A Combined Method Based on SVM and Online Learning with HOG for Hand Shape Recognition
A Combined Method Based on SVM and Online Learning with HOG for Hand Shape Recognition
复制标题
DOI：
10.20965/jaciii.2012.p0687
复制
发表时间：
2012-09
期刊：
影响因子：
--
通讯作者：
Kazutaka Shimada;Ryo Muto;Tsutomu Endo
中科院分区：
文献类型：
--
作者：
Kazutaka Shimada;Ryo Muto;Tsutomu Endo
本文提出了一种手形识别的组合方法。它由支持向量机和基于感知器的在线学习算法组成。我们将HOG特性应用于每种方法。首先，我们的方法通过使用支持向量机估计输入图像的手形。这里，带有感知器的在线学习方法使用输入图像作为新的训练数据，如果该数据对于识别过程中的重新学习是有效的。接下来，我们利用支持向量机的得分从支持向量机和感知器的输出中选择最终的手形。该方法解决了用户变化时精度下降的问题。联合应用在线感知器，提高了精度。我们将组合方法与仅使用支持向量机的方法进行了比较。实验结果表明了该方法的有效性。
In this paper, we propose a combined method for hand shape recognition. It consists of support vector machines (SVMs) and an online learning algorithm based on the perceptron. We apply HOG features to each method. First, our method estimates a hand shape of an input image by using SVMs. Here the online learning method with the perceptron uses the input image as new training data if the data is effective for relearning in the recognition process. Next, we select the final hand shape from the outputs of the SVMs and perceptron by using the score of SVMs. The combined method deals with a problem about decrease of the accuracy in the case that users change. Applying the online perceptron jointly leads to improvement of the accuracy. We compare the combined method with a method using only SVMs. The experimental result shows the effectiveness of the proposed method.