The Smart Agriculture based on Reconstructed Thermal Image

JITCE (Journal of Information Technology and Computer Engineering), Journal Year: 2022, Volume and Issue: 6(01), P. 8 - 13

Published: March 31, 2022

The utilization of thermal image in supporting precision agriculture is tremendous nowadays. There are many applications images agricultural fields, such as detecting crop water stress, monitoring free-range rabbits, measuring canopy temperature and so on. Furthermore, the importance camera became urgent need perform smart agriculture. Otherwise, price very expensive todays. Then, this kind not easy to find market. Therefore, it makes implementation difficult. In order handle problem, proposed method intends generate from visible images. Further, information concerning with agriculture, especially fertility leaves paddy fields stress can be monitored. uses deep learning architecture learn dataset. It applies Generative Adversarial Network architecture. This GAN pre-trained model trained using 150 training dataset tested testset. obtained used for generating results show constructed has high accuracy. assessment metric SSIM PSNR methods. Their indexes that have visual shows reconstructed also precision. Finally, implemented purposes.

Language: Английский

The High Efficiency of Induction Motor Failure using Predictive Maintenance with Deep Learning Model

International Journal of Research Publication and Reviews, Journal Year: 2023, Volume and Issue: 4(5), P. 1279 - 1283

Published: May 4, 2023

The rotary parts failure is very common problem in industrial world.It includes induction motors.The damaged motors make the process halted.This situation caused high loses.In order to handle this problem, implementation of predictive maintenance one optimal solution.The task able scheduling frequency and prevent incidental unpredicted machine failure, especially various data type have used, such as analog sensor, thermal data, image so on.Accordingly, methods also varied mathematic computation up smart or intelligent computation.The proposed method uses a kind deep learning architecture, Residual Network Resnet 50 model.The dataset selected image, motor with defect rate.It 10% 30 % respectively.The obtained prediction accuracy 100%.It means every true.Therefore, predict rate motor.It will help operator plan cost repairing suitable operational duration accurately.This use many sensors complicated electronic circuit.This easy accurate implement world.

Language: Английский

Implementation of Environmental Monitoring and Controlling for The Oyster Mushroom Based on The Internet of Things

Published: Dec. 8, 2023

The oyster mushroom is an edible fungus that requires precision environmental conditions in its growth. critical factors Oyster mushrooms' growth are temperature, humidity, and light. Farmers can use the Internet of Things (IoT) to gain farming. IoT a physical device collect data with sensors transmit it cloud over internet. This paper proposes IoT-based farming system for control mushrooms. proposed consists sensors, microcontrollers, actuators, virtual instrument LabView. get data, send microcontroller be processed, LabView's server. server equipped interface help visualize farmer. farmer adjust some thresholds. Thus, device's actuator take action maintain optimal farm. method promising improve efficiency profitability cultivation.

Language: Английский