This is gonna be the main service to face the client Side , its gonna be built using spring boot and posgresql
- Type of Work Project
- Supervisor: Maria João Varanda
- Co-Supervisor(s) Paulo Alves
- Date of proposal: December 2019
- Remarks This work will be developed by Benarous Farouq
Automatic Detection of Almond Variety based on image processing and artificial intelligence techniques
In the Trás-os-Montes region, where the IPB is located, is rich in almond production and several varieties have been detected. It is possible to identify to which variety an almond belongs based on the evaluation of a set of parameters. The idea will be to automate this process using almond images in various contexts and positions. This will require collecting previously identified images to train and test variety identification models using image processing techniques and artificial intelligence techniques such as deep learning. Work Methodology The methodology that will be adopted is based on the collection of a large number of images, their classification through the established parameters to be used by machine learning techniques, the choice of a model that can be trained with these images and which allows the identification of the almond variety in future images. For this purpose, will be used image processing tools, the python language and data science libraries.
- January – State-of-the art about similar problems and data collection (for training and testing);
- February – State-of-art about machine learning approaches and study of a specific deep learning architecture,
- March – Creation of a data set and image classification using deep learning;
- April – Model implementation using training data;
- May – Test the model with test data;
- June – Optimize the model;
- July - Write of the thesis.
Some experience in machine learning techniques and image processing.