Monocular vision box body volume measuring method based on multi-line structural light image recognition

Abstract

The invention relates to a monocular vision box body volume measuring method based on multi-line structural light image recognition. The monocular vision box body volume measuring method comprises thefollowing steps that (1) a camera is calibrated, and multi-line laser plane parameters are acquired; (2) calculation of mapping of two-dimensional coordinates to three-dimensional spatial coordinatesis conducted; (3) the box edges are detected and the laser central line is extracted by using a deep learning method; and (4) structural light with a well shape, structural light with a dual-parallel-line and structural light with a cross shape are correspondingly used for measuring different box bodies to obtain the volumes of inerratic box bodies. According to the monocular vision box body volume measuring method based on multi-line structural light image recognition, the multi-line structural light (the well shape, the cross shape and the parallel line shape) is casted on the measured boxbody plane, intersection points of the line structural light and the edges of the boxed are detected through imaging processing, the points are converted from the two-dimensional coordinates to three-dimensional spatial coordinates according to a structural light triangulation measuring theory to calculate the lengths, the widths and the heights of the inerratic box bodies, and the volumes are further calculated. According to the monocular vision box body volume measuring method based on multi-line structural light image recognition, the complexity of the algorithm is low, the method can be greatly applied to PC terminals and mobile terminal devices, real-time on-line measuring can be conducted, and the requirement of portability is met.

Description

technical field [0001] The invention belongs to a method for measuring the volume of a box, in particular to a method for measuring the volume of a box with monocular vision based on multi-line structured light image recognition. Background technique [0002] Box volume measurement is mainly used in logistics, e-commerce, warehouse storage, aviation warehouses and other fields, providing various services such as freight calculation, cargo placement, box recommendation, etc. Its application scenarios are relatively complex, so how to quickly, accurately and efficiently measure the volume of the box in a complex environment is an urgent problem to be solved in these industries. In addition, how to develop portable volume measuring instruments on the basis of good performance is also a problem for these industries. Urgent needs. At present, there are many measurement methods: manual detection, manual measurement by ruler, which is not only inefficient but also prone to errors;...

AI Technical Summary

Problems solved by technology

Its application scenarios are relatively complex, so how to quickly, accurately and efficiently measure the volume of the box in a complex environment is an urgent problem to be solved in these industries. In addition, how to develop portable volume measuring instruments on the basis of good performance is also a problem for these industries. Urgent needs

At present, there are many measurement methods: manual detection, manual measurement by ruler, which is not only inefficient but also prone to errors; grating measurement method, automatic measurement by equipment, high measurement accuracy, but it requires manpower to move the box to a fixed position. It is very difficult to measure a box higher than 1 meter, which is labor-intensive and the cost of the instrument is high; the volume measurement method based on binocular stereo vision has high algorithm complexity, difficult stereo matching, low accuracy in measuring large-sized boxes, and is not suitable for logistics, etc. Complex Measurement Scenarios

Images