Principe J C, Kim M, Fisher M
Dept. of Electr. and Comput. Eng., Florida Univ., Gainesville, FL 32611, USA.
IEEE Trans Image Process. 1998;7(8):1136-49. doi: 10.1109/83.704307.
This paper addresses target discrimination in synthetic aperture radar (SAR) imagery using linear and nonlinear adaptive networks. Neural networks are extensively used for pattern classification but here the goal is discrimination. We show that the two applications require different cost functions. We start by analyzing with a pattern recognition perspective the two-parameter constant false alarm rate (CFAR) detector which is widely utilized as a target detector in SAR. Then we generalize its principle to construct the quadratic gamma discriminator (QGD), a nonparametrically trained classifier based on local image intensity. The linear processing element of the QCD is further extended with nonlinearities yielding a multilayer perceptron (MLP) which we call the NL-QGD (nonlinear QGD). MLPs are normally trained based on the L(2) norm. We experimentally show that the L(2) norm is not recommended to train MLPs for discriminating targets in SAR. Inspired by the Neyman-Pearson criterion, we create a cost function based on a mixed norm to weight the false alarms and the missed detections differently. Mixed norms can easily be incorporated into the backpropagation algorithm, and lead to better performance. Several other norms (L(8), cross-entropy) are applied to train the NL-QGD and all outperformed the L(2) norm when validated by receiver operating characteristics (ROC) curves. The data sets are constructed from TABILS 24 ISAR targets embedded in 7 km(2) of SAR imagery (MIT/LL mission 90).
本文探讨了使用线性和非线性自适应网络对合成孔径雷达（SAR）图像进行目标辨别。神经网络广泛用于模式分类，但这里的目标是辨别。我们表明这两种应用需要不同的代价函数。我们首先从模式识别的角度分析双参数恒虚警率（CFAR）检测器，它在SAR中被广泛用作目标检测器。然后我们推广其原理来构建二次伽马辨别器（QGD），这是一种基于局部图像强度进行非参数训练的分类器。QGD的线性处理元件通过非线性进一步扩展，产生一个多层感知器（MLP），我们将其称为NL-QGD（非线性QGD）。MLP通常基于L(2)范数进行训练。我们通过实验表明，不建议使用L(2)范数来训练MLP以辨别SAR中的目标。受奈曼-皮尔逊准则的启发，我们基于混合范数创建了一个代价函数，以不同方式权衡误报和漏检。混合范数可以很容易地纳入反向传播算法，并带来更好性能。应用了其他几种范数（L(8)、交叉熵）来训练NL-QGD，并且在通过接收者操作特征（ROC）曲线验证时，所有这些范数的性能都优于L(2)范数。数据集由嵌入在7平方公里SAR图像（麻省理工学院/林肯实验室任务90）中的24个逆合成孔径雷达（ISAR）目标的TABILS构建而成。
