一种新型的城市火灾检测方法
|
摘要
在图像型火灾检测方法中,火灾特征的选取有一定的随机性和复杂性,仅仅依靠低层次的图像特征难以完整地描述复杂背景下的火灾图像。将深度学习技术应用到火灾检测领域,提出基于卷积神经网络的火灾检测方法,搭建包含3层全连接层的网络模型,使用Relu函数作为激活函数;然后基于Tensorflow平台实现该网络结构模型。在公开的火灾数据库上进行实验,结果表明,所提方法的火灾检测效果优于传统的图像型火灾检测算法,避免了由于选取特定火灾特征进行检测识别带来的局限性。
In image-based fire detection methods,the fire feature selection has a certain randomness and complexity,and it is difficult to completely describe the fire images in complex background by only relying on low level image features. Therefore,a fire detection method based on the convolutional neural network is proposed by applying the deep learning technology to the fire detection field. A network model including three full connection layers is built,taking the Relu function as the activation function. The network structure model is implemented based on the Tensorflow platform. An experiment was carried out on the public fire database. The experimental results show that the proposed method has a better fire detection effect in comparison with traditional image-based fire detection algorithms,which can avoid the limitations brought by detection recognition using selection of specific fire characteristics.
In image-based fire detection methods,the fire feature selection has a certain randomness and complexity,and it is difficult to completely describe the fire images in complex background by only relying on low level image features. Therefore,a fire detection method based on the convolutional neural network is proposed by applying the deep learning technology to the fire detection field. A network model including three full connection layers is built,taking the Relu function as the activation function. The network structure model is implemented based on the Tensorflow platform. An experiment was carried out on the public fire database. The experimental results show that the proposed method has a better fire detection effect in comparison with traditional image-based fire detection algorithms,which can avoid the limitations brought by detection recognition using selection of specific fire characteristics.
引文
[1]郑胤,陈权崎,章毓晋.深度学习及其在目标和行为识别中的新进展[J].中国图象图形学报,2014,19(2):175-184.ZHENG Yin,CHEN Quanqi,ZHANG Yujin.Deep learning and its new progress in object and behavior recognition[J].Journal of image and graphics,2014,19(2):175-184.
[2]胡勤,陈琛,刘敏.一种基于动态纹理的烟雾和火焰检测方法[J].消防科学与技术,2014,33(6):667-669.HU Qin,CHEN Chen,LIU Min.A smoke and fire detection method based on dynamic texture[J].Fire science and technology,2014,33(6):667-669.
[3]傅天驹,郑嫦娥,田野,等.复杂背景下基于深度卷积神经网络的森林火灾识别[J].计算机与现代化,2016(3):52-57.FU Tianju,ZHENG Chang’e,TIAN Ye,et al.Forest fire recognition based on deep convolutional neural network under complex background[J].Computer and modernization,2016(3):52-57.
[4]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems.New York:Curran Associates,2012:1097-1105.
[5]KIM Y.Convolutional neural networks for sentence classification[EB/OL].[2014-09-03].https://arxiv.org/pdf/1408.5882.pdf.
[6]KARPATHY A,TODERICI G,SHETTY S,et al.Large-scale video classification with convolutional neural networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Columbus:IEEE,2014:1725-1732.
[7]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C/OL].[2015-04-10].https://arxiv.org/pdf/1409.1556.pdf.
[8]章敏敏,徐和平,王晓洁,等.TensorFlow机器学习框架及应用[J].微型机与应用,2017,36(10):58-60.ZHANG Minmin,XU Heping,WANG Xiaojie,et al.Application of Google TensorFlow machine learning framework[J].Microcomputer&its applications,2017,36(10):58-60.
[9]KO B C,JUNG J H,NAM J Y.Fire detection and 3D surface reconstruction based on stereoscopic pictures and probabilistic fuzzy logic[J].Fire safety journal,2014,68:61-70.
[10]RAZAVIAN A S,AZIZPOUR H,SULLIVAN J,et al.CNNfeatures off-the-shelf:an astounding baseline for recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington:IEEE Computer Society,2014:512-519.
[11]卢英,王慧琴,柴茜,等.基于多尺度LBP与GBP特征的火焰识别[J].计算机科学,2015,42(10):316-320.LU Ying,WANG Huiqin,CHAI Qian,et al.Fire image detection based on LBP and GBP features with multi-scales[J].Computer science,2015,42(10):316-320.
[2]胡勤,陈琛,刘敏.一种基于动态纹理的烟雾和火焰检测方法[J].消防科学与技术,2014,33(6):667-669.HU Qin,CHEN Chen,LIU Min.A smoke and fire detection method based on dynamic texture[J].Fire science and technology,2014,33(6):667-669.
[3]傅天驹,郑嫦娥,田野,等.复杂背景下基于深度卷积神经网络的森林火灾识别[J].计算机与现代化,2016(3):52-57.FU Tianju,ZHENG Chang’e,TIAN Ye,et al.Forest fire recognition based on deep convolutional neural network under complex background[J].Computer and modernization,2016(3):52-57.
[4]KRIZHEVSKY A,SUTSKEVER I,HINTON G E.ImageNet classification with deep convolutional neural networks[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems.New York:Curran Associates,2012:1097-1105.
[5]KIM Y.Convolutional neural networks for sentence classification[EB/OL].[2014-09-03].https://arxiv.org/pdf/1408.5882.pdf.
[6]KARPATHY A,TODERICI G,SHETTY S,et al.Large-scale video classification with convolutional neural networks[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Columbus:IEEE,2014:1725-1732.
[7]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[C/OL].[2015-04-10].https://arxiv.org/pdf/1409.1556.pdf.
[8]章敏敏,徐和平,王晓洁,等.TensorFlow机器学习框架及应用[J].微型机与应用,2017,36(10):58-60.ZHANG Minmin,XU Heping,WANG Xiaojie,et al.Application of Google TensorFlow machine learning framework[J].Microcomputer&its applications,2017,36(10):58-60.
[9]KO B C,JUNG J H,NAM J Y.Fire detection and 3D surface reconstruction based on stereoscopic pictures and probabilistic fuzzy logic[J].Fire safety journal,2014,68:61-70.
[10]RAZAVIAN A S,AZIZPOUR H,SULLIVAN J,et al.CNNfeatures off-the-shelf:an astounding baseline for recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington:IEEE Computer Society,2014:512-519.
[11]卢英,王慧琴,柴茜,等.基于多尺度LBP与GBP特征的火焰识别[J].计算机科学,2015,42(10):316-320.LU Ying,WANG Huiqin,CHAI Qian,et al.Fire image detection based on LBP and GBP features with multi-scales[J].Computer science,2015,42(10):316-320.