Multi-task processing system and method for two-stage remote sensing images

Abstract

The invention discloses a multi-task processing system and method for two-stage remote sensing images, and the system comprises a double-path branch feature extraction module, a pyramid fusion module,a semantic guidance module, a feature aggregation module, a double-path attention mechanism module, a boundary improvement module, and an up-sampling module. The invention carries out feature extraction of two-stage remote sensing images by constructing a double-channel network, an information flow module is designed to guide high-level semantic information to guide bottom-layer space informationto perform feature learning, and two stages of image feature maps are effectively fused through a feature aggregation module, so that the relevance and difference between two stages of images are fully captured. And subsequently, the expression capability of the feature information is enhanced through an attention mechanism module, so that multi-task intelligent processing is realized.

Description

technical field [0001] The invention relates to a technology in the field of image processing, in particular to a multi-task processing system and method for two-phase remote sensing images. Background technique [0002] With the successful application of deep learning in remote sensing image processing, the multi-task intelligent processing of remote sensing images has become a research hotspot. Classification, segmentation, change detection, and more. The existing deep network is introduced into the change category information extraction of the change area of ​​the two remote sensing images, mainly including Fully Convolutional-Earlyfusion (FC-EF) based on semantic segmentation, deepSiamese multi-scale convolutional network (DSMS-CN), etc., based on target detection Faster R-CNN, and AggregationNet based on two-way network and block-level detection method, etc. However, in addition to the changing regions of the two-period remote sensing images, the invariant regions als...

AI Technical Summary

Problems solved by technology

[0003] The defect of the existing technology is that when completing multi-tasks, it is necessary to send two phases of images to the model for classification, and compare the classification results, which greatly increases the time of the testing phase

On the other hand, this method analyzes the two-phase images as two separate objects, and cannot take advantage of the differences and correlations between the two-phase image features, so the implementation effect is poor.

Images