Visual identification-based grain transport vehicle grain tank loading state detection method

Abstract

The invention relates to a grain transport vehicle grain tank loading state detection method based on visual recognition. The method comprises steps of arranging a plurality of depth cameras at the edge of a grain tank, and collecting the loading state information in the grain tank; constructing a 3D model graph at the corner of the grain tank body, and performing multi-view sampling on the 3D model graph to form a template database which is used for template matching; calibrating the plurality of depth cameras, and constructing a mapping relationship between the depth cameras and a grain tankcoordinate system; performing filtering and noise reduction processing on the point clouds acquired by the plurality of depth cameras to obtain effective point cloud information of the grain tank; and performing point cloud fusion on the effective point cloud information of the grain tank to obtain the spatial distribution information of grains loaded in the grain tank. According to the calibration method of the depth camera based on template matching, filtering and noise reduction processing is carried out on the point cloud, and a loading state can be effectively detected; the loading stateof the grain tank is detected through visual identification, the spatial distribution state of grain in the grain tank can be accurately described, applicability is high, and maintenance is convenient.

Description

technical field [0001] The invention relates to visual recognition three-dimensional reconstruction in the field of cooperative operation of agricultural machinery, in particular to a visual recognition-based method for detecting the loading state of a grain tank of a grain transport vehicle. Background technique [0002] Intelligent management of agricultural machinery operation and operation and maintenance is an important technical support for the transformation of agricultural production methods, the transformation and upgrading of agricultural machinery enterprises, and the improvement of the quality and efficiency of agricultural machinery operation services. The mechanization of agricultural production in my country is developing rapidly, but there is a lack of decision-making methods for agricultural machinery scheduling and collaborative operations, insufficient accumulation of data on agricultural machinery operating conditions, low accuracy of fault warning and dia...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the prior art, the present invention provides a visual recognition-based detection method for the loading state of the grain tank of the grain truck, which solves the problem that the grain loading status of the grain tank of the grain truck needs to be detected and the spatial distribution status information of the grain in the grain tank is difficult to detect The problem

Images