The accuracy of iris recognition is lowered due to the shelter by part of eyelid and eyelash during the process of iris detection.A noise detection method using the 2D-Gabor filter is proposed for it.Firstly,the eyelid and eyelash are detected using 2D-Gabor filter in the normalized iris images.Then the noise regions are marked.Finally,the interferential bits in the iris codes are eliminated to avoid the noise effect to the recognition as much as possible.The experiments by using the CASIA(3.0)database are finished.The result shows that the recognition rate with the noise detection is 99.02%,but the recognition rate without the noise detection is 98.46%.