CN-121767353-B - Bridge construction tower deviation control method and system based on image detection

CN121767353BCN 121767353 BCN121767353 BCN 121767353BCN-121767353-B

Abstract

The invention discloses a bridge construction tower deviation control method and system based on image detection, which relates to the technical field of tower deviation control, and comprises the steps of image data acquisition, multi-source information access, image feature extraction, effectiveness screening, geometric alignment and posture conclusion generation, multi-source collaborative quality assessment and trend judgment, control instruction generation and execution, closed loop rechecking and evidence chain retention, forming unified input, extracting a trusted edge, removing occlusion and artifacts, improving alignment precision, generating a grading conclusion based on gesture parameters and attaching proportion, obtaining a final result through weight weighted synthesis, carrying out trend judgment in combination with continuous periods, ensuring stable and reliable judgment, ensuring unified generated control instruction set fields and traceability of receipt, forming a differential evaluation report and filing a versioned evidence chain after execution, and realizing closed loop control and quality traceability of the whole process.

Inventors

ZHANG JIJIN
WANG XIANGYU
XIAO YU
XIAO QIAN
ZHU WENWEI
YU JIANG
LI YONGJIANG
ZHANG BIN
LONG JINWEN

Assignees

贵州路桥集团有限公司
沿河土家族自治县交通运输局

Dates

Publication Date: 20260512
Application Date: 20260302

Claims (7)

1. The bridge construction tower deviation control method based on image detection is characterized by comprising the following steps of: Step S1, reading a construction site monitoring image, a main tower related image and additional information thereof, synchronously accessing BIM geometric baseline data, solar illumination parameters, crane operation records and weather records, and carrying out time unification and position unification on all data to output a unified data set; s2, identifying the outline of the main tower, a section boundary and an image shadow area in the image, generating a predicted shadow area according to sun illumination parameters, comparing the predicted shadow area with the image shadow area to reserve a trusted edge, generating and eliminating a shielding area according to crane operation records, and outputting a trusted edge set; S3, aligning the trusted edge set with BIM geometric baseline data and outputting matching degree data, judging and outputting attitude parameters of a roll direction, a pitch direction and a top horizontal offset direction according to the relative relation between the trusted edge set and a design vertical line, and further generating a graded attitude conclusion according to alignment continuity; Step S4, judging and giving weight to the image definition and the edge effectiveness, the illumination and shadow consistency, the shielding condition and the site environment according to the trusted edge set, the sun illumination parameters, the image shadow area, the shielding area and the weather record, generating an image conclusion, carrying out weighted synthesis on the image conclusion to generate a final posture conclusion, and judging and outputting a continuous trend or fluctuation state according to the final posture conclusion in a continuous period; s5, generating a control instruction set comprising creeping formwork oil cylinder differential adjustment, temporary inhaul cable tension adjustment and outer side pull rod tension adjustment according to posture conclusion and trend judgment, expressing the control instruction in a unified field, inputting an execution object, amplitude gears and execution sequence to a construction control end, and recording receipt information and a state mark; and S6, collecting a new period image after execution and repeating the processing process, generating posture conclusion differences and trend labels before and after execution to form a differential evaluation report, simultaneously storing an image screenshot, a trusted edge mark, a posture conclusion and a control receipt, and carrying out versioning evidence chain archiving.
2. The method for controlling the deviation of the bridge construction tower based on the image detection as set forth in claim 1, wherein the step S1 comprises the following sub-steps: Step S101, reading a construction site monitoring image and a main tower related image, wherein the construction site monitoring image and the main tower related image are added with corresponding shooting time, shooting position information, lens orientation information, lens visual angle information and image resolution data; Step S102, synchronously accessing BIM geometric baseline data, solar illumination parameters, crane operation records and weather records, and establishing corresponding relations between the BIM geometric baseline data, the solar illumination parameters, the crane operation records and weather records, shooting time and shooting position information; Step S103, performing time unification and position unification on the construction site monitoring image, the main tower related image, the BIM geometric baseline data, the sun illumination parameters, the crane operation record and the weather record, and outputting a unified data set, wherein the time unification adopts the same time standard to perform time alignment on the construction site monitoring image, the main tower related image, the BIM geometric baseline data, the sun illumination parameters, the crane operation record and the weather record, and the position unification uses the BIM geometric baseline data as a position reference to perform normalized description on the lens orientation information and the lens visual angle information.
3. The method for controlling the deviation of the bridge construction tower based on the image detection as set forth in claim 2, wherein the step S2 comprises the following sub-steps: Step S201, identifying the outline of the main tower, the section boundary line and the image shadow area in the construction site monitoring image and the related image of the main tower; preprocessing a construction site monitoring image and a main tower related image in the identification process, and extracting position information of a section boundary; Step S202, a predicted shadow area is generated according to sun illumination parameters, the image shadow area is compared, and an edge consistent with the predicted shadow area is reserved as a trusted edge; In the contrast process, matching the shadow condition corresponding to the sun illumination parameter with the image shadow region, marking as credible if the image shadow region is consistent with the predicted shadow region, otherwise, regarding as a pseudo edge and removing; Step S203, generating a shielding area according to the crane operation record, removing the shielding area, and outputting a trusted edge set; The rotation angle and the trolley displacement data in the crane operation record are mapped to the planes of the construction site monitoring image and the main tower related image to form an occlusion tower body azimuth area, and the edges in the occlusion tower body azimuth area do not participate in calculation.
4. The method for controlling the deviation of the bridge construction tower based on the image detection according to claim 3, wherein the step S5 comprises the following sub-steps: step S501, a control instruction set is generated according to the posture conclusion and the trend judgment; The control instruction set determines an action level and a priority object according to the gesture conclusion and the trend judgment as unique basis, and expresses each control instruction as an action type, an execution object, an amplitude gear, an execution sequence and a rechecking mode; step S502, the control instruction set comprises a climbing die oil cylinder differential adjustment, a temporary cable tension adjustment and an outer pull rod tension adjustment, wherein the climbing die oil cylinder differential adjustment, the temporary cable tension adjustment and the outer pull rod tension adjustment form a control instruction item in a unified field; Step S503, inputting an execution object, an amplitude gear and an execution sequence, and inputting a control instruction set into a construction control end for execution; Before input, arranging the control instruction items according to the execution sequence, sending the control instruction items to a construction control end one by one, receiving and recording receipt information of the construction control end, establishing a reference relation between each receipt and a corresponding control instruction and a posture conclusion thereof, and attaching executed, to-be-executed and rechecked state marks to the instruction control instruction items.
5. The method for controlling the deviation of the bridge construction tower based on the image detection as set forth in claim 4, wherein the step S6 comprises the following sub-steps: step S601, collecting a new period construction site monitoring image and a main tower related image after execution is completed, and repeating the processing process; In the acquisition process, corresponding an execution receipt returned by a construction control end to a timestamp of a newly acquired construction site monitoring image and a main tower related image, and simultaneously running the steps S1 to S5 again to generate data output consistent with the previous period; step S602, generating a differential evaluation report, and recording posture conclusion changes before and after execution; the prime number difference evaluation report comprises differences of posture conclusions before execution and after execution, compares the change trend of the rolling direction, the pitching direction and the top horizontal offset, and records trend labels; Step S603, saving the image screenshot, the trusted edge mark, the gesture conclusion and the control receipt, and generating a versioned evidence chain archive; The image screenshot is provided with a corresponding trusted edge label, a grade label is attached to the gesture conclusion, the control receipt comprises an action execution object, an amplitude gear and sequence information, the image screenshot, the trusted edge label, the gesture conclusion and the control receipt are packaged into a unified archive file, and the version number and the generation time are labeled.
6. The bridge construction tower deviation control method based on image detection according to claim 5, wherein the differential evaluation report comprises a posture conclusion before and after execution, a trend label, an image list of participation conclusion, a weight score and a control receipt; The image list participating in the conclusion comprises shooting time, shooting position and availability weight description of each construction site monitoring image and main tower related image, and the control receipt comprises execution objects, amplitude gears, sequences and feedback information of a construction control end of each control instruction item.
7. The bridge construction tower deviation control system based on image detection is applied to the bridge construction tower deviation control method based on image detection as claimed in any one of claims 1-6, and is characterized by comprising a data acquisition access module, a feature extraction and screening module, a para-position posture generation module, a collaborative evaluation judgment module, an instruction generation execution module and a rechecking evidence retention module; the data acquisition access module is used for reading the construction site monitoring image, the main tower related image and the additional information, synchronously accessing BIM geometric baseline data, solar illumination parameters, crane operation records and weather records, and carrying out time unification and position unification on all the data to output a unified data set; The characteristic extraction screening module is used for identifying the outer contour of the main tower, the section boundary and the image shadow area in the image, generating a predicted shadow area according to the sun illumination parameters and comparing the predicted shadow area with the image shadow area to reserve the credible edge, generating and eliminating the shielding area according to the crane operation record, and outputting a credible edge set; The alignment gesture generating module is used for aligning the trusted edge set with BIM geometric baseline data and outputting matching degree data, judging and outputting gesture parameters of a rolling direction, a pitching direction and a top horizontal offset direction according to the relative relation between the trusted edge set and a designed vertical line, and further generating a graded gesture conclusion according to alignment continuity; The collaborative evaluation judging module is used for judging and giving weight to the image definition and the edge effectiveness, the illumination and shadow consistency, the shielding condition and the site environment according to the trusted edge set, the solar illumination parameters, the image shadow area, the shielding area and the weather record, generating an image conclusion, carrying out weighted synthesis on the image conclusion to generate a final posture conclusion, and judging and outputting a continuous trend or a fluctuation state according to the final posture conclusion in a continuous period; the instruction generation execution module is used for generating a control instruction set comprising creeping formwork oil cylinder differential adjustment, temporary inhaul cable tension adjustment and outer side pull rod tension adjustment according to posture conclusion and trend judgment, wherein the control instruction is expressed in unified fields, an execution object, amplitude gears and execution sequence are input to a construction control end, and receipt information and state marks are recorded; The rechecking evidence retention module is used for collecting new periodic images after execution and repeating the processing process, generating posture conclusion differences and trend labels before and after execution to form a differential evaluation report, simultaneously storing image screenshots, trusted edge marks, posture conclusions and control receipt, and carrying out versioning evidence chain archiving.

Description

Bridge construction tower deviation control method and system based on image detection Technical Field The invention relates to the technical field of tower deviation control, in particular to a bridge construction tower deviation control method and system based on image detection. Background In the bridge construction process, the perpendicularity control of the main tower is directly related to the overall stability and construction safety of the structure. The conventional monitoring method mainly relies on a total station or a laser theodolite to measure a single point, when wind-induced swing or heat disturbance occurs in the construction process, the single point measurement result needs to be confirmed through repeated measurement for many times, so that judgment delay is caused, on-site hoisting equipment, temporary components and personnel activities can possibly form shielding on measurement points, monitoring interruption or result deviation is caused, the effectiveness of laser and optical measurement is obviously reduced under the condition of low illumination, the attitude of a tower body cannot be continuously tracked at night, the conventional method usually only depends on single equipment or single point data, the information such as weather, construction progress and equipment operation state cannot be fused, and stable monitoring and trend judgment of the whole process are difficult to realize, and therefore, how to establish a continuous, anti-interference and traceable tower deviation monitoring and controlling method through image detection on the premise of not depending on additional hardware transformation becomes a urgent problem in the bridge construction field. At present, the Chinese patent application No. CN202310567252.1 discloses an automatic deviation correcting method for a cable tower in the bridge construction process, deformation conditions of the cable tower are detected in real time, the telescopic states of the buckling ropes at the two sides of the cable tower are regulated in real time according to the deformation conditions, so that the deformation values of the cable tower are always positioned in a preset deformation range, the perpendicularity of the cable tower in the bridge construction process and the stress balance at the two sides of the cable tower are ensured, the problem of construction risk caused by deviation of the cable tower is solved, meanwhile, in the telescopic process of the buckling ropes at the two sides of the cable tower is regulated, the stress conditions of all the buckling ropes are detected and controlled in real time, so that the stress of all the buckling ropes is in a corresponding preset stress range, the problem of large deviation between actual stress and design stress of the buckling ropes is avoided, the problem of premature failure of the buckling ropes in the later bridge use process is solved, the whole regulation process is fused in the bridge construction process, and the construction period is reduced. The technology is difficult to realize continuous monitoring, dynamic evaluation and closed-loop control of the tower body gesture in the construction process, and the invention establishes a reliable edge judgment and differential evaluation mechanism by fusing multi-source data, thereby realizing real-time discovery, trend judgment and traceable correction of gesture offset. Disclosure of Invention The invention solves the technical problems that the continuous monitoring, dynamic evaluation and closed-loop control of the body gesture are difficult to realize in the construction process in the prior art, and the invention establishes a credible edge judgment and differential evaluation mechanism by fusing multi-source data, thereby realizing the real-time discovery, trend judgment and traceable correction of gesture offset. In order to solve the technical problems, the invention provides the following technical scheme: A bridge construction tower deviation control method based on image detection comprises the following steps: Step S1, reading a construction site monitoring image, a main tower related image and additional information thereof, synchronously accessing BIM geometric baseline data, solar illumination parameters, crane operation records and weather records, and carrying out time unification and position unification on all data to output a unified data set; s2, identifying the outline of the main tower, a section boundary and an image shadow area in the image, generating a predicted shadow area according to sun illumination parameters, comparing the predicted shadow area with the image shadow area to reserve a trusted edge, generating and eliminating a shielding area according to crane operation records, and outputting a trusted edge set; S3, aligning the trusted edge set with BIM geometric baseline data and outputting matching degree data, judging and outputting attitude parameters of a roll direction, a pitch d