Ship loader level detecting method based on laser radar

Abstract

The invention discloses a ship loader level detecting method based on a laser radar, and provides a method with high recognition precision and capable of being used in adverse weathers. A two-dimension laser radar is mounted on a two-axis scanning drive mechanism, a length value in a polar coordinate system utilizing a laser radar detection center as a pole is obtained by decoding real-time scanned data of the two-dimension laser radar, an angle value is obtained according to a division value acquired during detection, a measured value in the polar coordinate system is converted into a right-angle coordinate system by a conversion equation, images are subjected to smooth filtering by a gaussian filter, edge strength of the images is obtained by combination of a difference algorithm, Hough transform and wavelet transform, and then retaining the maximum point of local gradients according direction of the gradients; edge recognition is carried out; two-dimensional space point cloud is obtained by final matching according to a least square method; three-dimensional space point cloud is then obtained by advancing the two-dimension laser radar along a storage yard; and finally point cloud of discrete data is subjected to triangulation.

Description

technical field [0001] The invention relates to the technical field of port and wharf equipment, in particular to a laser radar-based material level detection method for a ship loader. Background technique [0002] During port bulk cargo loading and unloading operations, real-time monitoring must be carried out on the distribution, maximum height, minimum height, volume, etc. of the material in the cabin to determine the subsequent material drop point and material trajectory. At present, these data are all observed and estimated by the staff on board with the naked eye, or assisted by the driver through the camera installed in the driver's cab of the cart. However, the manual observation results will be interfered by many factors such as staff experience, weather, etc., and it is difficult to guarantee the accuracy. If the staff is slightly negligent or the measurement accuracy is inaccurate, it is easy to cause accidents such as burying the throwing shovel, capsizing, sink...

AI Technical Summary

Problems solved by technology

The vision-based 3D inspection system needs to use a binocular system or a structured light plus monocular system. The binocular vision system needs to match binocular images, and the algorithm is complex. Currently, it is rarely used in large-scale real-time 3D measurement in research.

Structured light plus monocular systems are generally studied for fine and short-distance measurement. For large-scale measurement of bulk cargo, the imaging of laser beams on bulk cargo with long-distance low reflectivity is poor, and image processing is complicated, so it is not suitable for this environment.

Using the principle of distance measurement, affected by the physical characteristics of the sensor, only single point or two-dimensional measurement can be performed each time. For three-dimensional measurement, two-dimensional rangefinders and rotating equipment are generally used. Currently, the research on three-dimensional measurement systems focuses on the three-dimensional measurement of the measured object. Reconstruction, focusing on accuracy, eliminating measuring instrument errors, long measuring time, while the position of the measuring system does not change during the measuring process

Images