Detection Method And Application Of Lifting Point Of Prefabricated Component Based On Improved Guide Filter

With the rapid development of construction machinery,the automation of construction machinery has become the current focus.Reasonable use of machine vision technology in construction machinery can not only speed up the process of mechanical automation,improve the working efficiency of construction,but also save manpower and ensure personal safety,and complete more complex mobile lifting tasks.In the current hoisting process of construction machinery,it is necessary to observe the target position by people and manually operate the hook to reach the designated position.Its operation is difficult and its precision is not high,which greatly affects the construction efficiency.In view of the above complex problems,this paper proposes a method based on the improved guided filter to detect the position of the wall panel lifting point.The method of visual recognition replaces the existing method of artificial recognition,and the control system is designed by using this method.In view of this method,the main contents of the thesis are as follows:In order to ensure the reliability and stability of the lifting point detection and recognition system,the system is designed as a whole,using industrial camera as image acquisition equipment,using Ethernet,485 bus and other communication methods to connect the hardware equipment and build a lifting point detection system.The camera imaging model and the cause of distortion are analyzed,and the camera is calibrated for the distortion of the camera.Two commonly used calibration boards are analyzed and compared.By comparing the advantages and disadvantages of different calibration methods,the camera calibration is carried out and the internal and external parameters of the camera are obtained.When taking photos at the construction site,It is need to reduce the uneven illumination and serious noise pollution under the construction site environment.The image quality is guaranteed by prepossessing the image.Compared with the traditional filtering method and the improved guided filtering method proposed in this paper,the superiority of the improved guided filtering method is proved.The filtering method proposed in this paper is used to process the obtained image,which improves the accuracy of target recognition in the later stage.Template matching is achieved by obtaining the edge contour of the lifting point CAD model and the edge contour of the actual image.The edge contour image after processing is compared with the two-dimensional template contour image obtained by virtual camera to obtain the location information of the lifting point.The experimental results show that the accuracy of the measurement results is high,the accuracy increase when the camera is close to the target the change of light intensity has little effect on the measurement accuracy.Finally,the control system is designed according to the obtained position information.Through application verification,it is proved that the system has high stability and reliability,and has engineering application value.