Automatic combustible recognition method for surface fire model based on field pictures

The invention relates to an automatic combustible recognition method for a surface fire model based on field pictures. The method comprises the steps of combustible classification model generation, combustible automatic classification and combustible area map generation. The method comprises the fol...

Full description

Saved in:
Bibliographic Details
Main Authors LI JIANWEI, CAO LUYANG
Format Patent
LanguageChinese
English
Published 10.12.2021
Subjects
CALCULATING
COMPUTING
COUNTING
HANDLING RECORD CARRIERS
PHYSICS
PRESENTATION OF DATA
RECOGNITION OF DATA
RECORD CARRIERS
Online AccessGet full text

Cover

More Information
Summary:The invention relates to an automatic combustible recognition method for a surface fire model based on field pictures. The method comprises the steps of combustible classification model generation, combustible automatic classification and combustible area map generation. The method comprises the following steps: training a convolutional neural network by inputting the collected data and the marked combustible data, generating a combustible classification model, classifying different combustible data by using the model, outputting a corresponding combustible category probability map, performing color marking and region merging on different combustible categories to generate a complete ground surface combustible category region map, and superposing the ground surface combustible category region map with a real ground surface image to obtain a ground surface real combustible category region image. According to the method, real-time, automatic and intelligent classification of surface combustibles can be realized
Bibliography:Application Number: CN202111073099