Latest Issue
Study on Mechanical Structure Design for Plug-and-play Wheel Mobile Robot
Published: December 31,2023PI Controller for Velocity Controller Design based on Lumped Parameter Estimation: Simulation and Experiment
Published: December 31,2023Attitude Estimation by using Unscented Kalman Filter with Constraint State
Published: December 31,2023Characterization Study of Cambodian Natural Rubber and Clay Composites for Shock Absorption Floor Mat
Published: December 31,2023Selection of Observed Gridded Rainfall Data for different Analysis Purposes in Cambodia
Published: December 31,2023An Empirical Investigation of Gold Price Forecasting Using ARIMA Compare with LSTM Model
Published: December 31,2023Prediction of California Bearing Ratio with Soil Properties of Road Subgrade Materials in Cambodia
Published: December 31,2023Non-intrusive Load Monitoring Classification Based on Multi-Scale Electrical Appliance Load Signature
Published: December 31,2023Development of Control Framework Based on ROS Platform for a 3-Axis Gimbal
Published: December 31,2023Crop Disease Dataset and Recognition using Convolutional Neural Networks
-
1. Department of Information and Communication Engineering, Institute of Technology of Cambodia, Russian Federation Blvd., P.O. Box 86, Phnom Penh, Cambodia
Received: August 26,2022 / Revised: Accepted: April 20,2022 / Published: June 30,2023
Crop diseases, unfavorable growth, and nutritional deficiencies have a significant impact on the quality and quantity of agricultural income. According to the United Nations’ Food and Agriculture Organization, it is estimated that pre- and post-harvest diseases alone destroy at least 20–40% of global agricultural production. In developing countries like Cambodia, farmers tend to have a limited understanding of crop diseases and how to treat them, therefore AI solutions can assist farmers in detecting crop irregularities and diseases, which are now lacking in the agricultural sector. With the advancement and developments in AI Deep Learning Algorithms, this paper focuses on how disease dataset was collected and generated, as well as experimenting with four CNN models to detect and recognize the disease name across thirteen disease classes and evaluate the model performance such as accuracy, confusion matrix and computing resource consumptions that operate best with this dataset and can be integrated into mobile phone applications and microprocessor devices. The result of this research is a labeled crop disease dataset and experimental results, this dataset can achieve highest accuracy of 89.210%, 90.558%, 92.100%, 91.136% for InceptionV3, InceptionResNetV2, MobileNetV2 and EfficientNetB0, respectively.