基于Memetic算法的编码曝光最优码字序列搜索方法
|
摘要
提出了一种基于Memetic算法的编码曝光最优码字序列搜索方法。分析了编码曝光成像理论模型,建立了最优码字选取的适应度函数准则。引入Memetic算法框架并开展了最优编码序列搜索,利用遗传搜索算法进行了全局最优解搜索,并在此基础上利用模拟退火算法进行了局部最优解求解,通过适应度函数的阈值约束及种群和最优解的更新迭代,得到了最优码字搜索结果。研究结果表明,相比其他方法,所提算法兼顾了全局最优与局部最优的求解,得到的最优码字序列具有更优性能指标,算法执行效率高,复原图像的主客观评价质量更好。
A searching method for an optimal code sequence of coded exposure is proposed based on the Memetic algorithm. The theoretical model for coded exposure imaging is analyzed and the criteria of fitness function for the optimal codeword selection is established. The Memetic algorithm framework is introduced to carry out the optimal code sequence search, and the genetic search algorithm is utilized to implement the global optimal solution search. On this basis, the simulated annealing algorithm is used to conduct the local optimal solution. The optimal codeword search results are obtained by the threshold constraint of the fitness function and the updated iteration of population and optimal solution. The research results show that, compared with other methods, the proposed algorithm can take into account both the global and the local optimal solutions, the obtained optimal code sequence has a better performance index, the execution efficiency is high, and the restored image has superior subjective and objective evaluation quality.
A searching method for an optimal code sequence of coded exposure is proposed based on the Memetic algorithm. The theoretical model for coded exposure imaging is analyzed and the criteria of fitness function for the optimal codeword selection is established. The Memetic algorithm framework is introduced to carry out the optimal code sequence search, and the genetic search algorithm is utilized to implement the global optimal solution search. On this basis, the simulated annealing algorithm is used to conduct the local optimal solution. The optimal codeword search results are obtained by the threshold constraint of the fitness function and the updated iteration of population and optimal solution. The research results show that, compared with other methods, the proposed algorithm can take into account both the global and the local optimal solutions, the obtained optimal code sequence has a better performance index, the execution efficiency is high, and the restored image has superior subjective and objective evaluation quality.
引文
[1] Fergus R, Singh B, Hertzmann A, et al. Removing camera shake from a single photograph[J]. ACM Transactions on Graphics, 2006, 25(3): 787-794.
[2] Shan Q, Jia J Y, Agarwala A. High-quality motion deblurring from a single image[J]. ACM Transactions on Graphics, 2008, 27(3): 73.
[3] Cho S, Lee S. Fast motion deblurring[J]. ACM Transactions on Graphics, 2009, 28(5): 145.
[4] Tang C Y, Chen Y T, Li Q, et al. Motion detection and image restoration based on video reconstruction from a single coded exposure photograph[J]. Acta Optica Sinica, 2015, 35(4): 0410002. 唐超影, 陈跃庭, 李奇, 等. 基于视频重建的颤振探测与图像复原方法[J]. 光学学报, 2015, 35(4): 0410002.
[5] Yang A P, Wang J B, Yang B W, et al. Joint deep denoising prior for image blind deblurring[J]. Acta Optica Sinica, 2018, 38(10): 1010003. 杨爱萍, 王金斌, 杨炳旺, 等. 联合深度去噪先验图像盲去模糊[J]. 光学学报, 2018, 38(10): 1010003.
[6] Raskar R, Agrawal A, Tumblin J. Coded exposure photography: motion deblurring using fluttered shutter[J]. ACM Transactions on Graphics, 2006, 25(3): 795-804.
[7] Agrawal A, Raskar R. Optimal single image capture for motion deblurring[C]. IEEE Conference on Computer Vision and Pattern Recognition, 2009: 2560-2567.
[8] Jelinek J. Designing the optimal shutter sequences for the flutter shutter imaging method[J]. Proceedings of SPIE, 2010, 7701: 77010N.
[9] McCloskey S. Velocity-dependent shutter sequences for motion deblurring[C]. European Conference on Computer Vision, 2010: 309-322.
[10] McCloskey S. Temporally coded flash illumination for motion deblurring[C]. International Conference on Computer Vision, 2011: 683-690.
[11] McCloskey S, Ding Y Y, Yu J Y. Design and estimation of coded exposure point spread functions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(10): 2071-2077.
[12] Tendero Y, Osher S. On a mathematical theory of coded exposure[J]. Research in the Mathematical Sciences, 2016, 3(1): 4.
[13] Jeon H G, Lee J Y, Han Y, et al. Fluttering pattern generation using modified legendre sequence for coded exposure imaging[C]. IEEE International Conference on Computer Vision, 2013: 1001-1008.
[14] Jeon H G, Lee J Y, Han Y, et al. Generating fluttering patterns with low autocorrelation for coded exposure imaging[J]. International Journal of Computer Vision, 2017, 123(2): 269-286.
[15] Jeon H G, Lee J Y, Han Y, et al. Multi-image deblurring using complementary sets of fluttering patterns[J]. IEEE Transactions on Image Processing, 2017, 26(5): 2311-2326.
[16] Xu S K, Zhang J, Tu D, et al. A constant acceleration motion blur image deblurring based on hybrid coded exposure[J]. Journal of National University of Defense Technology, 2011, 33(6): 78-83, 94. 徐树奎, 张军, 涂丹, 等. 基于混合编码曝光的匀加速运动模糊图像复原方法[J]. 国防科技大学学报, 2011, 33(6): 78-83, 94.
[17] Huang K H, Zhang J, Xu S K, et al. Searching for the optimal code of coded exposure photography in the presence of CCD noise[J]. Journal of National University of Defense Technology, 2012, 34(6): 72-78. 黄魁华, 张军, 徐树奎, 等. 考虑CCD噪声条件下的编码曝光最优码字搜索方法[J]. 国防科技大学学报, 2012, 34(6): 72-78.
[18] Huang K H, Zhang J, Li G H. Noise-optimal capture for coded exposure photography[J]. Optical Engineering, 2012, 51(9): 093202.
[19] Huang K H, Hou J X, Zhang J. Optimal binary sequences for coded exposure photography using genetic algorithm[C]. IEEE International Conference on Signal Processing, Communications and Computing, 2014: 287-291.
[20] He L R, Cui G M, Feng H J, et al. Fast image restoration method based on coded exposure and vibration detection[J]. Optical Engineering, 2015, 54(10): 103107.
[21] He L R, Cui G M, Feng H J, et al. The optimal code searching method with an improved criterion of coded exposure for remote sensing image restoration[J]. Optics Communications, 2015, 338: 540-550.
[22] He F B. Research on motion deblurring using coded exposure[D]. Dalian: Dalian University of Technology, 2016. 何富斌. 编码曝光图像的运动去模糊方法研究[D]. 大连: 大连理工大学, 2016.
[23] Mao B Q. Pushframe imaging technology based on coded exposure[D]. Suzhou: Soochow University, 2016. 毛保奇. 编码曝光推帧成像技术研究[D]. 苏州: 苏州大学, 2016.
[24] Gao L, Liu W J, Zhou C. Solving of TSP with time restraints by Memetic algorithm[J]. Journal of Huazhong University of Science and Technology(Nature Science), 2008, 36(7): 93-96. 高亮, 刘文君, 周驰. 用Memetic算法求解有时间约束的TSP问题[J]. 华中科技大学学报(自然科学版), 2008, 36(7): 93-96.
[25] He L R, Huang J Z, Feng H J, et al. Research on fast restoration of blurred vibration image based on vibration detection and coded exposure[J]. Acta Optica Sinica, 2014, 34(9): 0910003. 何丽蓉, 黄加紫, 冯华君, 等. 基于颤振探测和编码曝光的快速图像复原方法[J]. 光学学报, 2014, 34(9): 0910003.
[2] Shan Q, Jia J Y, Agarwala A. High-quality motion deblurring from a single image[J]. ACM Transactions on Graphics, 2008, 27(3): 73.
[3] Cho S, Lee S. Fast motion deblurring[J]. ACM Transactions on Graphics, 2009, 28(5): 145.
[4] Tang C Y, Chen Y T, Li Q, et al. Motion detection and image restoration based on video reconstruction from a single coded exposure photograph[J]. Acta Optica Sinica, 2015, 35(4): 0410002. 唐超影, 陈跃庭, 李奇, 等. 基于视频重建的颤振探测与图像复原方法[J]. 光学学报, 2015, 35(4): 0410002.
[5] Yang A P, Wang J B, Yang B W, et al. Joint deep denoising prior for image blind deblurring[J]. Acta Optica Sinica, 2018, 38(10): 1010003. 杨爱萍, 王金斌, 杨炳旺, 等. 联合深度去噪先验图像盲去模糊[J]. 光学学报, 2018, 38(10): 1010003.
[6] Raskar R, Agrawal A, Tumblin J. Coded exposure photography: motion deblurring using fluttered shutter[J]. ACM Transactions on Graphics, 2006, 25(3): 795-804.
[7] Agrawal A, Raskar R. Optimal single image capture for motion deblurring[C]. IEEE Conference on Computer Vision and Pattern Recognition, 2009: 2560-2567.
[8] Jelinek J. Designing the optimal shutter sequences for the flutter shutter imaging method[J]. Proceedings of SPIE, 2010, 7701: 77010N.
[9] McCloskey S. Velocity-dependent shutter sequences for motion deblurring[C]. European Conference on Computer Vision, 2010: 309-322.
[10] McCloskey S. Temporally coded flash illumination for motion deblurring[C]. International Conference on Computer Vision, 2011: 683-690.
[11] McCloskey S, Ding Y Y, Yu J Y. Design and estimation of coded exposure point spread functions[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(10): 2071-2077.
[12] Tendero Y, Osher S. On a mathematical theory of coded exposure[J]. Research in the Mathematical Sciences, 2016, 3(1): 4.
[13] Jeon H G, Lee J Y, Han Y, et al. Fluttering pattern generation using modified legendre sequence for coded exposure imaging[C]. IEEE International Conference on Computer Vision, 2013: 1001-1008.
[14] Jeon H G, Lee J Y, Han Y, et al. Generating fluttering patterns with low autocorrelation for coded exposure imaging[J]. International Journal of Computer Vision, 2017, 123(2): 269-286.
[15] Jeon H G, Lee J Y, Han Y, et al. Multi-image deblurring using complementary sets of fluttering patterns[J]. IEEE Transactions on Image Processing, 2017, 26(5): 2311-2326.
[16] Xu S K, Zhang J, Tu D, et al. A constant acceleration motion blur image deblurring based on hybrid coded exposure[J]. Journal of National University of Defense Technology, 2011, 33(6): 78-83, 94. 徐树奎, 张军, 涂丹, 等. 基于混合编码曝光的匀加速运动模糊图像复原方法[J]. 国防科技大学学报, 2011, 33(6): 78-83, 94.
[17] Huang K H, Zhang J, Xu S K, et al. Searching for the optimal code of coded exposure photography in the presence of CCD noise[J]. Journal of National University of Defense Technology, 2012, 34(6): 72-78. 黄魁华, 张军, 徐树奎, 等. 考虑CCD噪声条件下的编码曝光最优码字搜索方法[J]. 国防科技大学学报, 2012, 34(6): 72-78.
[18] Huang K H, Zhang J, Li G H. Noise-optimal capture for coded exposure photography[J]. Optical Engineering, 2012, 51(9): 093202.
[19] Huang K H, Hou J X, Zhang J. Optimal binary sequences for coded exposure photography using genetic algorithm[C]. IEEE International Conference on Signal Processing, Communications and Computing, 2014: 287-291.
[20] He L R, Cui G M, Feng H J, et al. Fast image restoration method based on coded exposure and vibration detection[J]. Optical Engineering, 2015, 54(10): 103107.
[21] He L R, Cui G M, Feng H J, et al. The optimal code searching method with an improved criterion of coded exposure for remote sensing image restoration[J]. Optics Communications, 2015, 338: 540-550.
[22] He F B. Research on motion deblurring using coded exposure[D]. Dalian: Dalian University of Technology, 2016. 何富斌. 编码曝光图像的运动去模糊方法研究[D]. 大连: 大连理工大学, 2016.
[23] Mao B Q. Pushframe imaging technology based on coded exposure[D]. Suzhou: Soochow University, 2016. 毛保奇. 编码曝光推帧成像技术研究[D]. 苏州: 苏州大学, 2016.
[24] Gao L, Liu W J, Zhou C. Solving of TSP with time restraints by Memetic algorithm[J]. Journal of Huazhong University of Science and Technology(Nature Science), 2008, 36(7): 93-96. 高亮, 刘文君, 周驰. 用Memetic算法求解有时间约束的TSP问题[J]. 华中科技大学学报(自然科学版), 2008, 36(7): 93-96.
[25] He L R, Huang J Z, Feng H J, et al. Research on fast restoration of blurred vibration image based on vibration detection and coded exposure[J]. Acta Optica Sinica, 2014, 34(9): 0910003. 何丽蓉, 黄加紫, 冯华君, 等. 基于颤振探测和编码曝光的快速图像复原方法[J]. 光学学报, 2014, 34(9): 0910003.