Use of artificial neural networks of different architecture and learning rate to predict soil humus content using normalized difference vegetation index
The program of remote sensing of the normalized difference vegetation index. Prediction of humus content in soil using normalized vegetation difference index. Using artificial neural networks to predict the normalized vegetation difference index.
| Рубрика | Сельское, лесное хозяйство и землепользование |
| Вид | статья |
| Язык | английский |
| Дата добавления | 21.03.2024 |
| Размер файла | 16,3 K |
Отправить свою хорошую работу в базу знаний просто. Используйте форму, расположенную ниже
Студенты, аспиранты, молодые ученые, использующие базу знаний в своей учебе и работе, будут вам очень благодарны.
Размещено на http://www.allbest.ru/
Размещено на http://www.allbest.ru/
Use of artificial neural networks of different architecture and learning rate to predict soil humus content using normalized difference vegetation index
Lykhovyd Pavlo Volodymyrovych
candidate of agricultural sciences, postdoc student
Institute of Climate-Smart Agriculture of NAAS, Ukraine
As far as remotely sensed normalized difference vegetation index (NDVI) applications extend to more specific subjects than plants vegetation conditions assessment, studies are conducted to adopt this spatial index for evaluation of soil properties, e.g., soil nutrients and organic matter content, electrical conductivity, pH, etc. [1 -3]. Our studies are directed to the development of mathematical model for derivation of soil humus content using normalized difference vegetation index values. Although some success has been achieved in this field by the means of regression analysis, the prediction accuracy and model fitting quality are still insufficient to provide it for practical implementation [4]. As it is known that artificial neural networks (ANN) in many cases provide much better results than traditional regression analysis, the study was performed with different ANN architecture and learning rates to establish the relationship and improve the quality of soil humus content prediction based on the values of spatial vegetation index [5].
The study was carried out in the fallow-field period of 2022-2023 for the fields of Kherson oblast. 1478 bare-soil NDVI values were collected and generalized by the major natural and agricultural districts of the region according to the zoning [6]. Average NDVI values were associated with the soil humus content of the corresponding district of Kherson oblast. The soil humus content was taken from the results of the regional soil surveys and investigations conducted in the Institute of Climate-Smart Agriculture (former Institute of Irrigated Agriculture). In total, 34 data pairs “NDVI (pts.) - soil humus content (%)” were created and analyzed.
Statistical analysis in the pilot study was conducted by the means of nine regression model types [4]. In this study we applied ANN approach for the same dataset. Tiberius XL software was used for ANN modeling [7]. The settings included different architecture of the network (with the minimum 1 and the maximum 5 neurons in the hidden layer), and different values of the learning rate (from 0.7 to 1.0). The training was conducted within 1000 epochs. The results of the modeling were analyzed using coefficients of correlation (by Pearson), coefficient of determination, and mean absolute percentage error (MAPE). The interpretation of MAPE values was conducted according to [8], while the coefficients values were interpreted according to [9].
The results of the modeling using different ANN are presented in the Table 1. The best fitting quality and prediction accuracy are in the variant of ANN with 5 neurons in the hidden layer and learning rate of 0.8. Less number of neurons in the hidden layer, as well as higher learning rate, resulted in statistically significant decrease of the fitting quality and accuracy. The best variant of ANN model surpassed the best regression model (cubic) both in terms of curve approximation (R2 of 0.29 vs. 0.25) and prediction accuracy (MAPE of 12.28% vs. 13.77%) [4]. In general, ANN-based algorithm (Tiberius XL uses back-propagation of the error) is superior traditional regression analysis in case of humus content prediction by the values of NDVI. However, the drawback is that we can use the results only within Tiberius XL model, as it is impossible to derive the mathematical equation out of the ANN.
As a solution, we tried a combined ANN-regression approach, that provides an opportunity of mathematical expression of the ANN-based model, however, with a bit lower accuracy. The methodology of such a combination of two mathematical techniques was successfully introduced in the study of beans yield prediction [10]. In the current study, there also was slight improvement in the model quality: MAPE of 13.22%, Pearson's correlation coefficient of 0.54, and coefficient of determination - 0.29. The new combined model for humus content in the soils of Kherson oblast looks as follows (1):
y = 43.164 - 1168.8x + 12122x2 - 54272x3 + 89044x4 (1)
where:
y - humus content in the soil, %; x - bare-soil NDVI value for the area.
Table 1
Results of modeling soil humus content by the bare-soil NDVI values in ANN with different architecture and learning rate
|
Variant |
R |
R2 |
MAPE |
||
|
Learning rate |
1.0 |
0.45 |
0.20 |
13.27% |
|
|
Number of neurons |
5 |
||||
|
Learning rate |
1.0 |
0.48 |
0.23 |
12.45% |
|
|
Number of neurons |
1 |
||||
|
Learning rate |
0.8 |
0.48 |
0.23 |
14.24% |
|
|
Number of neurons |
1 |
||||
|
Learning rate |
0.8 |
0.54 |
0.29 |
12.28% |
|
|
Number of neurons |
5 |
||||
|
Combined model |
0.54 |
0.29 |
13.22% |
Author's study results
neural network humus soil
According to [9], the combined mathematical model has moderate fitting quality, while the predictive accuracy of the model with accordance to [8] is good. Further extension of the number of inputs, and inclusion of new “NDVI (pts.) - soil humus content (%)" pairs in the mathematical model may result in additional enhancement of the model quality and precision. Even in its current state the model certifies about promising prospects for implementation of spatial NDVI in soil surveys.
References
[1] Mazur P., Gozdowski D., & Wojcik-Gront E. (2022) Soil electrical conductivity and satellite- derived vegetation indices for evaluation of phosphorus, potassium and magnesium content, pH, and delineation of within-field management zones. Agriculture, 12(6), 883.
[2] Mazur P., Gozdowski D., & Wnuk A. (2022) Relationships between soil electrical conductivity and Sentinel-2-derived NDVI with pH and content of selected nutrients. Agronomy, 12(2), 354.
[3] Zhang Y., Guo L., Chen Y., Shi T., Luo M.,Ju Q., Zhang H., & Wang S. (2019) Prediction of soil organic carbon based on Landsat 8 monthly NDVI data for the Jianghan Plain in Hubei Province, China. Remote Sensing, 11(14), 1683.
[4] Lykhovyd, P. V. (2023) Using normalized difference vegetation index to estimate humus content in the soils of the South of Ukraine. Sectoral research XXI: characteristics and features: collection of scientific papers "SCIENTIA" with Proceedings of the V International Scientific and Theoretical Conference. (pp. 116-118) February 3, 2023, Chicago, USA. European Scientific Platform.
[5] Vozhehova, R. A., Lykhovyd, P. V., Lavrenko, S. O., Kokovikhin, S. V., Lavrenko, N. M., Marchenko, T. Yu., Sydyakina, O. V., Hlushko, T. V., & Nesterchuk, V. V. (2019) Artificial neural network use for sweet corn water consumption prediction depending on cultivation technology peculiarities. Research Journal of Pharmaceutical, Biological and Chemical Sciences, 10(1), 354-358.
[6] Bychkov V. V., Ostapov V. I., Zhuravlev A. I., Kotliar N. M., Lirnyk V. A., Lomonosov P. I., Pisarenko V. A., & Funtov A. P. (1987) Scientifically based system of agriculture of Kherson oblast. Kherson. 448 pp.
[7] Brierley, P. D. (1998) Some practical applications of neural networks in the electricity industry. Cranfield University (United Kingdom).
[8] Blasco B. C., Moreno J. J. M., Pol A. P., & Abad A. S. (2013) Using the R-MAPE index as a resistant measure of forecast accuracy, Psicothema, 25(4), 500-506.
[9] Evans, J. D. (1996) Straightforward statistics for the behavioral sciences. Thomson Brooks/Cole Publishing Co.
[10] Lavrenko, S., Lykhovyd, P., Lavrenko, N., Ushkarenko, V., & Maksymov, M. (2022) Beans (Phaseolus vulgaris L.) yields forecast using normalized difference vegetation index. International Journal of Agricultural Technology, 18(3),1033-1044.
Размещено на Allbest.ru
...Подобные документы
Місце Англії за рейтингом "Global competitivness index", "Human Development Index", "Corruption Perceptions Index". Порівняльний аналіз обсягу та динаміки ВВП країни із середнім по Європейського Союзу. Аналіз ВВП на душу населення країни та у відсотках.
курсовая работа [4,4 M], добавлен 05.03.2013Principles of green analytical metrics. National environment method index. Application of GAC metrics. Complementary green analytical procedure index. Additive color model to analytical method evaluation. Examples of analytical eco-scale calculation.
дипломная работа [2,0 M], добавлен 27.11.2022Leadership and historical approach. Effect, which leader makes on group. Developing leadership skills. Exercise control as function of the leader is significant difference between managers and leaders. Common points of work of leader and manager.
доклад [37,7 K], добавлен 13.02.2012Понятие о нейронных сетях и параллели из биологии. Базовая искусственная модель, свойства и применение сетей. Классификация, структура и принципы работы, сбор данных для сети. Использование пакета ST Neural Networks для распознавания значимых переменных.
реферат [435,1 K], добавлен 16.02.2015Тестування і діагностика є необхідним аспектом при розробці й обслуговуванні обчислювальних мереж. Компанія Fluke Networks є лідером розробок таких приладів. Такими приладами є аналізатори EtherScope, OptіVіew Fluke Networks, AnalyzeAir та InterpretAir.
реферат [370,5 K], добавлен 06.01.2009Решение задач прогнозирования цен на акции "Мазут" на 5 дней, построение прогноза для переменной "LOW". Работа в модуле "Neural networks", назначение вкладок и их характеристика. Построение системы "Набор программистов" нечеткого логического вывода.
курсовая работа [3,2 M], добавлен 26.12.2016Модели оценки кредитоспособности физических лиц в российских банках. Нейронные сети как метод решения задачи классификации. Описание возможностей программы STATISTICA 8 Neural Networks. Общая характеристика основных этапов нейросетевого моделирования.
дипломная работа [1,4 M], добавлен 21.10.2013Сравнение макроэкономической статистики США 2011-2015 гг. Корреляция цены инструментов отдельных секторов. Развитие рынка страхования. Показатели степени их влияния на отраслевые индексы страхования. Макроэкономические индексы на DJ Insurance Index.
доклад [451,6 K], добавлен 25.03.2016Definition of Metaphor as a Figurative and Expressive Means of Language. Types and the Mechanism of Education of the Metaphor, its difference from comparison. Metaphor role in speech genres, its influence on emotions and imagination of the recipient.
реферат [43,8 K], добавлен 04.05.2012Борьба с коррупцией как общегосударственная задача. Отношение ко взяточничеству в дореволюционной и Советской России. Понятие коррупции и классификация ее видов. Corruption Perceptions Index по России. Области обогащения. Методы борьбы с коррупцией.
курсовая работа [32,1 K], добавлен 15.11.2013What’s a hippie? What’s the difference between an old hippie and a new hippie? Once a hippie, always a hippie? These and similar questions are the source of much debate today.
реферат [384,2 K], добавлен 03.02.2004Розробка сайту-візитки компанії, яка надає послуги в ІТ-галузі та оцінювання створеного сайту. Структурне розположення усіх html, css—файлів та зображень. Створення текстового документу з іменем index та розширенням .html. Тестування сторінки в браузері.
курсовая работа [1,9 M], добавлен 25.06.2015Un tour de France en passant par les villes et les regions. La difference entre "le palais" et le "chateau". Les chateaux de la Loire. La presentation "a table ronde". Le vocabulaire culturel. La poesie. La litterature. Observez le billet de train.
презентация [13,8 M], добавлен 04.12.2008Технологии решения задач с использованием нейронных сетей в пакетах расширения Neural Networks Toolbox и Simulink. Создание этого вида сети, анализ сценария формирования и степени достоверности результатов вычислений на тестовом массиве входных векторов.
лабораторная работа [352,2 K], добавлен 20.05.2013The oxygen consumption as the most important difference between pediatric and adult patients. Basic anesthetic technique, postoperative care, intra-operative problem. Anesthesia breathing systems and the efficiency of the present breathing circuits.
презентация [1,0 M], добавлен 15.04.2010Concept of competitiveness and competition, models. Russia’s endowment. Engendered structural dominance and performance. The state of Russian competitiveness according to the Global Competitiveness Index. Place in the world, main growth in detail.
курсовая работа [1,2 M], добавлен 28.05.2014Розробка сайту, який буде мати можливість наповнення інформацією про стан команд та їх гравців у лізі в режимі реального часу. Переваги використання технології web 2.0. Написання програмного коду веб-сайту та його реалізація, головна сторінка Index.php.
дипломная работа [3,4 M], добавлен 18.08.2014Архитектура базы данных. Требования к именованию файлов в операционной системе. Величина приращения при увеличении и максимальный размер. Выделение пространства для таблиц и индексов. Таблица Index Allocation Map. Принцип работы журнала транзакций.
презентация [48,3 K], добавлен 10.11.2013Маркетинговое исследование уровня удовлетворённости потребительских запросов. Индекс удовлетворенности покупателей - Сustomer Satisfaction Index. Обзор рынка информационно-справочных услуг. Выявление удовлетворённости клиентов городской справочной "323".
курсовая работа [391,7 K], добавлен 25.02.2012Characteristic of the intellectual movement that gained strength in reaction to the Industrial Revolution. Studying the difference between "Songs of Innocence" and "Songs of Experience". Description of the main authors belonged to London Romanticism.
контрольная работа [27,4 K], добавлен 30.04.2011
