In this paper

a weight based Polar code puncturing method is proposed and applied to the image transmission under the Gaussian channel.This method encodes with the same weighted bit of the gray value of the image

and determines the number of deleted bits according to the weighted value of the gray value of the image.The principle indicates that bigger weight means less number of punctured bits and smaller weight means larger number of punctured bits.Peak Signal to Noise Ratio（PSNR）is used to evaluate the performance of reconstructed images.The simulation results show that under the image transmission of Gaussian channel with signal to noise ratio of 6 dB

the PSNR for the improved punctured Polar codes is 18 dB higher than the Polar code with fixed punctured code rate.The performance improvement is more significant at the condition of larger signal to noise ratio.