Remote sensing ship identification method based on dense feature fusion and pixel-level attention

Abstract

The invention belongs to the field of image target recognition and provides a remote sensing ship identification method based on dense feature fusion and pixel-level attention, and aims to solve the problems that a classical neural network easily identifies a plurality of dense targets as one target under a remote sensing image ship target identification task, a large number of small targets are missed, boundary frames are easy to overlap and the like. According to the main scheme, data set division is carried out on a remote sensing image data set to obtain a training set and a test set, anddata enhancement of the training set is carried out. RGB three-channel average values r < mean >, g < mean > and b < mean > of the original remote sensing image data set are calculated and the RGB three-channel values of the images are correspondingly subtracted in the expanded data set from the r < mean >, g < mean > and b < mean >; the obtained data set is input into an improved Faster RCNN network to be trained, core modules of the network are a dense feature fusion network and a pixel-level attention network, and the network outputs candidate rotation boxes and category scores of the candidate rotation boxes; and skew IOU-based rotating frame non-maximum suppression is carried out on the obtained result to obtain an identification result of the ship target in the remote sensing image.

Description

technical field [0001] The invention relates to a remote sensing ship recognition method based on dense feature fusion and pixel-level attention, and belongs to the field of target recognition in remote sensing image processing. Background technique [0002] With the substantial increase in the real-time and operability of remote sensing technology, various remote sensing image products are developing towards the goal of multi-scale, multi-frequency, all-weather, high-precision, high-efficiency and fast. In the face of massive remote sensing images, manual interpretation is no longer competent, and data processing such as secondary information extraction and target recognition of remote sensing images has become more and more important, and has become the main research direction of remote sensing images. For the processing of remote sensing images It increasingly reflects the structure of the entire field and the most important measure of the level of software and hardware. ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to introduce a dense feature fusion network on the basis of the Fatser RCNN network for the problems of poor image quality, complex background, large scale span, extreme aspect ratio, and dense distribution of remote sensing images of ship targets. And pixel-level attention network and other improvement measures to overcome the limitations of classical neural networks in remote sensing image ship target recognition tasks, such as easy to identify multiple dense targets as one target, a large number of missed detection of small targets, and easy overlapping of bounding boxes. Improve recognition accuracy and robustness

Images