Classification and Identification of Tea Diseases Based on Improved YOLO V7 Model of MobileNeXt DOI Creative Commons
Yuxin Xia, Wenxia Yuan, Shihao Zhang

и другие.

Research Square (Research Square), Год журнала: 2023, Номер unknown

Опубликована: Ноя. 1, 2023

Abstract To overcome the constraints associated with conventional approaches used in classification and detection of tea diseases, which are characterized by their limited accuracy sluggish responsiveness, this study introduces an enhanced YOLOv7 lightweight model algorithm integrated MobileNeXt. This refinement not only bolsters model's capacity for extracting processing features but also effectively lightens computational load, expedites recognition, integrates a dual-layer routing attention mechanism visual converter to enhance capture crucial details textures within disease images. Consequently, these enhancements lead improved performance efficiency, ensuring precise rapid identification diseases. Furthermore, incorporates more appropriate SIoU as loss function, mitigating losses, minimizing omissions, reducing misclassifications, thus resulting superior even complex image backgrounds. Based on training outcomes, attains Precision, Recall mean Average Precision scores 93.5%, 89.9%, 92.1%, respectively, marking substantial 5.06%, 2.16%, 2.91% compared original model. Additionally, size is reduced 19.12%, its speed accelerates 11.13%. excels accurately expediting.

Язык: Английский

Where is tea grown in the world: A robust mapping framework for agroforestry crop with knowledge graph and sentinels images DOI
Yufeng Peng, Bingwen Qiu, Zhenghong Tang

и другие.

Remote Sensing of Environment, Год журнала: 2024, Номер 303, С. 114016 - 114016

Опубликована: Янв. 26, 2024

Язык: Английский

Процитировано

11
Analysis of Tea Plantation Suitability Using Geostatistical and Machine Learning Techniques: A Case of Darjeeling Himalaya, India DOI Open Access
Netrananda Sahu, Pritiranjan Das, Atul Saini

и другие.

Sustainability, Год журнала: 2023, Номер 15(13), С. 10101 - 10101

Опубликована: Июнь 26, 2023

This study aimed to identify suitable sites for tea cultivation using both random forest and logistic regression models. The utilized 2770 sample points map the plantation suitability zones (TPSZs), considering 12 important conditioning factors, such as temperature, rainfall, elevation, slope, soil depth, drainability, electrical conductivity, base saturation, texture, pH, normalized difference vegetation index (NDVI), land use cover (LULC). data were ArcGIS 10.2 models calibrated 70% of total data, while remaining 30% used validation. final TPSZ was classified into four different categories: highly zones, moderately marginally not-suitable zones. revealed that (RF) model more precise than model, with areas under curve (AUCs) 85.2% 83.3%, respectively. results indicated well-drained a pH range between 5.6 6.0 is ideal farming, highlighting importance climate properties in cultivation. Furthermore, emphasized need balance economic environmental considerations when expansion. findings this provide insights site selection can aid farmers, policymakers, other stakeholders making informed decisions regarding

Язык: Английский

Процитировано

13
Crop suitability analysis for the coastal region of India through fusion of remote sensing, geospatial analysis and multi-criteria decision making DOI Creative Commons

Nishtha Sawant,

Bappa Das, Gopal Ramdas Mahajan

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Март 13, 2025

Abstract Crop suitability analysis plays an important role in identifying and utilizing the areas suitable for better crop growth higher yield without deteriorating natural resources. The present study aimed to identify rice coconut cultivation across coastal region of India using analytic hierarchy process (AHP) integrated with geographic information systems (GIS) remote sensing. A total nine parameters were selected including elevation, slope, soil depth, drainage, texture, pH, organic carbon, rainfall, temperature a land use cover (LULC) constraint map. This represents first-ever application approach combining AHP, GIS, sensing entire India. weights subclasses assigned AHP method based on experts’ opinions. Subsequently, all thematic maps overlaid weighted overlay generate Separately, LULC mask map was used extract create crop-specific maps. final classified into four different classes: highly suitable, moderately marginally not production. findings revealed that approximately 13.68% area around 19.26% 18.35% being respectively, 13.76% cultivation. Similarly, cultivation, 11% 27.40% 18.34% suitable. However, about 35% deemed permanently unsuitable any type validated under receiver operating characteristic curve (AUROC). AUROC values found be 0.764 0.740 indicating high accuracy. By strategically cultivating locations identified current study, other crops, it is possible achieve financial viability agricultural production by increasing causing harm

Язык: Английский

Процитировано

0
Prediction and quality zoning of potentially suitable areas for Panax notoginseng cultivation using MaxEnt and random forest algorithms in Yunnan Province, China DOI Creative Commons

Fengzhi Wu,

Yong Wang, Mingming Zheng

и другие.

Industrial Crops and Products, Год журнала: 2025, Номер 229, С. 120960 - 120960

Опубликована: Апрель 10, 2025

Язык: Английский

Процитировано

0
Integrating extreme temperature metrics into tea land suitability models: a GIS-based assessment in the ecological and cultural tourism circle in Western Hubei, China DOI
Yarong CHEN, Bowen Fu,

Yuqing Xie

и другие.

Theoretical and Applied Climatology, Год журнала: 2025, Номер 156(5)

Опубликована: Май 1, 2025

Язык: Английский

Процитировано

0
Explainable artificial intelligence to estimate the Sri Lankan (Ceylon) Tea crop yield DOI Creative Commons
Lakindu Mampitiya,

Harindu S. Sumanasekara,

Namal Rathnayake

и другие.

Smart Agricultural Technology, Год журнала: 2025, Номер unknown, С. 100999 - 100999

Опубликована: Май 1, 2025

Язык: Английский

Процитировано

0
Evaluation of Land Suitability for Olive (Olea europaea L.) Cultivation Using the Random Forest Algorithm DOI Creative Commons
Ayşe Yavuz Özalp, Halil Akıncı

Agriculture, Год журнала: 2023, Номер 13(6), С. 1208 - 1208

Опубликована: Июнь 7, 2023

Many large dams built on the Çoruh River have resulted in inundation of olive groves Artvin Province, Turkey. This research sets out to identify suitable locations for cultivation using random forest (RF) algorithm. A total 575 plots currently listed Farmer Registration System, where is practiced, were used as inventory data training and validation RF model. In order determine areas can be carried out, a land suitability map was created by taking into account 10 parameters including average annual temperature, precipitation, slope, aspect, use capability class, sub-class, soil depth, other properties, solar radiation, cover. According this map, an area 53,994.57 hectares detected production within study region. To validate model, receiver operating characteristic (ROC) curve under ROC (AUC) utilized. As result, AUC value determined 0.978, indicating that method may successfully determining lands particular, well crop-based general.

Язык: Английский

Процитировано

9
Classification and identification of tea diseases based on improved YOLOv7 model of MobileNeXt DOI Creative Commons
Yuxin Xia, Wenxia Yuan, Shihao Zhang

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Май 23, 2024

Abstract To address the issues of low accuracy and slow response speed in tea disease classification identification, an improved YOLOv7 lightweight model was proposed this study. The MobileNeXt used as backbone network to reduce computational load enhance efficiency. Additionally, a dual-layer routing attention mechanism introduced model’s ability capture crucial details textures images, thereby improving accuracy. SIoU loss function employed mitigate missed erroneous judgments, resulting recognition amidst complex image backgrounds.The revised achieved precision, recall, average precision 93.5%, 89.9%, 92.1%, respectively, representing increases 4.5%, 1.9%, 2.6% over original model. Furthermore, volum reduced by 24.69M, total param 12.88M, while detection increased 24.41 frames per second. This enhanced efficiently accurately identifies types, offering benefits lower parameter count faster detection, establishing robust foundation for monitoring prevention efforts.

Язык: Английский

Процитировано

3
Advancing Agricultural Land Suitability in Urbanized Semi-Arid Environments: Insights from Geospatial and Machine Learning Approaches DOI Creative Commons
S. Sathiyamurthi, Subbarayan Saravanan,

M. Ramya

и другие.

ISPRS International Journal of Geo-Information, Год журнала: 2024, Номер 13(12), С. 436 - 436

Опубликована: Дек. 3, 2024

Rising food demands are increasingly threatened by declining crop yields in urbanizing riverine regions of Southern Asia, exacerbated erratic weather patterns. Optimizing agricultural land suitability (AgLS) offers a viable solution for sustainable productivity such challenging environments. This study integrates remote sensing and field-based geospatial data with five machine learning (ML) algorithms—Naïve Bayes (NB), extra trees classifier (ETC), random forest (RF), K-nearest neighbors (KNN), support vector machines (SVM)—alongside land-use/land-cover (LULC) considerations the food-insecure Dharmapuri district, India. A grid searches optimized hyperparameters using factors as slope, rainfall, temperature, texture, pH, electrical conductivity, organic carbon, available nitrogen, phosphorus, potassium, calcium carbonate. The tuned ETC model showed lowest root mean squared error (RMSE = 0.15), outperforming RF 0.18), NB 0.20), SVM 0.22), KNN 0.23). AgLS-ETC map identified 29.09% area highly suitable (S1), 19.06% moderately (S2), 16.11% marginally (S3), 15.93% currently unsuitable (N1), 19.21% permanently (N2). By incorporating Landsat-8 derived LULC to exclude forests, water bodies, settlements, these estimates were adjusted 19.08% 14.45% 11.40% 10.48% 9.58% Focusing on model, followed land-use analysis, provides robust framework optimizing planning, ensuring protection ecological social developing countries.

Язык: Английский

Процитировано

3
Effect of bioactive compounds in processed Camellia sinensis tea on the intestinal barrier DOI
Nan Chen, Peng Yao, Muhammad Salman Farid

и другие.

Food Research International, Год журнала: 2024, Номер 199, С. 115383 - 115383

Опубликована: Ноя. 20, 2024

Язык: Английский

Процитировано

2