Open AccesswUUNET: advanced fully convolutional neural network for multiclass fire segmentation
Author(s) -
Vladimir Sergeevich Bochkov,
Liliya Yurievna Kataeva
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1727/1/012003
Subject(s) - dice , computer science , segmentation , bounding overwatch , convolutional neural network , artificial intelligence , task (project management) , multiclass classification , cross entropy , pattern recognition (psychology) , binary number , artificial neural network , machine learning , mathematics , statistics , arithmetic , management , support vector machine , economics
This paper describes a solution of computer vision task concerning multiclass fire segmentation to get and show location of red, yellow and orange flame. We use UNet model as the best open-sourced convolutional neural network baseline. Based on this model we introduce UUNet-concatenative and wUUNet models. Since the multiclass fire segmentation task is solved for the first time in science, we collect the appropriate dataset and use the dataset-labeling alignment via look-up-tables. Also we compare models trained by Soft Dice and Jac-card indexes in combination with binary cross-entropy as a loss functions. Paper shows the problem of accuracy loss at bounding nodes of splitting the frame. As a solution we introduce combinational methods of partially intersected areas. The comparison of the used models and calculation schemes is demonstrated and the corresponding conclusions of the investigation are made.