Airport Target Automatic Recognition Method Based on Line Classification and Texture Classification

Abstract

The invention provides an airport target automatic identification method based on line classification and texture classification. The invention learns the multi-dimensional runway line feature of the airport target to obtain a line feature classifier, and filters the airport runway line through the line feature classifier without setting A large number of threshold conditions have wider applicability; screen the extracted straight line segments and determine the ROI area, instead of directly relying on a large amount of prior knowledge to determine the ROI area; according to the texture feature classifier learned from texture features, determine whether the ROI area is For the airport target, this method of integrated multi-classifier recognition effectively avoids the situation of airport target extraction errors caused by relying solely on the line feature classifier, and can effectively improve the accuracy of the automatic recognition of airport targets.

Description

Technical field [0001] The present invention belongs to the technical field of automatic target recognition, and particularly relates to an automatic recognition method of airport targets based on line classification and texture classification. Background technique [0002] As a common transportation facility and military facility, the airport has a very important position in economic construction and national defense construction. The automatic recognition of airport targets from remote sensing images has important practical value in the fields of aircraft automatic driving and airport positioning and navigation, and is a hot issue in the field of target recognition. At present, there are two main methods of airport target recognition. One is to determine the region of interest (ROI) of suspected airport targets through image segmentation or visual saliency mechanisms based on the gray features of the airport, and then use methods such as texture feature classification to verify...

AI Technical Summary

Problems solved by technology

[0003] In the above-mentioned airport target recognition methods, when processing through image segmentation or visual salience mechanism, it is difficult to achieve ideal results for images with uneven gray distribution and high resolution

However, the target recognition method based on the structural characteristics of the airport runway relies too much on prior knowledge, and it is difficult to guarantee its applicability.

Images