A method for automatic identification of military and civilian attributes of ship targets in optical remote sensing images

Abstract

The invention discloses a method for automatically discriminating military and civilian attributes of a ship target in optical remote sensing images, and relates to the field of automatic processing of optical remote sensing images. The present invention constructs the ship target by obtaining training samples of the ship target, and performing a series of operations on the training samples such as manual identification of military and civilian attributes, normalized resampling, generation of collections, calculation of average images, calculation of difference images, and calculation of orthogonal vectors. The military-civilian attribute discrimination feature space of the test image is projected into this space, and the military-civilian attribute of the ship target in the test image can be automatically discriminated by distance measurement. This method only needs to construct the military-civilian attribute discrimination feature space of the ship target once, and then use this feature space to automatically identify the military-civilian attribute of the ship target found in the test image in future work. Through this method, the military-civilian attribute of the ship target in the newly discovered optical remote sensing image can be automatically discriminated, which liberates the brainpower of professional interpreters and greatly improves the work efficiency of ship target recognition.

Description

technical field [0001] The invention relates to the field of automatic processing of optical remote sensing images, in particular to an automatic identification method for military and civilian attributes of ship targets in optical remote sensing images. Background technique [0002] Satellite remote sensing technology is an effective means to detect ship targets at sea. When the spatial resolution of remote sensing images is low, only ship targets can be found; when the spatial resolution of remote sensing images is high, not only ship targets can be found It can also distinguish and identify the attributes and types of ship targets. [0003] According to the characteristics of remote sensing images, researchers have proposed various ship target detection algorithms. Generally speaking, most of them are based on the difference between the gray level of the target and the gray level of the background. Such as target extraction method based on minimum error threshold segment...

AI Technical Summary

Problems solved by technology

[0006] In view of this, the main purpose of the present invention is to propose an automatic military-civilian attribute discrimination algorithm for a ship target through manual screening of a large number of ship targets obtained in optical remote sensing to find out the target of interest. Ship targets are automatically divided into military ships and civilian ships, so as to meet the target observation needs of different military and civilian users, and narrow the scope for further fine identification of ship targets

Images