WO2010009570A1

WO2010009570A1 - A hoist-positioning method and intelligent vision hoisting system

Info

Publication number: WO2010009570A1
Application number: PCT/CN2008/001348
Authority: WO
Inventors: 于起峰; 张小虎; 姜欣; 桂阳
Original assignee: Yu Qifeng; Zhang Xiaohu; Jiang Xin; Gui Yang
Priority date: 2008-07-21
Filing date: 2008-07-21
Publication date: 2010-01-28

Abstract

A hoist-positioning method and intelligent vision hoisting system in which the video cameras (C1, C2) are mounted on each side of the crane boom respectively, the cooperation marks (P1, P2) are attached to the hook and the object to be hoisted respectively. The hoist-positioning method includes: firstly catching the cooperation mark attached to the sling of the object to be hoisted by the camera (C) mounted on the top of the crane tower, determining the orientation of the object to be hoisted, and controlling the rotation of the boom so as to cause the cooperation mark of the sling into visual field of the two cameras (C1, C2), using the two cameras mounted on the boom to form image of the hook and object simultaneously, and obtaining three-dimensional spatial coordinates of the hook and object; and then according to the three-dimensional spatial coordinates of the hook and object, calculating the parameter and sending it into the controller, after positioning the hook of the tower crane, automatically hoisting the object. The invention has a high-digitalized system and can meet the need of the high reliability and intellectualized by using the video camera measurement technology in the hoisting controlling, it thus lowers the labor strength of the operator greatly and improves the working efficiency.

Description

A lifting positioning method and lifting intelligent vision system.

Technical field

The invention relates to engineering fields such as tower cranes and cranes, as well as digital photogrammetry, digital image processing, computer vision and the like, and further relates to a lifting positioning method and a lifting intelligent vision system. Background technique

As a heavy transport vehicle, the lifting device has been widely used in workshops, docks, warehouses, construction sites and the like.

At present, lifting devices, such as tower cranes, mostly use cranes with traditional control methods, which require the staff to operate in a fixed operating room on the crane. The operating distance is long, and the operator does not see the lifting of the cargo and the place where the goods need to be placed. It is not possible to hoist and place the goods accurately, and the cargo conveying process requires at least two people to complete, the cargo conveying rate is low, and the productivity is low. In addition, the operating conditions of the crane operators are particularly harsh. The operators have long worked in the harsh environment such as the sun and sand. They have to withstand exposure to the sun and heat and cold. They are prone to fatigue, causing adverse reactions or diseases such as heat stroke, resulting in long driving operations. Not safe.

In recent years, the emergence of some lifting equipment such as remote control and video surveillance has greatly improved the working environment of the operators, so that the operators only need to work in the place where the goods are lifted or in a special office. But when lifting, it is still inseparable from human operations.

At the same time, in the book "Image-based Precision Measurement and Motion Measurement" published by Science Press, a "two-image intersection photogrammetry method" is mentioned, which indicates: using two or more cameras from different Under the condition that the position is shot on the target, the light beam and the image point of each camera should pass through the space object point, that is, each ray should intersect at the object point. Double image intersection photogrammetry method utilizes this original The rationalization of the intersection of the spatial objects is realized, which is referred to as the line intersection. Summary of the invention

The technical problem to be solved by the present invention is that, in view of the deficiencies of the prior art, a lifting positioning method and a lifting intelligent vision system for implementing the method are proposed in combination with the dual image intersection photogrammetry method, and the camera measurement is applied to the lifting control system. The high degree of digitization can meet the requirements of high reliability and intelligence of lifting and loading requirements, thereby greatly reducing the labor intensity of operators and improving work efficiency.

In order to achieve the above object, the technical solution adopted by the present invention is a lifting positioning method, which includes the following steps:

a. A camera device is mounted on each side of the beam of the lifting device, and a cooperation mark is mounted on the hook and the lifting ring of the object to be lifted;

b. controlling the rotation of the beam so that the cooperation mark on the hook and the hoist enters the field of view of the two camera devices of the beam;

c. Simultaneously imaging the hook and the lifting ring with two camera devices on the beam, then automatically extracting and coordinating the coordinates of the cooperative logo image, and using the double image intersection photogrammetry method to obtain the spatial three-dimensional coordinates of the cooperation logo, and obtaining the hook And the spatial three-dimensional coordinates of the lifting ring; the height difference H between the hook and the lifting ring and the horizontal distance L between the two in the direction of the beam, and the vertical plane of the beam to the vertical plane of the object are obtained according to the three-dimensional coordinates of the space of the hook and the lifting ring. Angle α _;

d. Through the parameters obtained above, the hook of the lifting device controlled by the controller automatically runs to the lifting ring of the lifting object and is automatically lifted.

As a preferred solution, in the above step a, a plurality of camera devices may be first installed at the top of the tower so that the field of view can cover a range of 360 degrees in the vicinity, and the cooperation device is captured by the tower top camera device. To determine the orientation of the lifting object, to achieve the purpose of automatically controlling the beam to rotate to the corresponding position. Correspondingly, the present invention also provides a lifting intelligent vision system for implementing the above lifting positioning method, comprising: an image capturing device respectively disposed on two sides of the lifting beam and a controller connected thereto, and the hook and the object to be lifted in the lifting device Cooperative signs are installed on the rings. Correspondingly, in the lifting intelligent vision system, the lifting tower top or the top of the crane is further mounted with a plurality of camera devices. And the camera device is preferably a digital camera. The cooperation logo may be a circle, a crosshair, or a apex shape.

It can be seen from the above that the lifting positioning method and the lifting intelligent vision system of the invention can meet the requirements of high reliability and intelligence of lifting and loading requirements, thereby greatly reducing the labor intensity of the operator and improving the working efficiency; applying the method of the invention When only the lifting device is factory-installed, the camera device is calibrated, and the mutual correspondence between the image point position and the space marker point position in the image is established, and the field of view of the camera device is covered by the field of view of the lifting object, so that in actual use, , high efficiency and good intelligence. DRAWINGS

1 is a schematic diagram of a lifting intelligent vision system in the embodiment; FIG. 2 is a cooperative marking point pattern in FIG. 1; wherein (1) is a circle, (2) is a crosshair, and (3) is a vertex angle;

DETAILED DESCRIPTION OF THE INVENTION The present invention will be further described below in conjunction with the accompanying drawings and embodiments. The lifting device in this embodiment is a tower crane, and the scheme of the lifting intelligent vision system is shown in FIG. 1 . For cameras mounted on the top of the tower crane that cover a 360-degree field of view, ς and c ₂ are two cameras mounted on the beam with a common field of view, and P ₂ are rings mounted on the hook and the object to be hoisted, respectively. Cooperation logo on.

The detailed steps of the lifting positioning method in this embodiment include:

a. The lifting intelligent vision system that implements this method is shown in Figure 1. A plurality of cameras C covering a range of 360 degrees are fixedly mounted on the top of the tower crane. The angles of the cameras described above are fixed to the beams, and (:, C ₂ is mounted on the beam. Two cameras in the common field of view,

P ₂ is a cooperation mark installed on the hook and the lifting ring respectively;

b. When the tower crane is installed at the factory, the camera calibration method is used to establish the mutual correspondence between the image point position and the space marker point position in the image;

c The tower crane coordinate system, with the tower base as the origin 0, the direction of the beam is the X axis, the vertical direction is the Z axis, and the vertical XOZ direction is the Y axis;

d. Using the tower camera C to capture the cooperation mark P ₂ on the lifting ring, by capturing the position of the cooperation mark point, the orientation of the lifting object can be determined, and the rotation of the beam can be controlled, so that the ring cooperation sign P ₂ enters the beam image. Inside the field of view of the casing, c ₂ ;

E. cooperation while two landmarks ^, P ₂ forming, carried out with two cameras on the beam image with sub-pixel positioning of the image processing technique, extraction cooperation landmarks ¾, P ₂ of the image point position.

f. From the image point position of the extracted cooperation mark point, according to the two-image intersection photogrammetry method, the two cooperation mark points are obtained; and the space coordinates (^, 0, Ζ, ) in the coordinate system, (Χ ₂ , Υ ₂ , z ₂ .

g. From the spatial coordinates ( , 0, Ζ, ), ( Χ ₂ , Υ ₂ , ζ ₂ ) obtained above, you can get: = Ζ ₂ —

L = ^X ₂ ² + Y^ ~Χ a = arc ^ where Η is the height difference between the hook and the sling, L is the horizontal distance between the two in the direction of the beam, and α is the vertical plane of the beam to the vertical plane of the object Angle;

h. Send the parameters obtained above to the controller, control the tower crane hook to automatically run to the lifting ring of the lifting object, and automatically lift.

The cooperation logo is shown in Fig. 2. It can be a circle, a diagonal shape, or a crosshair, or any other easily recognizable figure.

By using various existing digital image processing techniques, it is possible to detect and locate the image point position with high precision, and then obtain the spatial coordinates of the marker point by the two-image intersection photogrammetry method. For example, sub-pixel image localization is one of the advanced image processing methods that makes target positioning accuracy in images higher than the physical resolution of images. The present invention can employ the following sub-pixel positioning techniques:

1, using adaptive template correlation filtering

The basic idea is to make a template that can be adjusted. For each coarse positioning point, first determine the parameters of the template to be selected, select the most appropriate template, and use the selected template to perform correlation operations on the coarse positioning point and its neighbor points. The correlation coefficient fits the surface to determine the maximum relative position. 2. Adaptive Threshold Center of Gravity

For some targets, a target area with a certain area can be extracted by a variety of image processing methods, and the gray scale center of gravity method is used to determine the target area by using the gray scale as the weight in the target area. The gray center of gravity is used as the target position, and the Gaussian distribution template with adaptive threshold is used to track and locate the feature target.

3, gray map fitting method

For some targets, you can also directly select the appropriate analytic surface according to the characteristics of the target image, and then fit the surface of the grayscale image, and then find the extremum position of the analytic surface to achieve the sub-pixel precision positioning of the target.

4. Automatic identification of designated areas or full fields based on grayscale features

As shown in FIG. 2, the apex angle logo image has certain features, such as the diagonal region is bright or dark and the average gradation difference is small, the adjacent corner region is bright and dark, and the average gradation difference is large, and the four angular regions are respectively The average gray level is significantly different from the average gray level of the entire area. The average gray level of the center area is similar to the average gray level of the entire area. The gray level difference between pixels in the same angle range is small, and the outer edge of the mark has an ellipse or an approximate ellipse. Exist, there are two step edges that intersect at the center. By making full use of these features, high-precision identification and positioning can be reliably achieved.

By implementing the invention, the hardware device with the camera device and the personal computer (or DSP processor) as the core can be used with low degree of hardware, high degree of digitization and high degree of automation, and can be obtained by using a computer as a controller for the camera device. Image data is stored, copied, transferred, and automated.

Selection and installation of camera devices: Digital camera devices are easy to store and easy to implement post-processing numbers It is more suitable for analog cameras than digital cameras. Since the lifting equipment is accompanied by vibration during the operation of the crane. In order to ensure the normal working state of the camera device and correctly obtain the cooperative marker image, the camera device needs to be fixed on the lifting device.

Processor selection design: In the lifting intelligent vision system, PC computer or DSP processor can be used as the image storage and data processing device in the controller. Because the DSP processor is fast and easy to operate, it is more suitable for use in practical devices. When the camera is selected, the DSP processor can be designed according to the interface scheme of the camera.

The lifting device may be any one of the existing lifting devices, such as a tower crane or a crane in the embodiment, and no additional design is required.

Claims

Rights request

A lifting positioning method, comprising the steps of: a. mounting a camera device on each side of the beam of the lifting device, and installing a cooperation mark on the hook and the lifting ring of the object to be lifted; b Rotating the beam of the lifting device so that the cooperation marks on the hook and the lifting ring enter the field of view of the camera on both sides of the beam; c. Simultaneously imaging the hook and the lifting ring with two camera devices on the beam, and then automatically extracting Collaborate to mark the image coordinates and locate them. Use the double image intersection photogrammetry method to obtain the spatial three-dimensional coordinates of the cooperation mark, and obtain the spatial three-dimensional coordinates of the hook and the lifting ring. According to the space three-dimensional coordinates of the hook and the lifting ring, the hook and the lifting ring are obtained. The height difference H between them and the horizontal distance L of the two in the direction of the beam, and the angle α of the vertical plane of the beam to the vertical plane of the object;

d. Through the parameters obtained above, the hook of the lifting device controlled by the controller automatically runs to the lifting ring of the lifting object and automatically lifts.

2. The lifting positioning method according to claim 1, wherein in the step a, a plurality of camera devices are first installed on the top of the tower so that the field of view can cover a range of 360 degrees in the vicinity, and the tower top camera device is used. Capture the cooperation mark on the lifting ring to determine the orientation of the lifting object and automatically control the beam to rotate to the corresponding position.

3. The lifting positioning method according to claim 1, wherein the step of obtaining the parameter comprises:

a. When the lifting device is installed, firstly, the camera calibration method is used to establish the mutual correspondence between the image point position and the spatial mark point position in the image.

b. The lifting equipment coordinate system, with the tower base as the origin 0, the direction of the beam is the X axis, vertical The direction is the Z axis, and the vertical XOZ direction is the Y axis;

c. Using the tower top camera C to capture the cooperation mark P ₂ on the lifting ring, by capturing the position of the cooperation mark point, the orientation of the lifting object can be determined, and the beam rotation can be controlled, so that the ring cooperation sign enters the beam two camera. And c ₂ in the field of view; d. using two cameras on the beam to simultaneously image the two cooperation mark points ^, ₂ , using the sub-pixel image localization technology to perform the image processing, extracting the image of the cooperation mark points ^, ₂ Point position; e. From the image point position of the extracted cooperation mark point, according to the double image intersection photogrammetry method, the space coordinates ( , 0, Z ) of the two cooperation mark points and P ₂ in the coordinate system are obtained, (X ₂ , Y ₂ , ζ ₂ ) ; f. From the spatial coordinates ( 0, ), (Χ ₂ , Υ ₂ , ζ ₂ ) obtained above, you can get: = Ζ ₂ -Ζ _{

a - arc tan-

X ₂ where H is the height difference between the hook and the sling, L is the horizontal distance of the two in the direction of the beam, and a is the angle from the vertical plane of the beam to the vertical plane of the object.

4. The lifting positioning method according to claim 1, wherein the image point position of the cooperation mark point is image processed by a sub-pixel image localization technique, thereby extracting the image point position.

5. A lifting intelligent vision system, comprising: a camera device respectively located on both sides of the lifting beam and a controller connected thereto, and the hooks in the lifting device and the lifting ring of the object to be lifted are respectively respectively With a cooperation logo.

A lifting intelligent vision system according to claim 5, wherein a plurality of imaging devices are mounted on the hoisting tower top or the top of the crane.

7. The lifting intelligent vision system according to claims 5 and 6, characterized in that the camera device is a digital camera.

8. The lifting intelligent vision system according to claims 5, 6, wherein the cooperation mark is a circular shape, a crosshair, or a apex shape.