Remote sensing object interpretation method based on focusing weight matrix and variable-scale semantic segmentation neural network

Abstract

The invention discloses a remote sensing object interpretation method based on a focusing weight matrix and a variable-scale semantic segmentation neural network. The remote sensing object interpretation method comprises the following steps: obtaining and preprocessing data; making a thematic map; cutting the sample; designing a multi-spatial-scale remote sensing image labeling strategy; making alabel of the sample set; constructing a multi-scale remote sensing image semantic interpretation model; selecting a training set and a verification set; setting training parameters; training model; designing based on a remote sensing object recognition algorithm of a focusing weight matrix; and verifying and analyzing the effect of the variable-scale remote sensing image semantic interpretation model. According to the invention, the LSTM is constructed, so that the relation between nouns in the obtained semantic description and an object mask graph obtained by semantic segmentation is obtained, and the spatial relationship in semantic description is transferred between the object mask graphs, so that the variable-scale semantic segmentation of the remote sensing object and the end-to-end identification of the spatial relationship are realized, and the image classification and identification work in the remote sensing application field is guided to step forward to a higher step.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a remote sensing object interpretation method based on a focusing weight matrix and a variable-scale semantic segmentation neural network. Background technique [0002] Remote sensing image classification and remote sensing object recognition are the current research hotspots in remote sensing technology. With the development of artificial intelligence technology, deep neural network has been widely used in high-resolution remote sensing image analysis, and has become an effective processing method. [0003] At present, the traditional LSTM model based on the Attention mechanism is mainly applied to the semantic description of ordinary digital images. In the process of implementing the present invention, the inventor of the present application found that the method of the prior art has at least the following technical problems: [0004] Spatial location uncertainty: At d...

AI Technical Summary

Problems solved by technology

[0008] In view of this, the present invention provides a remote sensing object interpretation method based on focusing weight matrix and variable-scale semantic segmentation neural network, to solve or at least partially solve the technical problem of inaccurate recognition in the methods in the prior art

Images