CN-116309777-B - Tower crane suspended object height estimation method and device based on target identification

CN116309777BCN 116309777 BCN116309777 BCN 116309777BCN-116309777-B

Abstract

The invention provides a tower crane lifting object height estimation method and device based on target identification, comprising the steps of obtaining tower crane lifting hook size information and lifting object density information, obtaining lifting object weight, distance from the bottom end of a lifting hook to the top end of a lifting object and lifting hook ground clearance, obtaining a lifting object nodding video image, wherein the video image comprises a complete lifting hook image and a lifting object image, calculating pixel ratio of the lifting hook to the lifting object in the video image, calculating lifting object area according to the lifting hook size information, distance from the bottom end of the lifting hook to the top end of the lifting object, lifting hook ground clearance and pixel ratio, judging the type of the lifting object through target identification, and further calculating lifting object height and lifting object bottom ground clearance. The invention realizes the identification of the type of the suspended object, the identification of the plane size and the estimation of the height of the suspended object, realizes the judgment of the height of the suspended object from the ground, provides finer and more practical data support for the safe operation of the tower crane, and improves the safety management and control efficiency.

Inventors

YANG XIAOJIAO
HE YUECHUAN
PENG YAN
YANG GUANG

Assignees

四川省建筑科学研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20221207

Claims (9)

1. The method for estimating the height of the crane object based on the target identification is characterized by comprising the following steps of: acquiring tower foundation lifting hook size information and lifting object density information; Acquiring weight information of a suspended object, judging the distance from the bottom end of a lifting hook to the top end of the suspended object after the suspended object is lifted, and acquiring the ground clearance of the lifting hook; The method comprises the steps of acquiring a video image of a hung object in a nodding mode, wherein the video image comprises a complete lifting hook image and a hung object image; Calculating the area of the suspended object according to the size information of the suspended object, the distance from the bottom end of the suspended object to the top end of the suspended object, the height of the suspended object and the pixel ratio, and combining the weight information of the suspended object and the density information of the suspended object, wherein the step of calculating the area of the suspended object comprises the following steps: ; Wherein S d is the overlooking area of the lifting hook calculated according to the size information of the lifting hook, H is the distance from the lifting arm to the ground, H j is the height of the lifting hook, H g is the height of the lifting hook from the ground, H s is the distance from the bottom end of the lifting hook to the top end of the lifting object, and S '/S d ' is the pixel ratio of the lifting hook to the lifting object in the video image; and calculating the ground clearance of the bottom end of the suspended object according to the height of the suspended object, the distance from the bottom end of the suspended object to the top end of the suspended object and the ground clearance of the suspended object.
2. The method for estimating a height of a tower crane based on object recognition according to claim 1, wherein the tower base hook size information includes a hook length, a hook width, and a hook height.
3. The method for estimating a height of a crane suspended object based on object recognition according to claim 1, wherein the step of acquiring the suspended object density information comprises: pre-storing hanging object density information corresponding to different hanging object types; and automatically acquiring hanging object density information according to the current hanging object type.
4. The method for estimating a height of a crane according to claim 1, wherein the step of calculating the height of the bottom end of the crane from the ground according to the height of the crane, the distance from the bottom end of the hook to the top end of the crane, and the height of the hook from the ground comprises: ; wherein, H g is the height of the lifting hook from the ground, H s is the distance from the bottom end of the lifting hook to the top end of the lifting hook, M is the weight of the lifting hook, ρ is the density information of the lifting hook, S d ' is the pixel area of the lifting hook in the video image, and S d is the overlooking area of the lifting hook calculated according to the size information of the lifting hook.
5. The method of claim 1, wherein the step of identifying hook and hook boundaries in the video image using a target recognition algorithm comprises: Dividing the image of each frame of video image into N grid units, identifying the boundary of a target suspended object and the boundary of a lifting hook in the grid, judging the type of the suspended object, acquiring the density of the corresponding type in a database, identifying, calculating the occupied pixels, and realizing the detection and positioning of the target suspended object and the lifting hook; The relative coordinates of the target with respect to the bounding box of the grid cell in which it is located, as well as the probability that the target tag and target are present in the cell, are predicted.
6. A tower crane height estimation device based on the target recognition method according to any one of claims 1-5, comprising a controller, a camera module and a ranging module, wherein, The camera module is arranged on the tower crane and used for shooting a video image of a hung object in a down-shooting mode and sending the video image to the controller; The distance measuring modules are arranged at different positions of the lifting hook and are used for obtaining the distance from the bottom end of the lifting hook to the top end of the suspended object and the height of the lifting hook from the ground; The controller is used for receiving, storing and acquiring tower foundation lifting hook size information and lifting object density information, identifying a lifting hook boundary and a lifting object boundary in the video image, calculating the pixel ratio of a lifting hook to a lifting object in the video image, calculating the lifting object area according to the lifting hook size information, the distance from the bottom end of the lifting hook to the top end of the lifting object, the lifting hook ground clearance height and the pixel ratio, calculating the lifting object height according to the lifting object weight and the lifting object density information, and calculating the lifting object bottom ground clearance height according to the lifting object height, the distance from the bottom end of the lifting hook to the top end of the lifting object and the lifting hook ground clearance height.
7. The tower crane hoist height estimation apparatus of claim 6, further comprising a display module electrically coupled to the controller for displaying video images, and hook and hoist boundaries in the video images.
8. The tower crane hoist height estimation apparatus of claim 6, further comprising a power module electrically coupled to the controller for powering the controller, the camera module, and the ranging module.
9. The tower crane elevation estimation apparatus of claim 6, wherein the ranging module employs an infrared sensor.

Description

Tower crane suspended object height estimation method and device based on target identification Technical Field The invention belongs to the technical field of intelligent tower cranes, relates to a method and a device for estimating the height of a tower crane object based on target identification, and particularly relates to a method and a device for identifying the object and estimating the height based on image feature analysis. Background The tower crane is used as common vertical transportation equipment for construction sites, has the advantages of reducing labor intensity, improving production efficiency and being wide in coverage, but the safety accident risk is always high along with the continuous expansion of the tower crane, and the core reasons are that the management and the operators are slightly conscious of safety, the adventure operation and no evidence on duty are realized. The tower crane on the current construction site is widely applied, the safety monitoring system and the anti-collision system are generally installed, more detailed supervision data and suspended object visual information can be provided for a driver by installing the safety monitoring system on the tower crane, and part of hidden dangers are reduced. In the prior art, only the height of the lifting hook (the distance between the bottom of the lifting hook and the ground) can be obtained, but the distance between the bottom of the lifting hook and the ground cannot be obtained. Therefore, the existing anti-collision function can only use the bottom of the lifting hook as a judgment boundary, the anti-collision judgment in the vertical direction can not be realized, the actual anti-collision precision of the tower crane is lower, and potential safety hazards caused by overlarge lifting objects can possibly occur. Therefore, how to provide a method and a device for estimating the height of a crane based on target identification, which can realize the judgment of the height of the bottom of the crane from the ground, is a problem that needs to be solved by those skilled in the art. Disclosure of Invention In view of the above, the invention provides a method and a device for estimating the height of a crane based on target identification, which solve the technical problems in the prior art. In order to achieve the above purpose, the present invention adopts the following technical scheme: The invention discloses a tower crane object height estimation method based on target identification, which comprises the following steps: acquiring tower foundation lifting hook size information and lifting object density information; Acquiring weight information of a suspended object, judging the distance from the bottom end of a lifting hook to the top end of the suspended object after the suspended object is lifted, and acquiring the ground clearance of the lifting hook; the method comprises the steps of acquiring a video image of a hung object shot from the top, wherein the video image comprises a complete lifting hook image and a hung object image; calculating the area of the suspended object according to the size information of the suspended object, the distance from the bottom end of the suspended object to the top end of the suspended object, the height of the suspended object and the pixel ratio, and calculating the height of the suspended object by combining the weight information of the suspended object and the density information of the suspended object; and calculating the ground clearance of the bottom end of the suspended object according to the height of the suspended object, the distance from the bottom end of the suspended object to the top end of the suspended object and the ground clearance of the suspended object. Preferably, the tower foundation lifting hook size information comprises a lifting hook length, a lifting hook width and a lifting hook height. Preferably, the step of acquiring the hanging object density information includes: pre-storing hanging object density information corresponding to different hanging object types; and automatically acquiring hanging object density information according to the current hanging object type. Preferably, the step of calculating the area of the suspended object according to the information of the size of the suspended object, the distance from the bottom end of the suspended object to the top end of the suspended object, the height of the suspended object from the ground and the pixel ratio comprises the following steps: Wherein S d is the overlooking area of the lifting hook calculated according to the size information of the lifting hook, H is the distance from the lifting arm to the ground, H j is the height of the lifting hook, H g is the height of the lifting hook from the ground, H s is the distance from the bottom end of the lifting hook to the top end of the lifting object, and S '/S d' is the pixel ratio of the lifting hook to the lifting object in the video im