One-stage direction remote sensing image target detection method based on student-T distribution assistance

Abstract

The invention relates to a one-stage direction remote sensing image target detection method based on student-T distribution assistance, and solves the problem of frames in any direction by using a geometric method based on a horizontal frame. The method comprises the following steps: S1, converting a remote sensing image by using a geometric conversion method based on the horizontal frame; S2, extracting remote sensing image features; S3, carrying out the regression and classification on feature maps obtained from a Convolutional Neural Network (CNN) and a Feature Pyramid Network (FPN) respectively, and extracting the feature maps of the feature maps of the CNN and the FPN from the feature maps of the Feature Pyramid Network; S4, carrying out the result optimization and output: adopting student-T distribution as a result of joint distribution, synthesizing a classification branch and a regression branch, optimizing the one-stage direction detection model based on the student-T distribution, and outputting a target detection result. According to the student-T distribution assistance-based one-stage direction remote sensing image target detection method, the special rigidity and bird's-eye view characteristics of the remote sensing image target are fully utilized, and the CNN and FPN models based on deep learning and Gaussian distribution and inverse gamma distribution are adopted, so that more accurate remote sensing image target detection is realized.

Description

technical field [0001] The present invention relates to the field of intelligent processing of satellite remote sensing information, in particular to the target detection technology of remote sensing images of multispectral remote sensing data and Convolutional Neural Networks (CNN), specifically a one-stage orientation based on Student-T distribution assistance Object detection method in remote sensing images. Background technique [0002] The research of target detection is mainly applied to the target detection of remote sensing image data and to determine its category, but the complexity of remote sensing image background and the diversity of targets greatly increase the difficulty of target detection, especially the small and dense targets. Remote sensing images are usually taken from a bird's-eye view, which means that objects are always oriented arbitrarily, making them difficult to describe with horizontal bounding boxes. This problem can usually be solved with orie...

AI Technical Summary

Problems solved by technology

[0008] A one-stage directional remote sensing image object detection method based on Student-T distribution assistance, solves the problem of arbitrary directional borders by using a geometric method based on horizontal borders, and then performs the following steps:

Images