中文摘要：

基于传统神经网络训练速度慢、易陷入局部极小值和泛化性能低等问题,提出采用极端学习机（ELM,extreme learning machine）对立体图像质量进行了客观评价。ELM是单隐层前馈神经网络（SLFNs）的泛化,输入权重可以随机赋值并通过解析获得输出权值。与传统神经网络算法相比,ELM算法具有参数选择简单、学习速度快及泛化性能好等优点。实验结果表明,以sigmoid为激励函数,对241幅不同等级的立体图像测试样本进行测试,其正确等级分类率达到93.85%。研究了不同激励函数条件下不同隐藏层节点数对极端学习机网络性能的影响,且将ELM和传统BP及支持向量机（SVM）在立体图像质量评价中的性能进行了分析比较。

英文摘要：

The stereoscopic image quality assessment has become one of the popular topics in the field of three-dimensional imaging. Due to the slow training speed, easily falling into local minimum and the low generalization in traditional neural network, the extreme learning machine （ELM） algorithm is presented for objective stereoscopic image quality assessment in this paper. ELM works for generalized single-hidden layer feedforward neural networks （SLFNs）, which randomly chooses the input weights and analytically determines the output weights of SLFNs. Compared with traditional neural network algorithms, ELM not only is easier to select the parameters,but also keeps the advantage of extremely fast Learning speed and achieves better generalization performance, and is widely applied in the field of function approximations and pattern recognition. Experimental results show that the correct classification rate of 241 different levels of test samples with sigmoid activation function is 93.85%. At the same time,the paper not only studies the effects of different hidden layer nodes for ELM in different activation functions, but also performs an analysis and comparison of the performance among ELM, traditional back propagation （BP） and support vector machine （SX/WI） in stereoscopic image quality assessment.