Our research focuses on the question of classifiers that are capable of processing images rapidly and accurately\nwithout having to rely on a large-scale dataset, thus presenting a robust classification framework for both facial\nexpression recognition (FER) and object recognition. The framework is based on support vector machines (SVMs) and\nemploys three key approaches to enhance its robustness. First, it uses the perturbed subspace method (PSM) to\nextend the range of sample space for task sample training, which is an effective way to improve the robustness of a\ntraining system. Second, the framework adopts Speeded Up Robust Features (SURF) as features, which is more\nsuitable for dealing with real-time situations. Third, it introduces region attributes to evaluate and revise the\nclassification results based on SVMs. In this way, the classifying ability of SVMs can be improved.\nCombining these approaches, the proposed method has the following beneficial contributions. First, the efficiency of\nSVMs can be improved. Experiments show that the proposed approach is capable of reducing the number of samples\neffectively, resulting in an obvious reduction in training time. Second, the recognition accuracy is comparable to that\nof state-of-the-art algorithms. Third, its versatility is excellent, allowing it to be applied not only to object recognition\nbut also FER.
Loading....