The invention relates to the technical field of computers, in particular to a method for conducting three-dimensional point cloud denoising based on a two-dimensional binary image processing. The method comprises the following steps of S1, obtaining data; inputting a point cloud file path to obtain the data in a point cloud file; S2, rotating obtained three-dimensional point cloud coordinates fora certain angle; S3, projecting the rotated three-dimensional point cloud image to a two-dimensional plane; S4, storing the projected two-dimensional pixel point data into a bmp format; S5, utilizingopenCV to remove noisy points in a two-dimensional plane; S6, obtaining a picture pixel of which the noisy point is removed; S7, obtaining files of pictures with a noisy point and without a noisy point respectively through the picture pixel obtained in the step S6; S8, analyzing the picture before the noisy point is removed and the picture after the noisy point is removed. Through a multi-angle two-dimensional binary image, three-dimensional point cloud denoising is conducted, and a large quantity of noisy points in three-dimensional point cloud can be effectively removed.