Projection coding method for surface texture enhancement of aviation luggage in 3D modeling

Abstract

A projection coding method for surface texture enhancement of aviation luggage in 3D modeling is disclosed. The method comprises: setting a coding rule of pixels of a restraint random encoded image, encoding the pixels of a projected image by using the projected image pixel coding rule under a condition of satisfying uniqueness constraint of local feature constraint, to generate a restraint randomencoded image of high-intensity texture distribution projected onto an aviation luggage surface; and using an improved particle swarm optimization algorithm of dynamic speed-dynamic population to optimize and solve a threshold value of gray difference among pixels of the restraint random encoded image, to obtain an optimized restraint random encoded image; projecting the optimized restraint random encoded image to a weak texture aviation luggage surface, and obtaining 3D point cloud information of aviation baggage shapes after 3D reconstruction. The method can improve reliability of stereo matching of aviation luggage and accuracy of 3D reconstruction.

Description

[0001] technology neighborhood [0002] The invention belongs to the technical field of civil aviation, and in particular relates to a projection encoding method for enhancing the surface texture of air luggage in three-dimensional modeling. Background technique [0003] Civil aviation self-service check-in machines have been widely used in major airports, and self-service baggage drop-off systems are rapidly being popularized. This system needs to automatically detect the luggage dropped by passengers and judge whether its size and shape meet the requirements for check-in. It is a reliable solution to perform size detection and shape recognition based on the three-dimensional information of luggage. In the luggage 3D information collection solution, the 3D laser scanner is expensive and the line laser scanning speed is slow. The traditional stereo vision method is not reliable for 3D modeling of weakly textured luggage, and cannot meet the technical or price requirements of t...

AI Technical Summary

Problems solved by technology

However, due to its randomness, it will cause a certain probability of mismatching, which increases the uncertainty of reconstruction; the projection gray code method performs three-dimensional reconstruction on the surface of a static object, and its reconstruction accuracy is high; the projection is based on the M-array template The spatial domain encoding method concentrates a variety of encoding information in the projected image, and performs real-time 3D reconstruction after projecting onto the target surface.

However, the cost of precision optical instruments required by these two encoding methods is relatively high, which is not suitable for large-scale promotion and application.

Images