Ba


Methods of image modification
for neural network image classification

Introduction

Methods of image modification for neural network image classification

SK: Metódy modifikácie obrázkov určených na klasifikáciu neurónovou sieťou

Student: Martina Kuchtová

Tutor: RNDr. Miroslav Opiela, PhD.

Goals

Research

To review image classification tasks solved by neural networks, input datasets and preprocessing methods applied on images.

Implementation

To propose and implement methods for image preprocessing with a goal to improve the accuracy of image classification on the specified task.

Evaluation

To evaluate the impact of proposed preprocessing methods on the selected neural network with focus on the classification accuracy.

DOCUMENTS

PRESENTATION


December 2022


PAPER


SSBa (January 2023)


PAPER (ŠVK)


in process


PRESENTATION

(SSBb)


in process


CONTACT








Location:

Institute of Computer Science, Faculty of Science

University of Pavol Jozef Šafárik in Košice