A holographic stereoscopic image matching algorithm provided by the invention belongs to the technical field of stereoscopic display. Decomposition coefficients and image information of low frequencyand high frequency parts are obtained by image wavelet transform, and at the same time, the noise visibility function is used for performing threshold processing to extract a part of wavelet coefficients which contribute most to participate in matching, and the noise visibility function is used to select the parallax with the least noise in each parallax image to effectively improve the final accuracy of the parallax image. The noise visibility function has been frequently used in various fields of image compression processing. Aiming at an original image pairs, wavelet wavelet optimization isperformed, which is an innovative point different from the traditional algorithm, and can effectively enhance the matching effect and accuracy of the algorithm.