Monitorització de les activitats d’animals de granja mitjançant intel·ligència artificial
An important factor for farms earnings is to detect animals illness on its early stages. The sooner the disease is detected, the less cost it takes to treat it. One way to determine if a cow isn’t healthy is that the animal won’t drink water. So, it would be interesting to have a system to determine...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Fecha de publicación: | 2019 |
| País: | España |
| Institución: | Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya) |
| Repositorio: | Recercat. Dipósit de la Recerca de Catalunya |
| OAI Identifier: | oai:recercat.cat:10459.1/67736 |
| Acceso en línea: | http://hdl.handle.net/10459.1/67736 |
| Access Level: | acceso abierto |
| Palabra clave: | Inteligencia artificial Visión artificial Aprendizaje automático Intel·ligència artificial Visió artificial |
| Sumario: | An important factor for farms earnings is to detect animals illness on its early stages. The sooner the disease is detected, the less cost it takes to treat it. One way to determine if a cow isn’t healthy is that the animal won’t drink water. So, it would be interesting to have a system to determine which of the cows from the farm are not drinking water. Knowing this, we have implemented an object classifier using Convolution Neural Networks (CNNs) with Keras. The motivation of this project is understanding clearly about deep learning, particularly CNNs, and put in on real life. Therefore, we also tunned the hyper parameter of each models such as learning rate, batch size, and number of epochs. In addition, we also used techniques to optimize networks, acting as activation function, dropout and max pooling. During this process, several models have been generated in order to observe the relationship between number of layers, input data and accuracy. |
|---|