| In recent years,with the continuous development of shield machine technology,the use of shield method for tunnel construction has gradually replaced the traditional methods and become more and more mature.Shield machine construction is widely used in tunnel construction because of its high construction efficiency and one-time tunnel forming.Shield tail clearance is an important tunneling parameter in shield construction.It refers to the distance between the inner wall of shield tail and the outer edge of segment.As the shield machine needs to adjust the direction according to the overall planning and design scheme of the tunnel during tunneling,it will lead to inconsistent shield tail clearance.When the shield tail clearance is too large,it will increase the friction force between the shield tail brush and the segment;In severe cases,external water and earth pressure will leak from the rupture of segment lining to the tunnel,resulting in serious construction accidents such as ground collapse.Based on vision technology,a shield tail clearance measurement method based on multi algorithm fusion is proposed.Combined with the actual environment,performance indicators and common interference problems,a shield tail clearance measurement system is designed.The hardware part selects the image and data acquisition equipment and control and transmission equipment.The software part takes Python language as the development language,realizes the development of processing algorithm and system software based on Open CV image processing library and pyqt5,and solves the functional relationship between the measured distance and the physical distance corresponding to each unit pixel through pre calibration,which is used to solve the actual shield tail gap value.By projecting the laser line to the shield tail gap,the industrial camera and laser ranging module are used to collect the characteristic information of the laser line and measure the distance respectively,the filtering algorithm in image processing is used for denoising,and the feature extraction of the laser centerline is realized through the extraction algorithm;In view of the existing interference,the connected area marking method based on area threshold is used to propose the interference pixel points,and the laser centerline coordinates are optimized based on RANSAC algorithm,and the pixel distance of shield tail gap is calculated.On this basis,the solution of shield tail gap value is realized by combining the measured distance and pre calibration function.After the hardware and software parts of the system are assembled and debugged,repeated experiments are carried out under different measured distances and shield tail clearances.The experimental results show that the maximum measurement error is 1.852 mm,which meets the setting requirements.Finally,the error is analyzed from two aspects of hardware and software,which provides suggestions for subsequent system optimization. |
