Please use this identifier to cite or link to this item:
https://sci.ldubgd.edu.ua/jspui/handle/123456789/3500Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.author | Maksymiv, O.P. | - |
| dc.contributor.author | Rak, T.Je. | - |
| dc.contributor.author | Menshykova, Olha | - |
| dc.date.accessioned | 2017-02-01T12:55:21Z | - |
| dc.date.available | 2017-02-01T12:55:21Z | - |
| dc.date.issued | 2016 | - |
| dc.identifier.uri | http://hdl.handle.net/123456789/3500 | - |
| dc.description.abstract | Paper describes the developed approaches to detecting emergency situations that primed based on color segmentation, frame difference and deep convolutional networks. The main objective was to test the interaction of computer vision traditional methods, combined with modern methods of machine learning. Experimentally proved that detection quality for the combination of such methods is 96.7%. In this work, in particular, was developed own images dataset with emergencies and conducted comparison of neural networks AlexNet and GoogLeNet. | uk |
| dc.publisher | м.Львів | uk |
| dc.subject | detection; convolutional neural network; machine learning; emergency situations | uk |
| dc.title | DEEP CONVOLUTIONAL NETWORK FOR DETECTING PROBABLE EMERGENCY SITUATIONS | uk |
| Appears in Collections: | 2016 2016 | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| -download=true.pdf | Тези | 3.05 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.