改进的神经网络非均匀性校正算法研究

针对传统神经网络非均匀性校正算法所存在的像素点期望值估计不准确、场景长时间静止时校正图像发生衰退以及校正图像出现鬼影的问题,在原有的神经网络算法的基础上做了进一步的改进。主要包括三个部分:首先,计算场景的局部方差,设置合理判断阈值,区别不同的场景区域;再通过比较前后两帧场景的绝对误差值,判别场景是否静止,来控制校正参数是否更新;然后利用图像配准的方法,计算出帧间位移,对校正步长做出修正。最后与传统神经网络算法进行对比实验,结果表明,改进的神经网络算法在鬼影抑制和非均匀性校正方面都具有较好的效果。

Research on Improved Neural Network Non-uniformity Correction Algorithm

Abstract] The traditional Neural network non-uniformity correction algorithms exists following questions: The expectations of pixels are estimated inaccurate, when the scene for a long time still the corrected image will degrade, and the ghosting will appear in corrected image. Improved neural network algorithm based on the original algorithm made further improvements to solve the above questions. Mainly includes three parts: First, calculate the local variance of scene, set a reasonable judgment threshold to distinguish between different areas of the scene; Next, compare the absolute error of two adjacent scenes ,discriminant the scene is stationary or not to determine whether to update the correction parameters; Then, using image registration to calculate the inter-frame displacement and correct the correction step. Finally, to do a experiment compare with the traditional neural network algorithm. Experimental results show that the improved neural network algorithm in this paper has better effect correction in suppressing ghosting and non-uniformity.