Intelligent Outlier Detection for Smart Farming Application using Deep Neural Network

Agriculture is backbone of India. In the emerge of human civilization, agriculture has been an essential part of every human society due to the basic fact that the sustenance of any civilization directly depends on agriculture. Agriculture has the great impact on the economy of the country. This pap...

Full description

Saved in:
Bibliographic Details
Published in2022 IEEE 2nd International Conference on Mobile Networks and Wireless Communications (ICMNWC) pp. 1 - 5
Main Authors Murali, N., Kumar, A. Sasi, Karunamurthy, A., Suseendra, R., Manikandan, S.
Format Conference Proceeding
LanguageEnglish
Published IEEE 02.12.2022
Subjects
Anomaly Detection
Big Sensor Data
Crops
Feature extraction
Feature Selection
IoT
Machine Learning
Neural networks
Productivity
Smart agriculture
Soil
Support vector machines
Online AccessGet full text

Cover

More Information
Summary:Agriculture is backbone of India. In the emerge of human civilization, agriculture has been an essential part of every human society due to the basic fact that the sustenance of any civilization directly depends on agriculture. Agriculture has the great impact on the economy of the country. This paper mainly concentrates on increasing crop production in large scale agricultural area. Due to changing climatic conditions, insufficient water level, insufficient or excessive use of fertilizers etc. the crop production may be affected. Astonishingly, agriculture has not been blessed with the latest advancements in the high-tech space unlike other areas like transport, education, finance, etc. Advancement in agriculture is necessary to balance the demands of people needs as the population grows day by day. The detection of outliers in large-scale area is most challenging tasks. Hence, we explore machine learning methodologies namely Deep Neural Network and Tucker Decomposition is used to detect those anomalies in large scale agriculture area to increase crop production. In this smart farming, the system automatically learns the soil features from IoT sensor data then provides anomaly detection accuracy which helps the farmer to understand the parameters affect the crop growth. This will enhance the crop productivity and economy of farmer. Our system automatically learns the data, analyzes the data and predicts the future.
DOI:10.1109/ICMNWC56175.2022.10031638