A comparison study on nonlinear dimension reduction methods with kernel variations: Visualization, optimization and classification
A comparison study on nonlinear dimension reduction methods with kernel variations: Visualization, optimization and classification
复制标题
DOI：
10.3233/ida-194486
复制
发表时间：
2020-01-01
影响因子：
1.7
通讯作者：
Wong, Samuel W. K.
中科院分区：
文献类型：
--
作者：
Kempfert, Katherine C.;Wang, Yishi;Wong, Samuel W. K.
由于数据的高维性、协变量之间的相关性以及数据中包含的噪声，降维（DR）技术经常被应用于机器学习算法的应用中。主成分分析 (PCA)、线性判别分析 (LDA) 及其核变体（KPCA、KLDA）是最流行的 DR 方法。最近，监督核主成分分析（SKPCA）被证明是另一种成功的替代方法。在本文中，首先对这些流行技术进行简要回顾。然后，我们基于三个模拟数据集进行性能比较研究，然后通过应用于人脸图像分析中的模式识别问题来评估技术的性能。在 MORPH-II 和 FG-NET 这两个流行的纵向人脸老化数据库上考虑了性别分类问题。使用了多种特征提取方法，包括生物启发特征 (BIF)、局部二值模式 (LBP)、定向梯度直方图 (HOG) 和主动外观模型 (AAM)。应用DR方法后，部署线性支持向量机（SVM），在MORPH-II上性别分类准确率超过95%，与基准结果具有竞争力。还提出了一种并行计算方法，在 MORPH-II 上获得更快的处理速度和相似的识别率。我们的计算方法可以应用于实际的性别分类系统，并推广到其他面部分析任务，例如种族分类和年龄预测。
Because of high dimensionality, correlation among covariates, and noise contained in data, dimension reduction (DR) techniques are often employed to the application of machine learning algorithms. Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), and their kernel variants (KPCA, KLDA) are among the most popular DR methods. Recently, Supervised Kernel Principal Component Analysis (SKPCA) has been shown as another successful alternative. In this paper, brief reviews of these popular techniques are presented first. We then conduct a comparative performance study based on three simulated datasets, after which the performance of the techniques are evaluated through application to a pattern recognition problem in face image analysis. The gender classification problem is considered on MORPH-II and FG-NET, two popular longitudinal face aging databases. Several feature extraction methods are used, including biologically-inspired features (BIF), local binary patterns (LBP), histogram of oriented gradients (HOG), and the Active Appearance Model (AAM). After applications of DR methods, a linear support vector machine (SVM) is deployed with gender classification accuracy rates exceeding 95% on MORPH-II, competitive with benchmark results. A parallel computational approach is also proposed, attaining faster processing speeds and similar recognition rates on MORPH-II. Our computational approach can be applied to practical gender classification systems and generalized to other face analysis tasks, such as race classification and age prediction.