Javascript must be enabled for the correct page display

Deep Architectures using the Bag of Words Model for Object and Handwritten Character Recognition

Groefsema, M. (2016) Deep Architectures using the Bag of Words Model for Object and Handwritten Character Recognition. Bachelor's Thesis, Artificial Intelligence.

[img]
Preview
Text
AI_BA_2016_MarcGroefsema.pdf - Published Version

Download (524kB) | Preview
[img] Text
Toestemming.pdf - Other
Restricted to Backend only

Download (475kB)

Abstract

This thesis describes an image classification system, using feature extraction, the Bag of Visual Words model (BoW), a Deep Bag of Visual Words model (DBoW) and a Support Vector Machine (SVM) classifier. The performance of the system using the BoW model is compared to the performance using the DBoW model. The performances are assessed using 10-fold cross-validations on the MNIST and CIFAR-10 datasets. First different configurations are explored of both the BoW model and the DBoW model. Finally, both architectures are compared based on their best performances. Results show a lower performance by using the DBoW architecture compared to the performance using the BoW model.

Item Type: Thesis (Bachelor's Thesis)
Degree programme: Artificial Intelligence
Thesis type: Bachelor's Thesis
Language: English
Date Deposited: 15 Feb 2018 08:13
Last Modified: 15 Feb 2018 08:13
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/14115

Actions (login required)

View Item View Item