用于HVS的拉普拉斯金字塔变换图像融合研究
|
摘要
平视视景系统(HVS)融合了增强视景系统(EVS)和合成视景系统(SVS),在飞机近陆运行阶段,为飞行员提供实时外景信息,提高飞行安全。提出了一种用于HVS的图像融合方法:首先将源图像进行拉普拉斯金字塔变换,然后对高频层采用绝对值取大的方法进行融合,低频层采用区域能量与改进的区域平均梯度相结合的方法进行融合,最后进行拉普拉斯反变换得到最终融合图像。实验表明该方法用于EVS与SVS图像的融合时具有良好的融合效果。
The Head-up Vision System( HVS) integrates the Enhanced Vision System( EVS) with the Synthetic Vision System( SVS) and provides real-time location information for the pilot to improve flight safety during the aircraft's near-land operation. A new image fusion algorithm used in HVS is proposed in this paper. First Laplacian pyramid transform is carried out for each source image. Then the method of selecting the larger absolute value is used on the high-frequency part and the method of local energy combined with local average gradient is used for the fusion of the low-frequency part. Finally the fused image is obtained by inverse Laplacian pyramid transform. The experimental results show that this algorithm has good fusion effect in the fusion of EVS image with SVS image.
The Head-up Vision System( HVS) integrates the Enhanced Vision System( EVS) with the Synthetic Vision System( SVS) and provides real-time location information for the pilot to improve flight safety during the aircraft's near-land operation. A new image fusion algorithm used in HVS is proposed in this paper. First Laplacian pyramid transform is carried out for each source image. Then the method of selecting the larger absolute value is used on the high-frequency part and the method of local energy combined with local average gradient is used for the fusion of the low-frequency part. Finally the fused image is obtained by inverse Laplacian pyramid transform. The experimental results show that this algorithm has good fusion effect in the fusion of EVS image with SVS image.
引文
[1] POHL CVAN GENDEREN J L. Multisensor image fusion in remote sensing:conceptsmethods and applications[J]. International Journal of Remote Sensing199819(5):823-854.
[2]贾欢.基于多尺度分解的图像融合算法及其实现[D].北京:北京理工大学,2014.
[3] BURT P JADELSON E H. The Laplacian pyramid as a compact image code[J]. IEEE Transactions on Communications1983COM-31(4):532-540.
[4] BURT P JKOLCZYNSKI P J. Enhanced image capture through fusion[C]//IEEE 4th International Conference on Computer Vision1993:173-182.
[5] DO M NVETTERLI M. Framing pyramids[J]. IEEE Transactions on Signal Processing200351(9):2329-2342.
[6]邢素霞,吴静珠,陈天华.图像融合质量评价方法研究[J].微电子学与计算机,2010,27(6):157-159.
[7]王跃华,陶忠祥.红外与可见光图像融合质量评价方法综述[J].红外,2012,33(6):7-11.
[8] GONZALEZ R C.数字图像处理(MATLAB版)[M].阮秋琦,译.北京:电子工业出版社,2005.
[9]张瑶,刘生辉.基于MATLAB的图像融合技术[J].中国科技信息,2016(1):55-56.
[10]韩潇,彭力.基于改进拉普拉斯金字塔的图像融合方法[J].自动化与仪器仪表,2014(5):191-194.
[2]贾欢.基于多尺度分解的图像融合算法及其实现[D].北京:北京理工大学,2014.
[3] BURT P JADELSON E H. The Laplacian pyramid as a compact image code[J]. IEEE Transactions on Communications1983COM-31(4):532-540.
[4] BURT P JKOLCZYNSKI P J. Enhanced image capture through fusion[C]//IEEE 4th International Conference on Computer Vision1993:173-182.
[5] DO M NVETTERLI M. Framing pyramids[J]. IEEE Transactions on Signal Processing200351(9):2329-2342.
[6]邢素霞,吴静珠,陈天华.图像融合质量评价方法研究[J].微电子学与计算机,2010,27(6):157-159.
[7]王跃华,陶忠祥.红外与可见光图像融合质量评价方法综述[J].红外,2012,33(6):7-11.
[8] GONZALEZ R C.数字图像处理(MATLAB版)[M].阮秋琦,译.北京:电子工业出版社,2005.
[9]张瑶,刘生辉.基于MATLAB的图像融合技术[J].中国科技信息,2016(1):55-56.
[10]韩潇,彭力.基于改进拉普拉斯金字塔的图像融合方法[J].自动化与仪器仪表,2014(5):191-194.