Infrared remote sensing image sea ship detecting method based on significance analysis

Abstract

The invention discloses an infrared remote sensing image sea ship detecting method based on significance analysis. The method can be used for sea ship detection in infrared remote sensing images in aerospace and aviation. The method comprises the following steps of: dividing an infrared remote sensing image to obtain sea regions; in the divided sea regions, detecting candidate targets which can be ships on a sea surface by a method based on the significance analysis; carrying out primary filtering on the candidate targets according to size information; carrying out secondary filtering on the candidate targets according to shape information; and determining the candidate target which passes the limit on the size and shape as the sea ship which is obtained by final detection. By combining image division and significance analysis techniques, the method disclosed by the invention realizes the division of the sea regions in the infrared remote sensing image and the detection on the sea ships, and avoids the defects of relatively narrow application range and poor detection performance of a single target detection algorithm.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a method for detecting ships on the sea surface of infrared remote sensing images based on saliency analysis. Background technique [0002] In modern warfare, information dominance is a key factor affecting the overall strategy, and imaging reconnaissance and target detection and recognition are the main ways to obtain information. As one of the existing multiple imaging reconnaissance methods, the infrared imaging sensor is only sensitive to the radiation of the target scene (mainly determined by the emissivity and temperature difference of the target scene), but not sensitive to the brightness change of the scene. When the target has a large temperature gradient or the background has a large thermal contrast with the target, the low-visibility target is easy to see in the infrared image, and it has certain advantages in target detection, especially in the detection of ...

AI Technical Summary

Problems solved by technology

[0004] Various ground objects and targets in infrared remote sensing images do not have fixed grayscale properties. Different ground objects and targets change with time, and the temperature changes differently, which brings certain difficulties to target detection. Moreover, it is affected by weather, The comprehensive influence of various factors such as light and sea conditions makes it difficult to distinguish the ship target from the background sea surface

Although some researchers have carried out research on the detection of ships on the sea surface in infrared remote sensing images in recent years, and achieved certain results, there is still a long way to go before practical application.

Images