FIELD: physics.
SUBSTANCE: invention relates to computer engineering and can be used for monitoring subsoil management areas of open type. System comprises a set of computer facilities based on a convolution neural network using data from spacecrafts for remote probing of the Earth. Convolutional neural network processes images of forest fund objects received from spacecraft for remote probing of the Earth, with formation by means of a sliding window method of rectangular matrixes of pixels of an image with a given pitch - crosses. For each rectangular matrix, by turning and mirroring, eight displays are generated with segmentation of each display and its classification by means of a convolution neural network as one of the subsoil areas of open type and obtaining a final image classified as one of the open-pit subsoil areas.
EFFECT: technical result is faster processing of data and fewer computational resources.
1 cl, 3 dwg
Title | Year | Author | Number |
---|---|---|---|
INTELLIGENT SPACE SYSTEM FOR MONITORING BUILDINGS AND STRUCTURES | 2018 |
|
RU2707138C1 |
INTELLIGENT SPACE SYSTEM FOR MONITORING FOREST RESOURCES | 2019 |
|
RU2703349C1 |
SYSTEM FOR SEGMENTING IMAGES OF SUBSOIL RESOURCES OF OPEN TYPE | 2019 |
|
RU2733823C1 |
SYSTEM FOR SEGMENTING IMAGES OF BUILDINGS AND STRUCTURES | 2019 |
|
RU2734058C1 |
METHOD FOR MAPPING INUNDATED FLOODPLAIN WETLANDS FROM SATELLITE MULTISPECTRAL DATA OF MEDIUM SPATIAL RESOLUTION | 2024 |
|
RU2839570C1 |
2021 |
|
RU2769575C1 | |
METHOD FOR MULTISPECTRAL RECONSTRUCTION OF NIGHT IMAGES OF URBAN AGGLOMERATIONS | 2024 |
|
RU2838134C1 |
METHOD FOR REMOTE DETERMINATION OF DYNAMICS OF PYROGENIC VEGETATION SUCCESSIONS | 2021 |
|
RU2814455C2 |
METHOD OF SEARCHING FOR UNDERGROUND WATER | 2011 |
|
RU2465621C1 |
IMAGE PROCESSING SYSTEM AND METHOD USING DEEP NEURAL NETWORKS | 2018 |
|
RU2743931C1 |
Authors
Dates
2020-04-02—Published
2018-12-21—Filed