论文部分内容阅读
作为一种新型的单隐层前馈型神经网络,极限学习机(Extreme Learning Machine:ELM)相比于传统的神经网络学习算法具有参数设置少、泛化性能强、训练和识别速度快等优点.为了有效提高交通标志的识别速度和识别率,提出一种基于加权ELM和AdaBoost融合优化的交通标志识别新算法.该算法通过迭代更新原始ELM的训练权重,并利用加权后的ELM作为AdaBoost的弱分类器,最终通过加权多数表决得到最优强分类器.最终实验结果表明,该算法能够取得的交通标志总识别率为99.12%,且单张交通标志的识别时间为7.1ms,可以满足实时识别应用的需求,较好的改善了交通标志的识别性能.
As a new type of single hidden layer feedforward neural network, Extreme Learning Machine (ELM) has the advantages of less parameter setting, extensive generalization performance, faster training and recognition speed than the traditional neural network learning algorithm In order to effectively improve the recognition speed and recognition rate of traffic signs, a new traffic sign recognition algorithm based on weighted ELM and AdaBoost fusion optimization is proposed, which updates iteratively the original ELM training weights and uses the weighted ELM as AdaBoost Finally, we obtain the optimal strong classifier by weighted majority vote.The final experimental results show that the total recognition rate of traffic signs obtained by this algorithm is 99.12%, and the recognition time of single traffic sign is 7.1ms, which can meet the real-time Identify the needs of applications, a better traffic sign recognition performance.