Fast DDL Classification for SAR Images With ${l}_{1,\infty}$ Constraint

Abstract

Synthetic aperture radar (SAR) image classification aims at labeling pixels with different categories and this is both, a fundamental step for automatic target recognition (ATR) and a prerequisite for further interpretation. In the past decades, various methods have been proposed for the classification of SAR targets and among them are discriminative dictionary learning (DDL) methods. These DDL methods have recently gained attention from researchers' community due to the fact that they are very powerful on both, representation and discrimination during the classification process of SAR images. However, most of the existing DDL methods adopt l 0-norm or l 1-norm to ensure the sparsity, but in general, these DDL methods suffer from a high computational burden. Furthermore, it is important to minimize the execution time in the phase of online testing for the scenario of onboard real-time or near real-time SAR automatic target recognition such as modern unmanned aerial vehicle SAR platforms. That said, on reducing execution time, we are confronted with the problem of enhancing recognition efficiency while maintaining its accuracy. In order to solve this problem, our paper proposes a fast DDL method (named as FaDDL) based on a nonlinear analysis co-sparse model by adopting an l 1,∞-norm ball as a constraint to replace l 0-norm or l 1-norm on the coding coefficient matrix. The experimental results show that our proposed method significantly reduces execution time, without losing the classification accuracy. INDEX TERMS Discriminative dictionary learning, synthetic aperture radar, automatic target recognition, l 1∞-norm, nonlinear analysis co-sparse model.

Decision fusion using virtual dictionary‐based sparse representation for robust SAR automatic target recognition

“…noise contamination including multiplicative speckle noise and additive white Gaussian noise, limited training resource, resolution variance, partial occlusion etc. [7–47]. In practice, however, it seems that most crucial EOCs are adding speckle noise, different depression angles and a limited number of the training samples.…”

“…Considering the above facts, efforts have been made to develop SRC algorithms. A summary can be presented as follows: (i) using kernel method to transfer samples to new higher dimension spaces where classes can be linearly discriminated [29–31], (ii) utilising manifold learning [32–34], (iii) fusing SRC with other classification methods [24, 35], (iv) using l2‐norm [36, 37] or other norms [38, 39] instead of l1‐norm in SRC, (v) acquiring a dictionary via DL methods instead of using training samples could be very effective in the SR and SRC results [20, 40–42]. Based on the latter facts, and making use of the Fisher criterion, Zhang and co‐authors [43, 44] introduced Fisher discriminative DL (FDDL).…”

Обработка Сигналов От Пространственно Распределённых Целей

“…Метод разреженных представлений на примере обнаружения объектов базы MSTAR демонстрирует точность распознавания, превышающую 90 % [200], в сочетании со сравнительно высокой скоростью обработки, причем в ряде случаев оказываясь точнее, чем иные методы классификации с учителем, например, метод опорных векторов [11,196] и метод k ближайших соседей [30,186,201]. Как и для метода SVM, точность классификации растет с уменьшением количества распознаваемых классов [200].…”

Locality Constraint Dictionary Learning With Support Vector for Pattern Classification

“…To integrate structured dictionary learning, analysis representation and analysis classifier training into a unified framework, Zhang et al [48] proposed an analysis discriminative dictionary learning (ADDL) algorithm. Inspired by the superiority of 1,∞ norm [18], Wei et al [49] developed a fast DDL (FaDDL) method for synthetic aperture radar (SAR) image classification. The ordinal locality of analysis dictionary is not fully exploited in the above DPL and its variants, to tackle this problem, Li et al [50] proposed a discriminative low-rank analysissynthesis dictionary learning (LR-ASDL) algorithm with the adaptively ordinal locality.…”