计算机科学 ›› 2017, Vol. 44 ›› Issue (Z11): 225-227.doi: 10.11896/j.issn.1002-137X.2017.11A.047
任云,程福林,黎洪松
REN Yun, CHENG Fu-lin and LI Hong-song
摘要： 提出基于频率敏感三维自组织映射的立体视频视差估计算法,视差预测采用基于亮度分类频率敏感三维自组织映射(Classified Frequency Sensitive Three-Dimensional Self-Organizing Map,CFS-3DSOM)的视差模式识别(Disparity pattern recognition,DPR)算法。其基本思想是对视差图像的低亮度区域和高亮度区域分别进行预测,在训练模式库时引入频率敏感方法。实验表明,与传统基于块的视差估计算法相比,CFS-3DSOM-DPR算法的视差预测图像的平均峰值信噪比提高了0.78~1.78dB,时间减少了70%。
| [1] HAUBLEIN K,REICHENBACH M,FEY D.Fast and generic hardware architecture for stereo block matching applications on embedded systems[C]∥2014 International Conference on Re-ConFigurable Computing and FPGAs(ReConFig).IEEE,2014:1-6. [2] SAHLBACH H,ERNST R,WONNEBERGER S,et al.Exploration of FPGA-based dense block matching for motion estimation and stereo vision on a single chip[C]∥IEEE Intelligent Vehicles Symposium.IEEE,2013:823-828. [3] WANG Y C,TUNG C P,CHUNG P C.Efficient disparity estimation using hierarchical bilateral disparity structure based graph cut algorithm with a foreground boundary refinement mechanism [J].IEEE Transactions on Circuits & Systems for Video Technology,2013,23(5):784-801. [4] 赵戈,蔺蘭,唐延东,等.一种基于曲率与置信度传播的视差估计方法[J].模式识别与人工智能,2013,26(12):1154-1160. [5] PODDAR S,SAHU H,BANGALE M R,et al.Feature based dense disparity estimation[C]∥2015 International Conference on Industrial Instrumentation and Control (ICIC).IEEE,2015:950-955. [6] KAMENCAY P,BREZNAN M,JELSOVKA D,et al.Sparse dis-parity map computation from stereo-view images using segment based algorithm[C]∥2012 22nd International Conference Radioelektronika (RADIOELEKTRONIKA).IEEE,2012:1-4. [7] RAMAN S,KANOJIA G,et al.FacialStereo:Facial depth estimation from a stereo pair[C]∥International Conference on Computer Vision Theory and Applications.2014:686-691. [8] KOHONEN T.Essentials of the self-organizing map[J].Neural Networks,2013,7:52-65. [9] LI N,CHENG X,ZHANG S,et al.Realistic human action re-cognition by Fast HOG3D and self-organization feature map[J].Machine Vision and Applications,2014,25(7):1793-1812. [10] KAMAL S,MUJEEB A,SUPRIYA M H.Novel class detection of underwater targets using self-organizing neural networks[C]∥Underwater Technology (UT),2015.IEEE,2015:1-5. [11] PARK S,RYU U S,CHOI Y,et al.A framework for baselineload estimation in demand response:Data mining approach[C]∥2014 IEEE International Conference on Smart Grid Communications (SmartGridComm).IEEE,2014:638-643. [12] MCLOUGHLIN F,DUFFY A,CONLON M.A clustering approach to domestic electricity load profile characterisation using smart metering data[J].Applied Energy,2015,141:190-199. |
| No related articles found! |
|
||
首页