计算机科学 ›› 2020, Vol. 47 ›› Issue (12): 197-204.doi: 10.11896/jsjkx.191000054
陈晓文, 刘光帅, 刘望华, 李旭瑞
CHEN Xiao-wen, LIU Guang-shuai, LIU Wang-hua, LI Xu-rui
摘要: 针对原始的局部相位量化(Local Phase QuantizationLPQ)算法对具有模糊不变性的相位特征描述不准确、缺少对图像重要细节信息描述的缺点提出了一种结合高斯拉普拉斯(Laplace of GaussianLoG)边缘检测和增强局部相位量化(Enhanced Local Phase QuantizationELPQ)的模糊图像识别算法记为MrELPQ&MsLoG(Multi-resolution ELPQand Multi-scaleLoG).首先在频域中将图像进行短时傅里叶变换后得到的实部与虚部进行正负量化和幅值量化得到互补的符号特征ELPQ_S和幅值特征ELPQ_M;其次在空间域中利用多尺度高斯拉普拉斯与图像进行卷积得到图像空间域的边缘特征;最后将频域上的符号特征ELPQ_S和幅值特征ELPQ_M与空间域上的边缘特征结合生成最终的特征直方图采用SVM进行识别.在有模糊干扰的Brodatz和KTH-TIPS纹理库中文中提出的ELPQ算法相比原始的LPQ算法有较大的性能提升且空间域和频域结合的MrELPQ&MsLoG算法能进一步提高算法的识别性能;在具有模糊的AR、Extend YaleB人脸库和实际拍摄的铁路扣件库中将MrELPQ&MsLoG算法与目前模糊鲁棒性较好的算法进行对比发现MrELPQ&MsLoG算法保持着较高的识别率.实验结果表明MrELPQ&MsLoG算法对模糊具有较强的鲁棒性且特征提取时间较短具有实时性.
中图分类号:
[1] SHEN Y P.Learning of image motion blur invariant features[D].Chongqing:Chongqing University,2017. [2] DALAL N,TRIGGS B.Histograms of Oriented Gradients for Human Detection[C]//International Conference on Computer Vision &Pattern Recognition (CVPR'05).2005:886-893. [3] OJALA T,PIETIKÁNTEN M,HAWOODD.A comparativestudy of texture measure with classification based on featured distributions[J].Pattern Recognition,1996,29(1):51-59. [4] FLUSSER J,SUK T,SAIC S.Recognition of blurred images by the method of moments[J].IEEE Transactions on Image Processing,1996,5(3):533-538. [5] DAI X,LIU T,SHU H,et al.Pseudo-zernike moment invariants to blur degradation and their use in image recognition[C]//Intelligent Science and Intelligent Data Engineering.Springer,2013:90-97. [6] OJANSIVU V,HEIKKILÄ J.Blur insensitive texture classification using local phase quantization[C]//Image and Signal Processing.Springer,2008:236-243. [7] ZHU M,CAO Z,XIAO Y,et al.Beyond local phase quantiza-tion:Mid-level blurred image representation using fisher vector[C]//2015 IEEE International Conference on Image Processing (ICIP).IEEE,2015:1890-1894. [8] LI L,SHI F L,XU N N.Face recognition based on adaptively weighted[J].Opto-Electronic Engineering,2012,39(12):138-142. [9] WANG W W,CAO Z G.Recognition of blurred faces using local phase pattern[J].Electronics Letters,2012,48(20):1269. [10] BHATTACHARYA S,DASGUPTA A,ROUTRAY A.Robust face recognition of inferior quality images using Local Gabor Phase Quantization[C]//2016 IEEE Students' Technology Symposium (TechSym).IEEE,2017:294-298. [11] NGUYEN H T,HUONG TT N.Gender classification by LPQ features from intensity and Monogenic images[C]//2017 4th NAFOSTED Conference on Information and Computer Science.IEEE,2017:96-100. [12] XIAO Y,CAO Z,WANG L,et al.Local phase quantization plus:A principled method for embedding local phase quantization into fisher vector for blurred image recognition[J].Information Sciences,2017,420:77-95. [13] CHAN C H,TAHIR M A,KITTLER J,et al.Multiscale local phase quantization for robust component-based face recognition using kernel fusion of multiple descriptors[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,35(5):1164-1177. [14] AHONEN T,RAHTU E,OJANSIVU V,et al.Recognition of blurred faces using local phase quantization[C]//ICPR.2008:1-4. [15] 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. [16] ZHENG L P.Research on Second-order Differential Feature Extraction and Face Recognition[D].Kunming:Kunming University of Science and Technology ,2015. [17] YANG H X,LIU F,HE D L.Face recognition based on Laplace filter and Center-Symmetric Local Binary Pattern[J].Computer Engineering and Applications,2017,53(7):165-170. [18] LIU J F,ZUO X Q,WUL M.Edge detection of remote sensing image based on LOG algorithm and Otsu algorithm[J].Science of Surveying and Mapping,2015,40(1):107-109. [19] NANNI L,BRAHNAM S,LUMINI A,et al.Ensemble of local phase quantization variants with ternary encoding[M]//Local Binary Patterns:New Variants and Applications.Springer,Berlin,Heidelberg,2014:177-188. [20] WANG R,LI W,QIN R,et al.Blur image classification based on deep learning[C]//2017 IEEE International Conference on Imaging Systems and Techniques (IST).IEEE,2017:1-6. [21] YU D,WU X J.2DPCANet:a deep leaning network for face re-cognition[J].Multimedia Tools and Applications,2018,77(10):12919-12934. |
[1] | 尹文兵, 高戈, 曾邦, 王霄, 陈怡. 基于时频域生成对抗网络的语音增强算法 Speech Enhancement Based on Time-Frequency Domain GAN 计算机科学, 2022, 49(6): 187-192. https://doi.org/10.11896/jsjkx.210500114 |
[2] | 赫晓慧, 邱芳冰, 程淅杰, 田智慧, 周广胜. 基于边缘特征融合的高分影像建筑物目标检测 High-resolution Image Building Target Detection Based on Edge Feature Fusion 计算机科学, 2021, 48(9): 140-145. https://doi.org/10.11896/jsjkx.200800002 |
[3] | 杨小琴, 刘国军, 郭建慧, 马文涛. 基于随机森林的空域-频域联合特征全参考彩色图像质量评价方法 Full Reference Color Image Quality Assessment Method Based on Spatial and Frequency Domain Joint Features with Random Forest 计算机科学, 2021, 48(8): 99-105. https://doi.org/10.11896/jsjkx.200700106 |
[4] | 李兴国, 任益枚, 田竞, 唐竟淇. SFRA方法在交流伺服系统中的应用研究 Application of SFRA Method in AC Servo System 计算机科学, 2020, 47(11A): 628-631. https://doi.org/10.11896/jsjkx.190600163 |
[5] | 刘泽康, 孙华志, 马春梅, 姜丽芬. 基于卷积神经网络多特征联合的车辆识别模型 Vehicle Recognition Model Based on Multi-feature Combination inConvolutional Neural Network 计算机科学, 2019, 46(6A): 254-258. |
[6] | 刘加存, 赵桂艳, 梅其祥. 迭代学习控制的最优学习律和简化学习律的频域研究 Study of Optimal Learning Law and Simplified Learning Law of Iterative Learning Control in Frequency Domain 计算机科学, 2019, 46(2): 327-332. https://doi.org/10.11896/j.issn.1002-137X.2019.02.050 |
[7] | 梁晏慧, 李国东, 王爱银. 基于分数阶Chen超混沌的频域自适应图像加密算法 Frequency Domain Adaptive Image Encryption Algorithm Based on Fractional Order Chen Hyperchaos 计算机科学, 2019, 46(11A): 488-492. |
[8] | 何晓俊, 吴梦麟, 范雯, 袁松涛, 陈强. 基于三维限制区域生长的SD-OCT中浆NRD病变区域分割 SD-OCT CSC NRD Region Segmentation Based on Region Restricted 3D Region Growing 计算机科学, 2018, 45(6A): 187-192. |
[9] | 袁小艳, 王安志, 王明辉. 融入频域信息聚焦特征的显著性目标检测算法 Saliency Object Detection Algorithm Integrating Focusness Feature of Frequency Domain Information 计算机科学, 2018, 45(10): 261-266. https://doi.org/10.11896/j.issn.1002-137X.2018.10.048 |
[10] | 俞晨琛,陈强,范雯,袁松涛,刘庆淮. 基于自适应阈值和区域生长的SD-OCT糖网图像亮斑分割 Segmentation of Bright Speckles in SD-OCT Diabetic Retinal Images Based on Self-adaption Threshold and Region Growing 计算机科学, 2015, 42(Z11): 123-125. |
[11] | 米晓萍,李雪梅. 基于信息融合度传递的频域徙动入侵特征挖掘算法 Mining Algorithm of Frequency Domain Migration Intrusion Feature Based on Information Fusion Transfer 计算机科学, 2015, 42(3): 224-227. https://doi.org/10.11896/j.issn.1002-137X.2015.03.046 |
[12] | 孙统风,丁世飞. 基于改进尺度不变特征的图像局域几何配准研究 Image Local Geometric Registration Based on Improved Scale Invariant Features 计算机科学, 2014, 41(1): 111-115. |
[13] | 文昌辞,王 沁,苗晓宁,刘向宏,彭阳翔. 数字图像加密综述 Digital Image Encryption:A Survey 计算机科学, 2012, 39(12): 6-9. |
[14] | 曲智国,高颖慧,王平,王鹏,沈振康. 基于空频域联合闭值分割的轮廓检测方法 Contour Detection by Thresholding the Gradient Image in Spatial-frequency Domain 计算机科学, 2012, 39(10): 286-289. |
[15] | 王建锋,马建,马荣贵,宋宏勋. 动位移的加速度精确测量技术研究 Study on Calculation of Dynamic Displacement from Time-frequency Integration of Acceleration 计算机科学, 2010, 37(12): 201-202. |
|