Research on road image fusion enhancement technique based on wavelet transform

This paper uses image sensor to gain images of low visibility road and discusses multi-distinguishing wavelet transform, reconstruction and image fusion rules of the low visibility roadpsilas images. Depending on the different spatial frequency character of the details components between the different decompositions, a choice fusion operator based on regional characteristic measure is adopted to enhance the images of low visibility road. Based on six indicators, including the standard deviation, entropy, RMSE, NLSE PSNR and the average gradient, a series of images are obtained by different fusion rules are analyzed and compared in this paper. We will find that, on the regional characteristic measure, the regional energy is best of all, the region max is second, the region variance and entropy are as good as region area. Through comparing above six indicators, we can see that the experiment results based on wavelet transform better than those, which are obtained through the weighted average arithmetic or Laplacian pyramid transform.