基于脉冲耦合神经网络的红外图像拼接

创建高分辨率的宽视角的拼接图像为图像处理、计 算机图形学等交叉学科学研究的新领域,针对传统的基于边缘的拼接算法对噪声比较敏感, 红外图像噪声干扰严重与信噪比低的特点,提 出一种基于脉冲耦合神经网络(PCNN,pulse coupled neural network)的红外图像拼接算法 。采用PCNN仿生物视觉角度,提取图像的边缘信息；采用Hausdorff距离作为配准的相似性 测度,计算出最优配准参数进行图像拼接；使用加 权平均方法,实现拼接图像的融合,提高拼接图像的视觉效果。实验结果表明,本文算法能 够实现图像的精确拼接,对噪声具有较好的鲁棒性,并提高了搜索效率,减小了计算量。

Infrared image mosaic based on pulse coupled neural network

The automatic construction of large,hi gh-resolution image mosaics is an active area of research in the fields of imag e processing,computer vision and computer graphics.Due to the traditional edge-based algorithm is sensitive to noise,also because of the infrared image noise inte rference and low signal to noise ratio characteristics,an image mosaic algorithm based on pulse coupled neural network (PCNN) is presented in this paper.First,t he pulse coupled neural network imitation of biological vision is used to extract t he edge information,the Hausdorff distance is applied as the registration similarity measure,and then the optimal reg istration parameters are calculated for image mosaicing,simultaneously using the weighted average method to achieve image mosa icing fusion,which results in an optimal image mosaic and significantly improved quality of the image mosaic.The experimental results show that the algorithm can achieve precise image mosaicing and has better r obustness to noise.