This discussion has two purposes: The first purpose consists in the analysis of the behavior of a neural network for the segmentation of images of colon polyps, which is the process of accurately delineating and discriminating the region that identifies the polyp from the background. Particularly, it will be studied the behavior of the model as it changes the loss function. The second one, on the other hand, consists in implementing and testing new loss functions in order to improve the performance of the network itself.
Questa trattazione si pone due obiettivi: Il primo obiettivo consiste nell’analisi del comportamento di una rete neurale volta alla segmentazione di immagini di polipi al colon, ovvero il processo di delineare e discriminare accuratamente la regione che identifica il polipo dallo sfondo. In particolare, ne verrà studiato il comportamento al variare della funzione di loss. Il secondo, invece, consiste nell’implementare e testare nuove loss function al fine di migliorare le prestazioni della rete stessa.
Loss function per image segmentation
LORENZON, NICOLA
2021/2022
Abstract
This discussion has two purposes: The first purpose consists in the analysis of the behavior of a neural network for the segmentation of images of colon polyps, which is the process of accurately delineating and discriminating the region that identifies the polyp from the background. Particularly, it will be studied the behavior of the model as it changes the loss function. The second one, on the other hand, consists in implementing and testing new loss functions in order to improve the performance of the network itself.File | Dimensione | Formato | |
---|---|---|---|
Lorenzon_Nicola.pdf
accesso aperto
Dimensione
4.45 MB
Formato
Adobe PDF
|
4.45 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/10496