Comparison of Different Convolutional Neural Network Architectures for Satellite Image Segmentation

Convolutional neural networks for detection geo-objects on the satellite images from DSTL, Landsat -8 and PlanetScope databases were analyzed. Three modification of convolutional neural network architecture for implementing the recognition algorithm was used. Images obtained from the Landsat -8 and PlanetScope satellites are used for estimation of automatic object detection quality. To analyze the accuracy of the object detection algorithm, the selected regions were compared with the areas by previously marked by experts. An important result of the study was the improvement of the detector for the class 'Forest'. Segmentation of satellite images has found application at urban planning, forest management, climate modelling, etc. © 2018 FRUCT Oy.

Авторы
Khryashchev V.1 , Ivanovsky L.1 , Pavlov V.1 , Ostrovskaya A. 2 , Rubtsov A.3
Редакторы
-
Сборник материалов конференции
Издательство
IEEE Computer Society
Номер выпуска
-
Язык
Английский
Страницы
172-179
Статус
Опубликовано
Подразделение
-
Ссылка
-
Номер
8588071
Том
2018-November
Год
2018
Организации
  • 1 P.G. Demidov Yaroslavl State University, Yaroslavl, Russian Federation
  • 2 People's Friendship University of Russia (RUDN University), Moscow, Russian Federation
  • 3 Russian Space Systems, Moscow, Russian Federation
Ключевые слова
Convolution; Forestry; Image segmentation; Neural networks; Object detection; Object recognition; Satellites; Convolutional neural network; GEO objects; LANDSAT; Object detection algorithms; Recognition algorithm; Satellite images; Network architecture
Дата создания
19.07.2019
Дата изменения
19.07.2019
Постоянная ссылка
https://repository.rudn.ru/ru/records/article/record/38271/