Wide remote sensing description generation method based on target detection

Abstract

The invention discloses a target detection-based wide remote sensing description generation method. The method comprises the steps of firstly obtaining a remote sensing image; constructing a trainingsample set and a test sample set, and normalizing the two sample sets; processing the remote sensing image by using a Faster-RCNN network model to obtain a corresponding target; clustering the targetsby using a K-means clustering algorithm; segmenting the wide remote sensing image according to the obtained clustering center; carrying out random segmentation on other unused areas; processing the segmented picture by using a ResNet101 network model; obtaining a corresponding picture description by using LSTM; and detecting whether the target detection result is in the description or not again to obtain a final result. According to the wide remote sensing image description method based on target detection, the description accuracy is improved, and information required by a user can be betterobtained.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image description generation, and in particular relates to a wide-width remote sensing description generation method based on target detection. Background technique [0002] With the development of satellite technology, modern aerospace remote sensing technology has greatly improved its ability to observe the earth. The obtained remote sensing data can play an important role in both military and civilian applications. In order to make full use of these remote sensing data and reduce the consumption of manpower and material resources, it is a very meaningful work to design a network to automatically understand remote sensing images and describe the content of remote sensing images accurately and fluently at the semantic level. However, in the field of high-resolution remote sensing, the main focus is still on target detection, object classification, semantic segmentation, etc. These studies ...

AI Technical Summary

Problems solved by technology

Moreover, in the process of collecting remote sensing images, it is easily affected by the surrounding environment such as illumination, occlusion, distance, etc., which will cause the characteristics of the target remote sensing image to be blurred, which makes it more difficult to understand the semantic level of remote sensing images.

Images