Turnover box automatic positioning method based on image geometrical characteristics

Abstract

The invention provides a turnover box automatic positioning method based on image geometrical characteristics, and the method comprises the steps: S1, obtaining a vertical edge and a horizontal edge according to an original image, and obtaining a region of interest (Roi image); s2, carrying out binarization processing; s3, removing noise points; s4, carrying out straight line detection on the binary image by adopting a Hough transformation strategy to obtain a plurality of approximate straight lines, and then fusing the approximate straight lines into a fused straight line by adopting a straight line fusion method; s5, obtaining to-be-selected edges, if the number of the to-be-selected edges is zero, returning to the step S2, reducing the offset C, and continuing iteration until at least one to-be-selected edge is obtained; s6, determining the color of the local image to filter out false edges; and S7, integrating the edges, and calculating a central point and a rotation angle to realize positioning. The turnover box can be positioned in the visible light image, consumed time is short, and manual participation is not needed.

Description

technical field [0001] The invention relates to the technical field of image analysis, in particular to an automatic positioning method for a turnover box based on image geometric features. Background technique [0002] Locating the target of interest in the image is a key technology in the fields of image processing, machine vision, etc. Its purpose is to find the target of interest in the image, extract its position information in the image, and then calibrate it through the camera Convert its position to the world coordinate system to lay the foundation for subsequent operations such as robot grasping. The short-distance transportation of materials is an essential link in scenarios such as warehouses and assembly lines. The turnover box is a standardized material container, which is widely used in the circulation and storage of materials. At present, its handling tasks are mainly completed manually, and the existing problems are: (1) high labor cost; (2) low efficiency;...

AI Technical Summary

Problems solved by technology

At present, its handling tasks are mainly completed manually, and the existing problems are: (1) high labor cost; (2) low efficiency; (3) error rate is uncontrollable, and the source of error cannot be traced

[0004] The segmentation-based method has poor positioning effect on images with small differences in grayscale between the target and the background, and it is difficult to completely separate the target from the background, so the positioning accuracy is low, and it is difficult to guide the robot arm to grasp

The edge-based method is sensitive to noise, and the accuracy of some turnover box images with no obvious difference between the gray level and the background is low, and the recognition rate is difficult to meet the industrial-grade requirements

The biggest problem with the feature-based method is that the process of extracting and analyzing features takes a long time, and it is difficult to distinguish some false targets that are similar to the target features.

The method based on deep learning requires a large number of labeled samples for training the model, and the final positioning accuracy is affected by the labeling accuracy, and the positioning results may be unstable, which is not suitable for precise grasping

[0005] The turnover box positioning problem has the following difficulties: (1) The image background of the turnover box is complex, and there are various types of internal materials, which may interfere with target recognition; (2) The distance between the turnover boxes is small, and adjacent boxes may interfere with each other

(3) For gray turnover boxes, positioning based on color features will lead to large errors

(4) The turnover box may be deformed during use, and the shape of each turnover box will not be exactly the same

[0006] Existing methods are difficult to achieve precise positioning of turnover boxes

Images