基于智能识别的激光成像特征控制
|
摘要
针对激光成像在重叠波段特征控制稳态性不好的问题,为了提高激光成像的像元特征表达能力,从而实现激光成像的智能识别,提出一种基于梯度像素差控制和智能识别的激光成像特征控制技术。运用WorldView-2激光图像处理系统进行激光遥感图像采集,对采集的灰度不均匀激光图像进行模板配准和分块匹配处理,建立激光成像的仿射不变区域分布结构模型,结合局部灰度项和局部梯度项特征融合方法进行激光成像的信息融合,对融合后的激光成像采用SURF算法实现梯度像素差特征控制,将激光图像特征分布集引入到极限学习机中进行自适应学习,在重叠波段实现对激光成像的智能识别和成像特征控制。仿真结果表明,采用该方法进行激光成像特征控制的稳定性较好,提高了激光成像识别能力和特征分辨力。
Aiming at the problem of poor steady-state control of laser imaging in overlapping band,in order to improve the pixel feature expression ability of laser imaging,the intelligent recognition of laser imaging can be realized. A laser imaging feature control technique based on gradient pixel difference control and intelligent recognition is proposed.The WorldView-2 laser image processing system is used to collect laser remote sensing images. The template registration and block matching of the grayscale uneven laser images are carried out,and the affine invariant region distribution structure model of laser imaging is established. Combining the local grayscale term and the local gradient feature fusion method,the information fusion of laser imaging is carried out. The SURF algorithm is used to realize the gradient pixel difference feature control for the fused laser imaging. The laser image feature distribution set is introduced into the extreme learning machine for adaptive learning,and the intelligent recognition and imaging feature control of laser imaging are realized in overlapping bands. The simulation results show that the stability of laser imaging feature control by this method is good,and the recognition ability and feature resolution of laser imaging are improved.
Aiming at the problem of poor steady-state control of laser imaging in overlapping band,in order to improve the pixel feature expression ability of laser imaging,the intelligent recognition of laser imaging can be realized. A laser imaging feature control technique based on gradient pixel difference control and intelligent recognition is proposed.The WorldView-2 laser image processing system is used to collect laser remote sensing images. The template registration and block matching of the grayscale uneven laser images are carried out,and the affine invariant region distribution structure model of laser imaging is established. Combining the local grayscale term and the local gradient feature fusion method,the information fusion of laser imaging is carried out. The SURF algorithm is used to realize the gradient pixel difference feature control for the fused laser imaging. The laser image feature distribution set is introduced into the extreme learning machine for adaptive learning,and the intelligent recognition and imaging feature control of laser imaging are realized in overlapping bands. The simulation results show that the stability of laser imaging feature control by this method is good,and the recognition ability and feature resolution of laser imaging are improved.
引文
[1]汪锋,孙开江,向小梅.基于光电倍增管的自动增益控制技术研究[J].激光技术,2015,39(4):510-514.
[2]朱建新,沈东羽,吴钪.基于激光点云的智能挖掘机目标识别[J].计算机工程,2017,43(1):297-302.
[3]牛英宇,聂瑞杰,李丽娟.基于FPGA的红外图像非均匀校正实现方法[J].激光与红外,2016,46(8):1028-1032.
[4]LIM B,SON S,KIM H,et al.Enhanced deep residual networks for single image super-resolution[C]//CVPRW2017:Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington,DC:IEEE Computer Society,2017:1132-1140.
[5]YANG L.Multi-focus image fusion method based on NSSTandⅡCM[C]//Proceedings of the 2017 International Conference on Emerging Internetworking,Data and Web Technologies.Berlin:Springer,2017:679-689.
[6]YANG L.Technique for image de-noising based on nonsubsampled shearlet transform and improved intuitionistic fuzzy entropy[J].Optik-International Journal for Light and Electron Optics,2015,126(4):446-453.
[7]陈波.送粉式激光3D打印数据采集与监控系统的设计与实现[J].电子设计工程,2017,25(4):66-70.
[8]赵坤,史学舜,刘长明,等.用于探测器中红外绝对光谱响应度测量的激光源[J].红外与激光工程,2016,45(7):705005-0705005(7).
[9]蒋理,李维勇.红外激光成像中的弱小目标的准确定位技术研究[J].现代电子技术,2016,39(20):83-86.
[10]卜燕,沈俊,张华,等.虚拟二维激光雷达成像积分算法[J].计算机测量与控制,2016,24(4):152-154.
[11]周靖鸿,周璀,朱建军,等.基于非下采样轮廓波变换遥感影像超分辨重建方法[J].光学学报,2015,35(1):106-114
[12]BHARADI V A,PADOLE L.Performance comparison of hybrid wavelet transform-I variants and contrast limited adaptive histogram equalization combination for image enhancement[C]//Proceedings of the 201714th International Conference on Wireless and Optical Communications Networks.Piscataway,NJ:IEEE,2017:1-8.
[13]LV D,JIA Z,YANG J,et al.Remote sensing image enhancement based on the combination of nonsubsampled shearlet transform and guided filtering[J].Optical Engineering,2016,55(10):103-104.
[14]周飞,贾振红,杨杰,等.基于剪切波域改进Gamma校正的医学图像增强算法[J].光电子·激光,2017,28(5):566-572.
[15]焦阿敏,董明利,娄小平,等.光条图像成像参数的模糊自适应调整方法研究[J].工具技术,2017,51(5):112-117.
[16]赵伟飞,康诚,钱登朝,等.城市隧道照明调光控制参数特征分析[J].低温建筑技术,2018(6).
[17]李春凯,石玗,顾玉芬,等.GTAW熔池三维表面激光点阵反射特征的仿真与优化[J].焊接学报,2017,38(9):42-46.
[2]朱建新,沈东羽,吴钪.基于激光点云的智能挖掘机目标识别[J].计算机工程,2017,43(1):297-302.
[3]牛英宇,聂瑞杰,李丽娟.基于FPGA的红外图像非均匀校正实现方法[J].激光与红外,2016,46(8):1028-1032.
[4]LIM B,SON S,KIM H,et al.Enhanced deep residual networks for single image super-resolution[C]//CVPRW2017:Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops.Washington,DC:IEEE Computer Society,2017:1132-1140.
[5]YANG L.Multi-focus image fusion method based on NSSTandⅡCM[C]//Proceedings of the 2017 International Conference on Emerging Internetworking,Data and Web Technologies.Berlin:Springer,2017:679-689.
[6]YANG L.Technique for image de-noising based on nonsubsampled shearlet transform and improved intuitionistic fuzzy entropy[J].Optik-International Journal for Light and Electron Optics,2015,126(4):446-453.
[7]陈波.送粉式激光3D打印数据采集与监控系统的设计与实现[J].电子设计工程,2017,25(4):66-70.
[8]赵坤,史学舜,刘长明,等.用于探测器中红外绝对光谱响应度测量的激光源[J].红外与激光工程,2016,45(7):705005-0705005(7).
[9]蒋理,李维勇.红外激光成像中的弱小目标的准确定位技术研究[J].现代电子技术,2016,39(20):83-86.
[10]卜燕,沈俊,张华,等.虚拟二维激光雷达成像积分算法[J].计算机测量与控制,2016,24(4):152-154.
[11]周靖鸿,周璀,朱建军,等.基于非下采样轮廓波变换遥感影像超分辨重建方法[J].光学学报,2015,35(1):106-114
[12]BHARADI V A,PADOLE L.Performance comparison of hybrid wavelet transform-I variants and contrast limited adaptive histogram equalization combination for image enhancement[C]//Proceedings of the 201714th International Conference on Wireless and Optical Communications Networks.Piscataway,NJ:IEEE,2017:1-8.
[13]LV D,JIA Z,YANG J,et al.Remote sensing image enhancement based on the combination of nonsubsampled shearlet transform and guided filtering[J].Optical Engineering,2016,55(10):103-104.
[14]周飞,贾振红,杨杰,等.基于剪切波域改进Gamma校正的医学图像增强算法[J].光电子·激光,2017,28(5):566-572.
[15]焦阿敏,董明利,娄小平,等.光条图像成像参数的模糊自适应调整方法研究[J].工具技术,2017,51(5):112-117.
[16]赵伟飞,康诚,钱登朝,等.城市隧道照明调光控制参数特征分析[J].低温建筑技术,2018(6).
[17]李春凯,石玗,顾玉芬,等.GTAW熔池三维表面激光点阵反射特征的仿真与优化[J].焊接学报,2017,38(9):42-46.