Compresion y analisis de imagenes por medio de algoritmos para la ganaderia de precision

TL;DR

This paper explores image compression algorithms combined with machine learning to analyze bovine images for health assessment, aiming to improve efficiency in cattle selection processes.

Contribution

It introduces the use of neural networks and bilinear interpolation for image compression in livestock health analysis, demonstrating improved execution speed.

Findings

Significant speed improvements with proposed algorithms

Effective identification of healthy vs. sick cattle images

Potential for further enhancement acknowledged

Abstract

The problem that we want to solve in this project of the subject of Data Structures and Algorithms, is to decipher some images, which have in them animals, being more specific, bovine animals; in which it is necessary to identify if the animal is healthy, that is to say, if it is in good conditions to be taken into account in the process of selection of the cattle, or if it is sick, to know if it is discarded. All this by means of an algorithm of compression, which allows to take the images and to take them to an examination of these in the code, where not always the results are going to be one hundred percent exact, but what allows this code to be efficient, is that it works with machine learning, which means that the more information it takes, the more precise the results are going to be without bringing with it general affectations. The proposed algorithms are NN and bilinear interpolation, where significant results were obtained on the execution speed. It is concluded that a better job could have been done, but with what was delivered, it is believed that it is a good result of the work.