摘要
针对图像烟雾识别存在高误报率和高错误率的问题,提出一种新的结合聚合Gabor核和局部二值模式(LBP)的烟雾识别方法.首先使用加权平均法生成聚合Gabor核,以减少Gabor核的数目,并将聚合Gabor核和原图像进行卷积,生成聚合Gabor特征图.为了增强分类性能,本文提出了一个自定义比较函数,用于原图像和Gabor特征图的局部二值编码.然后生成3种LBP映射模式的特征,并串联这3种特征得到LBP串联模式特征.最后串联所有的LBP串联模式特征作为原图像的局部聚合Gabor二值模式(LAGBP)特征.与传统方法的比较实验显示,在3个测试集上,本文方法都分别要优于传统方法,且具有最低的错误率;与深度学习方法的比较实验中,本文方法也有很好的表现.实验表明LAGBP方法具有很好的烟雾特征表达能力,非常适用于烟雾识别.
To reduce false alarm and error rates,we propose a novel smoke recognition method by combining aggregated Gabor kernels and Local Binary Patterns( LBPs). First,aggregated Gabor kernels are generated using the weighted sum of original Gabor kernels,and then aggregated Gabor feature maps are produced by convolving the aggregated Gabor kernels with an image. To enhance classification performance,we propose a self-defined comparison function,which is used to encode the original image and aggregated Gabor feature maps. Then we compute three histograms of each LBP code with three LBP modes,and concatenate them to produce an LBP mode connection histogram. Finally,we concatenate all LBP mode connection histograms to achieve a histogram of Local Aggregated Gabor Binary Pattern( LAGBP). Comparison experiments show that the proposed method outperforms traditional methods and has the lowest error rates among them on all test sets. In comparison experiments with several deep learning methods,our method also achieves an outstanding performance. Therefore,the proposed method has a good discriminative ability and is suitable for smoke recognition.
引文
[1]Luo Sheng,Jiang Yuzheng.State-of-art of video based smoke detection algorithms[J].Journal of Image and Graphics,2013,18(10):1225-1236.
[2]Yuan F N.Video-based smoke detection with histogram sequence of LBP and LBPV pyramids[J].Fire Safety Journal,2011,46(3):132-139.
[3]Tian H D,Li W Q,Wang L,et al.A novel video-based smoke detection method using image separation[C].IEEE International Conference on M ultimedia and Expo.M elboume,VIC,Australia,IEEE,2012:532-537.
[4]Pan Chao-feng,Yang Shu-seng,Chen Ni.Fire smoke recognition based on image entropy[J].Journal of Jiangsu University of Science and Technology,Natural Science Edition,2015,29(1):52-57.
[5]Yuan F,Shi J,Xia X,et al.Sub oriented histograms of local binary patterns for smoke detection and texture classification[J].KSIITransactions on Internet and Information Systems,2016,10(4):1807-1823.
[6]Yuan F,Shi J,Xia X,et al.High-order local ternary patterns with locality preserving projection for smoke detection and image classification[J].Information Sciences,2016,372(c):225-240.
[7]Zhao Liang,Luo Yan-min,Luo Xiang-yu.Based on dynamic background update and dark channel prior of fire smoke detection algorithm[J].Application Research of Computers,2017,34(3):957-960.
[8]Kamaruzaman F,Shafie A A.Recognizing faces with normalized local gabor features and spiking neuron patterns[J].Pattern Recognition,2016,53(C):102-115.
[9]Zhang Hong-tao,Ma Yan,Zhang Xiang-fen,et al.High discriminative SIFT and LGQP feature and its application to face recognition[J].Journal of Chinese Computer Systems,2017,38(11):2488-2493.
[10]Li C,Huang Y,Zhu L.Color texture image retrieval based on Gaussian copula models of Gabor wavelets[J].Pattern Recognition,2017,64(c):118-129.
[11]Kumar M,Singh K M.Retrieval of head-neck medical images using Gabor filter based on power-lawtransformation method and rank BHM T[J].Signal Image and Video Processing,2018,12(5):827-833.
[12]Shi Y,Yang X,Guo Y.Translation invariant directional framelet transform combined with Gabor filters for image denoising[J].IEEE Transactions on Image Processing,2014,23(1):44-55.
[13]Zhang W,Shan S,Gao W,et al.Local Gabor binary pattern histogram sequence(LGBPHS):a novel non-statistical model for face representation and recognition[C].10th IEEE International Conference on Computer Vision.IEEE Computer Society,2005:786-791.
[14]Zhang Wen-chao,Shan Shi-guang,Chen Xi-lin,et al.Local gabor binary patterns based on mutual information for face recognition[J].International Journal of Image&Graphics,2007,7(4):777-793.
[15]Gabor D.Theory of communication[J].Journal of Institute for E-lectrical Engineering,1946,93(26):429-457.
[16]Daugman J G.Uncertainty relatin for resolution in space,spatial frequency and orientation optimized by two-dimensional visual cortical filters[J].Journal of the Optical Society of America,1985,2(7):1160-1169.
[17]Ojala T,Pietik M,Harwood D.A comparative study of texture measures with classification based on feature distributions[J].Pattern Recognition,1996,29(1):51-59.
[18]Ojala T,Pietikinen M,MenpT.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[C].European Conference on Computer Vision,Springer-Verlag,2000:404-420.
[19]Yuan F,Fang Z,Wu S,et al.Real-time image smoke detection using staircase searching-based dual threshold AdaBoost and dynamic analysis[J].IET Image Processing,2015,9(10):849-856.
[20]Yin Z,Wan B,Yuan F,et al.A deep normalization and convolutional neural network for image smoke detection[J].IEEE Access,2017,5(99):18429-18438.
[21]Ojala T,Pietikinen M,MenpT.Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[C].European Conference on Computer Vision,Springer-Verlag,2000:404-420.
[22]Guo Z,Zhang L.Zhang D.A completed modeling of local binary pattern operator for texture classification[J].IEEE Transactions on Image Processing,2010,19(6):1657-1663.
[23]Murala S,Maheshwari R P,Balasubramanian R.Local tetra patterns:a newfeature descriptor for content-based image retrieval[J].IEEE Transactions on Image Processing,2012,21(5):2874-2886.
[24]Qian X,Hua X S,Chen P,et al.PLBP:An effective local binary patterns texture descriptor with pyramid representation[J].Pattern Recognition,2011,44(10-11):2502-2515.
[25]Zhang B,Gao Y,Zhao S,et al.Local derivative pattern versus local binary pattern:face recognition with high-order local pattern descriptor[J].IEEE Transactions on Image Processing,2010,19(2):533-544.
[26]Krizhevsky A,Sutskever I,Hinton G E.ImageNet classification with deep convolutional neural networks[C].Advances in Neural Information Processing Systems 25,2012:1097-1105.
[27]Zeiler M D,Fergus R.Visualizing and understanding convolutional networks[C].European Conference on Computer Vision,Springer,Cham,2014:818-833.
[28]Simonyan K,Zisserman A.Very deep convolutional networks for large-scale image recognition[EB/OL].https://arxiv.org/pdf/1409.1556v6.pdf,2018.
[29]Shi Zhi-guo,Yang Zhi-yong.Research on the measurement of information loss in the dimension reduction of deep learning[J].Journal of Chinese Computer Systems,2017,38(7):1590-1594.
[1]罗胜,Jiang Yuzheng.视频检测烟雾的研究现状[J].中国图象图形学报,2013,18(10):1225-1236.
[4]潘朝峰,杨树森,陈宁.基于图像熵的火灾烟雾识别[J].江苏科技大学学报:自然科学版,2015,29(1):52-57.
[7]赵亮,骆炎民,骆翔宇.一种基于背景动态更新与暗通道先验的火灾烟雾检测算法[J].计算机应用研究,2017,34(3):957-960.
[9]张洪涛,马燕,张相芬,等.高鉴别力SIFT和LGQP特征及其在人脸识别中的应用[J].小型微型计算机系统,2017,38(11):2488-2493.
[29]石志国,杨志勇.深度学习降维过程中的信息损失度量研究[J].小型微型计算机系统,2017,38(7):1590-1594.
1http://staff. ustc. edu. cn/~yfn/vsd. html