An aerial vehicle detection method and detection system based on multi-scale small samples

Abstract

The invention discloses an aerial photography vehicle detection method based on multi-scale small samples, comprising: firstly using a data enhancement method to expand the collected data set, and then using a multi-scale adaptation algorithm to allow a deep learning model to extract common features for targets of different sizes; At the same time, small sample learning is used to extract shallow features to generate weighted feature parameters with small sample information; finally, the two parts of the features are combined and input into the subsequent deep learning model to obtain the detection frame, and the Gaussian mixture model method, classification confidence and soft intersection are combined. The final result is extracted by the Soft‑IoU algorithm. The present invention effectively solves the technical problems of multi-scale, small sample and high density encountered in aerial vehicle image detection through the above technical solutions.

Description

technical field [0001] The invention relates to the technical field of computer vision, and more particularly, to an aerial vehicle detection method based on multi-scale small samples. Background technique [0002] In recent years, the target detection algorithm based on deep learning is a very popular research direction in the field of computer vision. At present, the target detection algorithm based on deep learning is mainly divided into a one-stage regression-based detection algorithm and a two-stage candidate frame-based detection algorithm. Both types of algorithms are based on deep learning network technology. By feeding the optical camera image to the network model, the location of the preset classification in the optical image is detected. Object detection is an important science and technology in the field of artificial intelligence, which has received extensive attention from industry and academia. Artificial intelligence technology has achieved very good result...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the three major technical difficulties encountered in aerial vehicle image detection: multi-scale problems caused by aerial heights and angles; high-density single-target detection loss of many detection frames; aerial images cannot be trained using conventional data sets Small sample problem, the present invention proposes an aerial vehicle detection method based on multi-scale small samples

Images