SAR image target discrimination method based on weakly supervised learning

Abstract

The invention discloses an SAR image target discrimination method based on weakly supervised learning, which is mainly used to solve problems of a prior art such as low discrimination performance and high sample marking costs. The SAR image target discrimination method comprises steps that in a training phase, locality-constrained linear coding LLC characteristics of a positive image sample set and locality-constrained linear coding LLC characteristics of a negative image sample set are respectively extracted, and the negative sample set is used to train a potential latent dirichlet allocation LDA model, which is used to select an initial positive sample set from the positive image sample set to iteratively train a second-class SVM discriminator, and an optimal discriminator is acquired; in a testing phase, LLC characteristics of a testing sample set are extracted, and the acquired optimal discriminator is used to discriminate the testing sample set. The SAR image target discrimination method based on the weakly supervised learning is advantageous in that the discrimination performance is close to the fully-supervised SVM discriminator, and at the same time, costs of manual marking are reduced, and therefore practicability is provided; by comparing with a first-class SVDD discriminator of clutter training, the discrimination performance in a complicated scene is better, and the discrimination method provided by the invention is suitable for the SAR image target discrimination.

Description

technical field [0001] The invention belongs to the technical field of radar, in particular to a target identification method, which can be used to effectively identify targets in synthetic aperture radar SAR images. Background technique [0002] Radar imaging technology was developed in the 1950s, and has been developed by leaps and bounds in the next 60 years. At present, it has been widely used in military, agriculture, forestry, geology, ocean, disaster, mapping and many other fields. Synthetic aperture radar (SAR) has the characteristics of all-weather, all-time, high resolution and strong penetrating power, and has become an important means of earth observation and military reconnaissance. [0003] Automatic target recognition in SAR image is the frontier subject of current SAR application, which has important research significance and broad application prospect. The Lincoln Laboratory of the United States proposed a three-level processing flow chart for automatic tar...

AI Technical Summary

Problems solved by technology

Training a fully supervised two-class SVM discriminator requires a large number of labeled training samples. If the first step of detection is based on unsupervised, manual labeling of the obtained detection results is time-consuming and labor-intensive. At the same time, for fuzzy or Occluded samples may lead to wrong labels and affect the learning of the discriminator; a class of SVDD discriminators only need one class of data to learn. Usually, clutter data is easier to get, although it reduces the tediousness of manual labeling , but compared to the target, the clutter data is more complex and has more types, and the trained model cannot be well adapted to the identification of SAR image targets and clutter in complex scenes, which will affect the final identification performance

Images