops, such as Chinese cabbage. The CNN-based classifier was integrated with UAV imagery, achieving an average sensitivity of 80.29%, average specificity of 93.88%, average precision of 75.45%, and average accuracy of 92.41%, outperforming the RF-based classifier.…”

Section: Crop Classification Using Uav Datamentioning

confidence: 91%

Deep Learning Models for the Classification of Crops in Aerial Imagery: A Review

et al. 2023

View full text Add to dashboard Cite

In recent years, the use of remote sensing data obtained from satellite or unmanned aerial vehicle (UAV) imagery has grown in popularity for crop classification tasks such as yield prediction, soil classification or crop mapping. The ready availability of information, with improved temporal, radiometric, and spatial resolution, has resulted in the accumulation of vast amounts of data. Meeting the demands of analysing this data requires innovative solutions, and artificial intelligence techniques offer the necessary support. This systematic review aims to evaluate the effectiveness of deep learning techniques for crop classification using remote sensing data from aerial imagery. The reviewed papers focus on a variety of deep learning architectures, including convolutional neural networks (CNNs), long short-term memory networks, transformers, and hybrid CNN-recurrent neural network models, and incorporate techniques such as data augmentation, transfer learning, and multimodal fusion to improve model performance. The review analyses the use of these techniques to boost crop classification accuracy by developing new deep learning architectures or by combining various types of remote sensing data. Additionally, it assesses the impact of factors like spatial and spectral resolution, image annotation, and sample quality on crop classification. Ensembling models or integrating multiple data sources tends to enhance the classification accuracy of deep learning models. Satellite imagery is the most commonly used data source due to its accessibility and typically free availability. The study highlights the requirement for large amounts of training data and the incorporation of non-crop classes to enhance accuracy and provide valuable insights into the current state of deep learning models and datasets for crop classification tasks.

show abstract

Section: Crop Classification Using Uav Datamentioning

confidence: 91%

Deep Learning Models for the Classification of Crops in Aerial Imagery: A Review

et al. 2023

View full text Add to dashboard Cite

show abstract

“…It is employed for detecting edges, counting objects, removing noise, and calculating growth, particle, color, texture, and shape analysis. For example, images of Chinese cabbage were analyzed by four image processing steps namely, image acquisition by UAV, segmentation by Simple Linear Iterative Clustering (SLIC) Superpixel algorithm, feature selection by Local Binary Pattern (LBP) for Random Forest (RF), and classification by RF and CNN [61].…”

Section: Image Analysismentioning

confidence: 99%

AI-Based UAV Swarms for Monitoring and Disease Identification of Brassica Plants Using Machine Learning: A Review

Ali¹,

Deng²,

Shaikh³

et al. 2024

CSSE

View full text Add to dashboard Cite

Technological advances in unmanned aerial vehicles (UAVs) pursued by artificial intelligence (AI) are improving remote sensing applications in smart agriculture. These are valuable tools for monitoring and disease identification of plants as they can collect data with no damage and effects on plants. However, their limited carrying and battery capacities restrict their performance in larger areas. Therefore, using multiple UAVs, especially in the form of a swarm is more significant for monitoring larger areas such as crop fields and forests. The diversity of research studies necessitates a literature review for more progress and contribution in the agricultural field. In this review, the comparative analysis of existing literature surveys is explored. This paper aims to provide an overview of AIbased UAV swarms, different cameras and sensors, image processing, and machine learning (ML) algorithms for image analysis having the purpose of monitoring and disease identification. Brassica plants are focused as they are grown on wider scales globally. Brassica species, the commonly infected diseases, and different types of disease detection methods are discussed. Investigations show the significance of using UAV swarms for growth monitoring growth for yield estimation, health monitoring, water status monitoring and irrigation management, nutrition disorders monitoring, pest and disease detection, and pesticide and fertilizer spraying in Brassica plants. Finally, some challenges of swarm-based applications are also addressed that require future consideration. The significance of this paper is that it suggests its readers embrace swarm-based technologies in the pursuit of more efficient production with relevant economic benefits. KEYWORDSArtificial intelligence (AI); Brassica; disease identification; image analysis; image processing; machine learning (ML); monitoring; swarm; remote sensors; unmanned aerial vehicles (UAVs) UAV Swarm Brassica Plants ML Approaches Research FocusFigure 1: Review framework with the research focusWe searched articles in different databases, such as Science Direct, Google Scholar, SpringerLink, SAGE Journals, IEEE Xplore, and Wiley Online Library Journal. In the advanced search section of these database repositories, the period from 2018 to 2023 was selected. While plotting the framework we ensured to include all the important studies. After selecting the first batch of research papers, their related articles were also searched. To extract relevant articles, we read the abstract of all the retrieved articles. The screening criteria include deleting non-English and duplicate papers. We included the papers which were focused on optical sensors and machine learning approaches. We excluded the papers which were focused on other remote sensing tools and sensors. This review is proposed for farmers and researchers who are inclined to study Brassica plants and UAV swarms. The major contributions of the paper include:

show abstract

“…RF is an algorithm based on decision trees and a self-service resampling method [46,47]. The principle of RF is to build multiple decision trees and fuse them to get a more accurate and stable model, which is a combination of the bagging idea and random selection features.…”

Section: Identification Model Of Rubber Tree Powdery Mildewmentioning

confidence: 99%

Recognition of Rubber Tree Powdery Mildew Based on UAV Remote Sensing with Different Spatial Resolutions

Zeng

Fang

Yin

et al. 2023

Drones

View full text Add to dashboard Cite

Rubber tree is one of the essential tropical economic crops, and rubber tree powdery mildew (PM) is the most damaging disease to the growth of rubber trees. Accurate and timely detection of PM is the key to preventing the large-scale spread of PM. Recently, unmanned aerial vehicle (UAV) remote sensing technology has been widely used in the field of agroforestry. The objective of this study was to establish a method for identifying rubber trees infected or uninfected by PM using UAV-based multispectral images. We resampled the original multispectral image with 3.4 cm spatial resolution to multispectral images with different spatial resolutions (7 cm, 14 cm, and 30 cm) using the nearest neighbor method, extracted 22 vegetation index features and 40 texture features to construct the initial feature space, and then used the SPA, ReliefF, and Boruta–SHAP algorithms to optimize the feature space. Finally, a rubber tree PM monitoring model was constructed based on the optimized features as input combined with KNN, RF, and SVM algorithms. The results show that the simulation of images with different spatial resolutions indicates that, with resolutions higher than 7 cm, a promising classification result (>90%) is achieved in all feature sets and three optimized feature subsets, in which the 3.4 cm resolution is the highest and better than 7 cm, 14 cm, and 30 cm. Meanwhile, the best classification accuracy was achieved by combining the Boruta–SHAP optimized feature subset and SVM model, which were 98.16%, 96.32%, 95.71%, and 88.34% at 3.4 cm, 7 cm, 14 cm, and 30 cm resolutions, respectively. Compared with SPA–SVM and ReliefF–SVM, the classification accuracy was improved by 6.14%, 5.52%, 12.89%, and 9.2% and 1.84%, 0.61%, 1.23%, and 6.13%, respectively. This study’s results will guide rubber tree plantation management and PM monitoring.

show abstract

UAV-based weed detection in Chinese cabbage using deep learning

Cited by 23 publications

References 33 publications

Deep Learning Models for the Classification of Crops in Aerial Imagery: A Review

Deep Learning Models for the Classification of Crops in Aerial Imagery: A Review

AI-Based UAV Swarms for Monitoring and Disease Identification of Brassica Plants Using Machine Learning: A Review

Recognition of Rubber Tree Powdery Mildew Based on UAV Remote Sensing with Different Spatial Resolutions

Contact Info

Product

Resources

About