Measuring method, device, system and medium based on binocular microscopic vision

Abstract

The present application discloses a measurement method, device, system and medium based on binocular microscopic vision. Constraint relationship, perform image correction to obtain the left and right images; down-sample the left and right images, perform stereo matching and matching point propagation on the pixels in the sampled image of the corresponding layer with the smallest resolution in the obtained left sampled image set, and perform layer-by-layer sampling and Stereo matching update and interpolation processing to obtain dense disparity results; then measure the object to be tested according to camera parameters and dense disparity results. The embodiment of the present application solves the problem of inapplicability and low accuracy of the measurement method on the macro scale in the prior art to the micro scale, and the measurement method provided is very suitable for binocular measurement in the micro scene, and can obtain good accuracy and precision of microscopic measurements.

Description

technical field [0001] The present application relates to the technical field of visual measurement, and in particular to a measurement method, device, system and medium based on binocular microscopic vision. Background technique [0002] Microscopic stereoscopic measurement technology is an important means for three-dimensional measurement and surface reconstruction of observed objects and exploration of the microscopic world. Typical methods of microscopic stereo measurement include defocus depth measurement, microscopic binocular imaging, laser scanning measurement, shadow shape acquisition, etc. Among them, microscopic binocular imaging is widely favored by workers in the fields of precision machining and failure analysis due to its advantages of simple and fast imaging, good robustness, and high precision. [0003] Microscopic binocular imaging measurement currently mainly relies on stereo microscopes (also known as dissecting microscopes, solid microscopes or stereo m...

AI Technical Summary

Problems solved by technology

At present, there are few methods for microscopic stereo matching, and most researchers directly apply the above macroscopic matching methods to microscopic matching.

However, the monotonous color and monotonous texture of the observation sample in the microscopic world lead to unsatisfactory results after the direct transfer method, and a more targeted microscopic stereo matching method is worthy of being proposed.

For example, the existing stereo matching method based on deep learning has weak generalization ability, and the microscopic scene is quite different from the traditional stereo matching scene, so the performance in the microscopic scene is poor.

Images