Microscopic three-dimensional reconstruction method based on Markov random field constraint

Abstract

The invention discloses a microscopic three-dimensional reconstruction method based on a Markov random field constraint. The method comprises the following steps: scanning and shooting an object by using a microscope to obtain a plurality of images with different focusing degrees, and forming a multi-focusing image sequence; performing non-subsampled contourlet transformation on each image in the image sequence, and constructing a low-pass filter bank distributed in a pyramid shape to process a non-subsampled contourlet transformation result; comparing different image processing results to obtain depth information and full-focus information implied in the image sequence; correcting the depth information by using the prior constraint of the Markov random field; and constructing a three-dimensional form of the microscopic image by using the depth information of the obtained multi-focus image sequence. The algorithm is simple in implementation principle and can be effectively applied to the fields of fine structure detection, ultra-precision machining, medical operation and the like. The method solves the problem that in the prior art, a focusing evaluation function is difficult to balance the detection sensitivity and the noise robustness, and improves the reconstruction capability of the weak texture sample.

Description

technical field [0001] The invention relates to technical fields such as digital image processing and computer vision, and in particular to a microscopic three-dimensional reconstruction method based on Markov random field constraints. Background technique [0002] The development of precision machining technology puts forward higher requirements for microscopic three-dimensional reconstruction technology. The key to 3D reconstruction technology is to accurately extract the depth information and all-focus information of the microscopic sample. In this case, due to the large measurement range of the focusing method and the low requirement for the smoothness of the surface of the sample to be tested, it has become a reliable method for quality inspection in the field of precision machining, especially for samples with large inclination angles such as cutting tools. detection. [0003] Although the focusing method has the ability to quickly obtain the depth information and al...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a microscopic three-dimensional reconstruction method based on Markov random field constraints, which solves the problem that the focus evaluation function in the prior art is difficult to balance detection sensitivity and robustness to noise, and improves the Reconstruction Ability of Weakly Textured Samples

Images