激光与光电子学进展, 2020, 57 (10): 101015, 网络出版: 2020-05-08
基于深度学习和最大相关最小冗余的火焰图像检测方法 下载: 1448次
Flame Image Detection Method Based on Deep Learning with Maximal Relevance and Minimal Redundancy
图像处理 火焰检测 卷积神经网络 动态特征 最大相关最小冗余 image processing flame detection convolutional neural network dynamic feature maximal relevance and minimal redundancy
摘要
为了解决基于浅层特征的火焰识别模型对环境变化敏感且鲁棒性较低的问题,提出了一种基于卷积神经网络串行特征融合模型与最大相关最小冗余(MRMR)的火焰图像检测方法。为了从有限样本集中训练卷积神经网络获取更加全局性的特征,对使用预训练方法提取的火焰图像深层特征进行串行融合;再针对融合后的特征维度高、冗余大且未包含动态特征的问题,利用MRMR特征选择算法,去除与火焰相关性低的特征,获得相关性高的串行特征后与动态特征进行融合,得到最优子集的重构特征向量;最后通过支持向量机分类器完成对火焰目标的检测。实验结果表明,所提方法具有良好的泛化能力,对火焰的检测效果较好。
Abstract
A flame image detection method is proposed based on convolutional neural network using serial feature fusion model with maximal relevance and minimal redundancy (MRMR) to address the issue that the flame recognition model based on shallow features is susceptible to environmental changes and has low robustness. First, to obtain more global features from the finite sample set training convolutional neural network, the pre-training method was used to extract the deep features from the flame image for serial fusion. Then, to solve the problem of high dimensions of fusion feature, large redundancy, and lack of dynamic features after fusion, the MRMR feature-selection algorithm was used to remove features with low relevance to the flame, obtain highly relevant serial features, and merge with dynamic features to obtain a superior subset of the reconstructed feature vector. Finally, the flame target was detected using the support vector machine classifier. Experimental results show that the proposed method has good generalization ability and flame detection capability.
李梓瑞, 王慧琴, 胡燕, 卢英. 基于深度学习和最大相关最小冗余的火焰图像检测方法[J]. 激光与光电子学进展, 2020, 57(10): 101015. Zirui Li, Huiqin Wang, Yan Hu, Ying Lu. Flame Image Detection Method Based on Deep Learning with Maximal Relevance and Minimal Redundancy[J]. Laser & Optoelectronics Progress, 2020, 57(10): 101015.