Logo Lanfrica

NourhanHassanEid/predictive-analysis-for-precision-farming

Domain:

agriculture
Creator:
Nou
Host:
This repository contains all the work that had been done in ONE LAB Egypt project 'Descriptive analysis on PA using ML'. # Predictive-analysis-for-Precision-Farming This repository contains all the work had been done in **ONE LAB Egypt** Descriptive analysis on PA using ML internship. The general scope of the proposed project lies in the field of agricultural development by using green electromagnetic-based techniques. In particular, the project focuses on the early prediction and control of the harmful green field living organisms by using **Machine Learning** models that capable to predict the numbers and conditions of these organisms then control the quasi-stationary magnetic fields. In this project, we target ##### The eggs of cotton leaf worm, scientifically known as Spodoptera littoralis. ##### Potato Blight. ##### Guava trees seasonal abundance of mealybug species and its associated predators and parasitoid . ### abstract ------------ Traditional ways of farming are no longer suitable for early and accurate detection of biotic stress. Recently, precision agriculture has been extensively used as a potential solution for agriculture problems using high resolution sensors and data analysis. In this paper, several methods of machine learning have been utilized in order to study pests' population, and agricultural conditions for some crops such as potatoes, guava, and cotton, which are among the main Egyptian crops. ### Dataset and setup The datasets used in this work for guava trees, cotton leafworm are a research study by Dr. Haitham Sharaf a professor at Cairo University, faculty of Agriculture, which contains data of weather conditions inside a controlled greenhouse system where Guava trees, cotton are planted. Seasonal abundance of mealybug species and their associated predators and parasitoid on guava trees in Egypt have been surveyed. The survey has been conducted in Giza, Egypt spanning two years (Jan. 2014 to Dec. 2015). Fifteen plants have been randomly chosen and five leaves have been biweekly collected, Each leave has been picked either from the middle of the in …