Soil Type Classification Through Image Processing and Machine Learning This project is in development process for our Thesis Project.

A web-based software that classifies type of soil based on its physical properties using CNN. Outputs soil information and amendment for soil improvment.

An open-source software for geotechnical engineering analysis and modeling.

• Designed a classification model with 5 classes and 98% testing accuracy using a Convolutional Neural Network. • Applied Data Augmentation and reduced validation losses by 30% by applying MobileNetV2 Architecture.

An application for Farmers to recommend them the best types of crops which can be cultivated on a certain piece of land using Soil Image.

Agro companion is a soil classification, crop recommendation and crop information website. It helps identify seven types of soils, recommend 20+ crop based on specific factors and also help search information required to grow them.

Desarollo movido a https://github.com/ppizarror/granulometria-gui

soilDT: the soil classification app

Working repository for the code of the investigation of minimum soil sample masses