Infrared remote sensing image ship detection method

Abstract

The invention belongs to the field of target identification in remote sensing image processing, provides an infrared remote sensing image ship detection method, and solves the problem that an existingdetection method based on artificial design features is complex in process, low in robustness, narrow in application range and the like and the problem that a large amount of missing detection and false detection exist due to small ship targets in a complex marine environment for a method based on deep learning infrared remote sensing images. According to the main scheme, the method comprises thesteps that an infrared remote sensing ship data set is created, offline data enhancement is conducted, and the data set is divided into a training set, a verification set and a test set; and the training set obtained by division is input into an improved RefineDet detection network to carry out training. The core of the detection network is a basic feature extraction network ResNet101 and a feature pyramid feature fusion module FPN combined with a visual saliency module CBAM. The detection process is a two-step cascade regression mode, the category recognition and position regression precision of ship targets can be improved, and a final ship target detection result can be obtained by outputting the category confidence coefficient and the regression frame after threshold filtering by thenetwork and performing non-maximum suppression.

Description

technical field [0001] An infrared remote sensing image ship detection method is used for infrared remote sensing image imaging ship target detection and belongs to the field of target detection in remote sensing and infrared image processing. Background technique [0002] The target detection task is to find out all the specific targets in the image, and determine their position and size. Since the targets in the image or video have different sizes and are easily affected by factors such as illumination, occlusion and similar object interference, the target detection problem It has always been a very challenging task in the field of machine vision. Object detection is widely used in monitoring equipment, face recognition, automatic driving, national defense and security and other fields. [0003] The task of object detection in remote sensing images is to mark one or more objects of interest in a given aerial or satellite image. Among them, ship detection is a research ho...

AI Technical Summary

Problems solved by technology

[0007] The object of the present invention is: the present invention provides a kind of ship detection method of infrared remote sensing image, overcomes the problem that the existing method can not realize real-time and accurate positioning of ship target position under the trade-off between detection accuracy and speed of remote sensing image ship problem, it can improve the detection ability in the case of real-time detection

Images