Unlock instant, AI-driven research and patent intelligence for your innovation.

Optical remote sensing image statement description generation method based on scene pre-classification

A technology for optical remote sensing image and scene classification, applied in the field of automatic generation of optical remote sensing image sentence description, to achieve accurate description, strong interpretability, and wide application space

Active Publication Date: 2020-04-10
BEIHANG UNIV +1
View PDF6 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Its technical implementation faces greater difficulties and challenges in data set construction, model building, training and testing, etc.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Optical remote sensing image statement description generation method based on scene pre-classification
  • Optical remote sensing image statement description generation method based on scene pre-classification
  • Optical remote sensing image statement description generation method based on scene pre-classification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] In order to better understand the technical solution of the present invention, the embodiments of the present invention will be further elaborated below in conjunction with the accompanying drawings:

[0043] The system environment that the present invention relies on is Windows 10, and the hardware configuration mainly includes 7700K 4-core 8-thread CPU, GTX1080 (11GB video memory) GPU and 64GB internal memory; the software configuration mainly includes deep learning framework TensorFlow, programming language Python, and compiler PyCharm. In this environment, neural networks can be efficiently trained and tested. Specifically, first write the data interface, model, training function, and test function, then build the training data stream, perform model training and save according to the set macro parameters, and then reload the trained model to generate images Descriptive statement.

[0044] The overall model structure of the method for generating sentence descr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an optical remote sensing image statement description generation method based on scene pre-classification, and the technical core is to jointly optimize a statement descriptiongeneration task and a scene pre-classification task, thereby improving the data domain adaptability of a model. The method comprises the following four steps: 1, making a data set; 2, building a network model; 3, training a network model step by step; 4, performing remote sensing image statement description generation. According to the method, the defects in the prior art are overcome, the adaptability problem of a remote sensing image statement description model and a remote sensing data domain is well solved, the automation degree and the statement description content accuracy are high, andthe method has wide application prospects and huge use value.

Description

technical field [0001] The invention relates to a method for generating sentence descriptions of optical remote sensing images based on scene pre-classification, and belongs to the technical field of automatic generation of sentence descriptions for optical remote sensing images. Background technique [0002] Remote sensing (remote sensing) can be divided into active remote sensing and passive remote sensing. The former forms images by transmitting and recovering signals, while the latter forms images by receiving sunlight reflected by ground objects. Optical remote sensing (optical remote sensing) generally refers to the latter. In the range of ultraviolet and infrared bands, the technology of long-distance acquisition of ground object information through detection instruments. At present, optical remote sensing technology generally relies on space-borne and airborne sensors to detect ground objects. Affected by the height of the shooting equipment, shooting angle, and sens...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06V20/13G06N3/044G06N3/045G06F18/241G06F18/214
Inventor 史振威马小锋赵睿陈科研张宁韩传钊章泉源朱新忠张瑞珏
Owner BEIHANG UNIV