Depth map denoising method based on color image segmentation

Abstract

The invention discloses a depth map denoising method based on color image segmentation. The depth map denoising method firstly performs chromaticity space conversion and grey-scale conversion on color maps; secondly, adopting Sobel operators to convert a grey-scale map into a gradient map, performing watershed segmentation based on the gradient map and dividing the color map into a plurality of areas; then calculating the Euclidean distances of adjacent domains under an LUV space, calculating gradient mean value differences of the junction of the adjacent domains according to the gradient map, performing domain fusion by utilizing the two groups of information, combining the resemble domains and marking the communicated domains; and finally marking the depth map communicated domains according to the color map and the depth map, and performing cavity compensation and denoising treatment on the depth map by utilizing the characteristics that the depth of the same areas of the depth map are approximately equal. The depth map denoising method has the advantages of being remarkable in denoising effect, capable of being achieved through hardwares and the like.

Description

technical field [0001] The invention relates to the technical field of image processing and natural interaction, in particular to the technology of hole compensation and denoising of image depth maps. Background technique [0002] Depth maps are indispensable information for converting 2D images into 3D scenes. The active vision mode based on structured light can obtain the depth information of the image more accurately. Compared with the binocular stereo camera, the acquired depth map information of this mode is more stable and reliable, not affected by ambient light, the stereo matching process is simple, and the algorithm calculation amount Small and other advantages. For example, Kinect, the somatosensory interactive device of Microsoft, adopts the active vision mode of infrared structured light, that is, the infrared laser projects a fixed pattern image onto the surface of the object, forms speckle spots through the diffuse reflection of the surface of the object, ...

AI Technical Summary

Problems solved by technology

The filter denoising method can achieve different filtering effects by selecting different filters and their parameters, which will damage a certain source image, blur the edge of the image, and cause image distortion

The non-local image denoising method has a good denoising effect on local small holes, but the denoising result on edge shape distortion is not ideal, and the edge information with large distortion in the image cannot be restored.

These denoising methods are not ideal for noise caused by large-area voids

Images