基于BP神经网络的数字图像相关非迭代灰度梯度算法
|
摘要
为了提高非迭代灰度梯度算法的亚像素位移求解精度与效率,提出一种基于神经网络的非迭代灰度梯度改进算法。该算法应用BP神经网络直接建立变形前后散斑图子图像的灰度及灰度梯度与亚像素位移之间的非线性映射关系,无需对相关系数进行最小二乘优化求解。应用模拟散斑图像对算法的亚像素位移求解精度与效率进行了验证。结果表明,作者提出的算法的计算精度与效率较传统的非迭代灰度梯度算法均有提高。最后进行了真实的刚体平移实验,其结果进一步证明了作者提出的算法的有效性。
In order to improve the accuracy and efficiency of solution of the subpixel displacement measurement of non-iterative gray-gradient algorithm in digital image correlation,an improved algorithm based on artificial neural network is proposed. The Back Propagation( BP) artificial neural network was employed to build directly the nonlinear mapping relationship between the gray and gray-gradient in the subimages of speckle images before and after deformation and subpixel displacements,without the need for least square analytical optimal solution of correleation coefficient. Speckle images were simulated and applied to verify the accuracy and efficiency of solution of the Subpixel displacement measurement of the algorithm. The results reveal that the proposed algorithm has higher computation accuracy and efficiency than traditional non-iterative gray-gradient algorithm. The validity of the proposed algorithm is verified further by rigid body translation experiment finally.
In order to improve the accuracy and efficiency of solution of the subpixel displacement measurement of non-iterative gray-gradient algorithm in digital image correlation,an improved algorithm based on artificial neural network is proposed. The Back Propagation( BP) artificial neural network was employed to build directly the nonlinear mapping relationship between the gray and gray-gradient in the subimages of speckle images before and after deformation and subpixel displacements,without the need for least square analytical optimal solution of correleation coefficient. Speckle images were simulated and applied to verify the accuracy and efficiency of solution of the Subpixel displacement measurement of the algorithm. The results reveal that the proposed algorithm has higher computation accuracy and efficiency than traditional non-iterative gray-gradient algorithm. The validity of the proposed algorithm is verified further by rigid body translation experiment finally.
引文
[1]PAN B,QIAN K M,XIE H M,et al.Two-dimensional Digital Image Correlation for In-plane Displacement and Strain Measurement:a Review[J].Measurement Science&Technology,2009,20(6):062001-17.
[2]刘小勇.数字图像相关方法及其在材料力学性能测试中的应用[D].长春:吉林大学,2012.
[3]WANG M,CEN Y W,HU X F,et al.A Weighting Window Applied to the Digital Image Correlation Method[J].Optics&Laser Technology,2009,41(2):154-158.
[4]潘兵,谢惠民,续伯钦,等.数字图像相关中的亚像素位移定位算法进展[J].力学进展,2005,35(3):345-352.PAN B,XIE H M,XU B Q,et al.Development of Sub-pixel Displacements Registration Algorithms in Digital Image Correlation[J].Advances in Mechanics,2005,35(3):345-352.
[5]PAN B,XIE H M,XU B Q,et al.Performance of Sub-pixel Registration Algorithms in Digital Image Correlation[J].Measurement Science&Technology,2006,17(6):1615-1621.
[6]XIONG L,LIU X Y,LIU G D,et al.Evaluation of Sub-pixel Displacement Measurement Algorithms in Digital Image Correlation[C]//The International Conference on Mechatronic Science,Electric Engineering and Computer,2011,1-4.
[7]BRUCK H A,MCNEILL S R,SUTTON M A,et al.Digital Image Correlation Using Newton-Raphson Method of Partial Differential Correction[J].Experimental Mechanics,1989,29(3):261-267.
[8]王琛影,何小元.相关识别中的曲面拟合法[J].实验力学,2000,15(9):281-285.WANG C Y,HE X Y.Curved Surface Approximation in Correlation Method[J].Journal of Experimental Mechanics,2000,15(9):281-285.
[9]DAVIS C Q,FREEMAN D M.Statistics of Subpixel Registration Algorithms Based on Spatiotemporal Gradients or Block Matching[J].Optical Engineering,1998,37(4):1290-1298.
[10]MENG L B,JIN G C,YAO X F.Application of Iteration and Finite Element Smoothing Technique for Displacement and Strain Measurement of Digital Speckle Correlation[J].Optics and Lasers in Engineering,2007,45(1):57-63.
[11]潘兵,续伯钦,李克景.梯度算子选择对基于梯度的亚像素位移算法的影响[J].光学技术,2005,31(1):26-31.PAN B,XU B Q,LI K J.Performance of Gradient Operators in Algorithm of Gradient-based Subpixel Registration[J].Optical Technique,2005,31(1):26-31.
[12]谢锋云,曹青松,黄志刚.基于小波包-BP神经网络的切削颤振监测[J].仪表技术与传感器,2015(10):88-90.XIE F Y,CAO Q S,HUANG Z G.Chatter Monitoring Based on Wavelet Packet and BP Neural Network[J].Instrument Technique and Sensor,2015(10):88-90.
[13]张楠,王亮清,葛云峰,等.基于因子分析的BP神经网络在岩体变形模量预测中的应用[J].工程地质学报,2016(1):87-95.ZHANG N,WANG L Q,GE Y F,et al.Application of BP Neural Network Based on Factor Analysis to Prediction of Rock Mass Deformation Modulus[J].Journal of Engineering Geology,2016(1):87-95.
[14]张靓,李铁军,宗银雪.融合视听信息的机电设备状态监测系统设计[J].仪表技术与传感器,2015(9):56-59.ZHANG L,LI T J,ZONG Y X.Electromechanical Equipment Condition Monitoring System Based on Audio-visual Information Fusion[J].Instrument Technique and Sensor,2015(9):56-59.
[15]杨兵兵,鞠建波,赵玉刚,等.基于BP神经网络的反潜巡逻机背景磁场补偿技术研究[J].计算机测量与控制,2016,24(1):146-152.YANG B B,JU J B,ZHAO Y G,et al.Technical Research of Anti-submarine Patrol Aircraft Magnetic Compensation Based on BP Neural Network[J].Computer Measurement&control,2016,24(1):146-152.
[16]ZHOU P,GOODSON K E.Subpixel Displacement and Deformation Gradient Measurement Using Digital Image/Speckle Correlation(DISC)[J].Optical Engineering,2001,40(8):1613-1620.
[17]LIU X Y,LI R L,ZHAO H W,et al.Quality Assessment of Speckle Patterns for Digital Image Correlation by Shannon Entropy[J].Optik,2015,126(23):4206-4211.
[2]刘小勇.数字图像相关方法及其在材料力学性能测试中的应用[D].长春:吉林大学,2012.
[3]WANG M,CEN Y W,HU X F,et al.A Weighting Window Applied to the Digital Image Correlation Method[J].Optics&Laser Technology,2009,41(2):154-158.
[4]潘兵,谢惠民,续伯钦,等.数字图像相关中的亚像素位移定位算法进展[J].力学进展,2005,35(3):345-352.PAN B,XIE H M,XU B Q,et al.Development of Sub-pixel Displacements Registration Algorithms in Digital Image Correlation[J].Advances in Mechanics,2005,35(3):345-352.
[5]PAN B,XIE H M,XU B Q,et al.Performance of Sub-pixel Registration Algorithms in Digital Image Correlation[J].Measurement Science&Technology,2006,17(6):1615-1621.
[6]XIONG L,LIU X Y,LIU G D,et al.Evaluation of Sub-pixel Displacement Measurement Algorithms in Digital Image Correlation[C]//The International Conference on Mechatronic Science,Electric Engineering and Computer,2011,1-4.
[7]BRUCK H A,MCNEILL S R,SUTTON M A,et al.Digital Image Correlation Using Newton-Raphson Method of Partial Differential Correction[J].Experimental Mechanics,1989,29(3):261-267.
[8]王琛影,何小元.相关识别中的曲面拟合法[J].实验力学,2000,15(9):281-285.WANG C Y,HE X Y.Curved Surface Approximation in Correlation Method[J].Journal of Experimental Mechanics,2000,15(9):281-285.
[9]DAVIS C Q,FREEMAN D M.Statistics of Subpixel Registration Algorithms Based on Spatiotemporal Gradients or Block Matching[J].Optical Engineering,1998,37(4):1290-1298.
[10]MENG L B,JIN G C,YAO X F.Application of Iteration and Finite Element Smoothing Technique for Displacement and Strain Measurement of Digital Speckle Correlation[J].Optics and Lasers in Engineering,2007,45(1):57-63.
[11]潘兵,续伯钦,李克景.梯度算子选择对基于梯度的亚像素位移算法的影响[J].光学技术,2005,31(1):26-31.PAN B,XU B Q,LI K J.Performance of Gradient Operators in Algorithm of Gradient-based Subpixel Registration[J].Optical Technique,2005,31(1):26-31.
[12]谢锋云,曹青松,黄志刚.基于小波包-BP神经网络的切削颤振监测[J].仪表技术与传感器,2015(10):88-90.XIE F Y,CAO Q S,HUANG Z G.Chatter Monitoring Based on Wavelet Packet and BP Neural Network[J].Instrument Technique and Sensor,2015(10):88-90.
[13]张楠,王亮清,葛云峰,等.基于因子分析的BP神经网络在岩体变形模量预测中的应用[J].工程地质学报,2016(1):87-95.ZHANG N,WANG L Q,GE Y F,et al.Application of BP Neural Network Based on Factor Analysis to Prediction of Rock Mass Deformation Modulus[J].Journal of Engineering Geology,2016(1):87-95.
[14]张靓,李铁军,宗银雪.融合视听信息的机电设备状态监测系统设计[J].仪表技术与传感器,2015(9):56-59.ZHANG L,LI T J,ZONG Y X.Electromechanical Equipment Condition Monitoring System Based on Audio-visual Information Fusion[J].Instrument Technique and Sensor,2015(9):56-59.
[15]杨兵兵,鞠建波,赵玉刚,等.基于BP神经网络的反潜巡逻机背景磁场补偿技术研究[J].计算机测量与控制,2016,24(1):146-152.YANG B B,JU J B,ZHAO Y G,et al.Technical Research of Anti-submarine Patrol Aircraft Magnetic Compensation Based on BP Neural Network[J].Computer Measurement&control,2016,24(1):146-152.
[16]ZHOU P,GOODSON K E.Subpixel Displacement and Deformation Gradient Measurement Using Digital Image/Speckle Correlation(DISC)[J].Optical Engineering,2001,40(8):1613-1620.
[17]LIU X Y,LI R L,ZHAO H W,et al.Quality Assessment of Speckle Patterns for Digital Image Correlation by Shannon Entropy[J].Optik,2015,126(23):4206-4211.