A two-stage classification approach for the detection of spider mite- infested cotton using UAV multispectral imagery

Huang, Huasheng; Deng, Jizhong; Lan, Yubin; Yang, Aqing; Deng, Xiaoling; Zhang, Lei; Wen, Sheng; Jiang, Yan; Suo, Gaoyu; Chen, Pengchao

doi:10.1080/2150704x.2018.1498600

Cited by 24 publications

(11 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike traditional feature extraction, CNN is capable of automatically extracting effective features and overcoming the over fitting problem. This architecture strongly improves the performance of remote sensing imagery classification [7], object detection [20], and segmentation [21].…”

Section: Feature Extractionmentioning

confidence: 98%

“…In this case, effective detection of the disease could provide detailed support information for the spraying machines. Traditional assessment of the HLB disease is made by manual investigation throughout the field, which is laborious and time-consuming [7]. Remote sensing (RS) has been frequently used as an effective, efficient, and safe tool for rapid detection of plant diseases [8][9][10][11].…”

Section: Study Sitementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Detection of Helminthosporium Leaf Blotch Disease Based on UAV Imagery

et al. 2019

Self Cite

View full text Add to dashboard Cite

Helminthosporium leaf blotch (HLB) is a serious disease of wheat causing yield reduction globally. Usually, HLB disease is controlled by uniform chemical spraying, which is adopted by most farmers. However, increased use of chemical controls have caused agronomic and environmental problems. To solve these problems, an accurate spraying system must be applied. In this case, the disease detection over the whole field can provide decision support information for the spraying machines. The objective of this paper is to evaluate the potential of unmanned aerial vehicle (UAV) remote sensing for HLB detection. In this work, the UAV imagery acquisition and ground investigation were conducted in Central China on April 22th, 2017. Four disease categories (normal, light, medium, and heavy) were established based on different severity degrees. A convolutional neural network (CNN) was proposed for HLB disease classification. The experiments on data preprocessing, classification, and hyper-parameters tuning were conducted. The overall accuracy and standard error of the CNN method was 91.43% and 0.83%, which outperformed other methods in terms of accuracy and stabilization. Especially for the detection of the diseased samples, the CNN method significantly outperformed others. Experimental results showed that the HLB infected areas and healthy areas can be precisely discriminated based on UAV remote sensing data, indicating that UAV remote sensing can be proposed as an efficient tool for HLB disease detection.

show abstract

Section: Feature Extractionmentioning

confidence: 98%

Section: Study Sitementioning

confidence: 99%

See 1 more Smart Citation

Detection of Helminthosporium Leaf Blotch Disease Based on UAV Imagery

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…These values depended on classification methods. Spider mite infestation levels were estimated based on plant damage (Huang et al 2018). As it is hard to reach 100% accuracy, especially when data are obtained on different days, in most studies, there are certain numbers of false positives (plants are classified as infested while they are healthy) and/or false negatives (plants are classified as healthy while they are infested) (Congalton 1991, Lowe et al 2017.…”

Section: Classification Accuracymentioning

confidence: 99%

Drones: Innovative Technology for Use in Precision Pest Management

Filho

Heldens

Kong

et al. 2019

Journal of Economic Entomology

123

View full text Add to dashboard Cite

Arthropod pest outbreaks are unpredictable and not uniformly distributed within fields. Early outbreak detection and treatment application are inherent to effective pest management, allowing management decisions to be implemented before pests are well-established and crop losses accrue. Pest monitoring is time-consuming and may be hampered by lack of reliable or cost-effective sampling techniques. Thus, we argue that an important research challenge associated with enhanced sustainability of pest management in modern agriculture is developing and promoting improved crop monitoring procedures. Biotic stress, such as herbivory by arthropod pests, elicits physiological defense responses in plants, leading to changes in leaf reflectance. Advanced imaging technologies can detect such changes, and can, therefore, be used as noninvasive crop monitoring methods. Furthermore, novel methods of treatment precision application are required. Both sensing and actuation technologies can be mounted on equipment moving through fields (e.g., irrigation equipment), on (un)manned driving vehicles, and on small drones. In this review, we focus specifically on use of small unmanned aerial robots, or small drones, in agricultural systems. Acquired and processed canopy reflectance data obtained with sensing drones could potentially be transmitted as a digital map to guide a second type of drone, actuation drones, to deliver solutions to the identified pest hotspots, such as precision releases of natural enemies and/or precision-sprays of pesticides. We emphasize how sustainable pest management in 21st-century agriculture will depend heavily on novel technologies, and how this trend will lead to a growing need for multi-disciplinary research collaborations between agronomists, ecologists, software programmers, and engineers.

show abstract

“…Martin and Latheef, (2017) evaluated a groundbased multispectral optical sensor for detecting spider mite damage in greenhouse condition on cotton production. The supervised classi cation approaches such as Support Vector Machine (SVM) and a transferred Convolutional Neural Network (CNN) was reported for mite-infestation using UAV multispectral imagery (Huang et al 2018). Species composition are another GIS and remote sensing approaches applied for modeling ecological niche of Tetranychoid mites (Acari: Tetranychoidea) in in different climates of Tehran Province, Iran (Ghasemi Moghadam et al 2016).…”

Section: Introductionmentioning

confidence: 99%

Influence of climatic Parameters on The Two-spotted Spider Mite population based on Remote Sensing in Southeast of the Caspian Sea

Jokar

2021

Preprint

View full text Add to dashboard Cite

Tetranychus urticae Koch (Acari: Tetranychidae) is a serious pest in cotton fields worldwide. Monitoring of T. urticae with time-series of vegetation index and climatic factors based satellite data was applied to near real-time assessing. The current study aimed to determine correlations between T. urticae population dynamic and effects of Aerosol index of Sentinel-5, Sentinal-2-NDVI (10m), The Land Surface Temperature (LST), MODIS-Evapotranspiration (ET) and CHIRPS-precipitation. Spider mite out-breaking has coincided with the wheat harvesting and where experienced several dusty days with high aerosol index 0.167. Rainfall had a significant negative correlation with T. urticae population (R2 = 0.378) and a threshold precipitation level was estimated at least 2 mm to clean up the canopy. We could not find a significant pattern between temperature and T. urticae population until August 2020 and then the significant positive relationships were observed during August 2020, R² = 0.3519, 0.1283, 0.1675 and 0.178, weekly. Evapotranspiration depicted a statistically synchronous relationship R2 = 0.637 with T. urticae dynamism. There was a positive correlation between increasing NDVI and T. urticae population until August 2020 and then was shifted to negative pattern R2 = 0.273 and 0.139. These findings, aerosol index of sentinel-5 and MODIS-evapotranspiration have potential to forecast spider mite population with high temporal resolution.

show abstract

A two-stage classification approach for the detection of spider mite- infested cotton using UAV multispectral imagery

Cited by 24 publications

References 12 publications

Detection of Helminthosporium Leaf Blotch Disease Based on UAV Imagery

Detection of Helminthosporium Leaf Blotch Disease Based on UAV Imagery

Drones: Innovative Technology for Use in Precision Pest Management

Influence of climatic Parameters on The Two-spotted Spider Mite population based on Remote Sensing in Southeast of the Caspian Sea

Contact Info

Product

Resources

About