Building settlement monitoring method and device, electronic equipment and storage medium

Abstract

The invention provides a building settlement monitoring method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring monitoring images of a building at at least two moments; performing corner detection on the monitoring images at the at least two moments to obtain corner positions of the building in the monitoring images at the at least two moments; and determining the settlement rate of the building based on the angular point positions at the at least two moments. According to the method, the device, the electronic equipment and the storage medium provided by the invention, the angular point position of the building in the monitoring image can be positioned through angular point detection, so that settlement monitoring for the building is realized based on the angular point positions at at least two moments. In the process, observation points do not need to be specially arranged for the building, the appearance of the building is not affected, settlement monitoring can be achieved only by using monitoring equipment around the building to shoot monitoring images containing the building, extra hardware is not added, and precise settlement monitoring which is low in cost and capable of achieving continuous observation is achieved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a method, device, electronic device and storage medium for building settlement monitoring. Background Art

[0002] In order to ensure the normal use and safety of buildings, and to provide reliable data and corresponding settlement parameters for future exploration, design and construction, it is necessary to monitor the settlement of buildings during the construction process and after being put into use.

[0003] Currently, the methods for building settlement monitoring mainly include manually collecting settlement data, or obtaining the building height in real time through the satellite positioning system to achieve settlement monitoring, or setting up laser emitters and targets inside the building for settlement monitoring.

[0004] In the above solutions, manually collecting settlement data requires a large amount of manpower, while using the satellite positioning system or laser emitters for settlement monitoring will incur a large amount of overhead costs. The high cost is not conducive to the realization of long-term monitoring. Summary of the Invention

[0005] The present invention provides a method, device, electronic device and storage medium for building settlement monitoring, so as to solve the defect of high cost in building settlement monitoring in the prior art.

[0006] The present invention provides a method for building settlement monitoring, including:

[0007] Obtaining monitoring images of the building at at least two moments;

[0008] Performing corner detection on the monitoring images at the at least two moments to obtain the corner positions of the building in the monitoring images at the at least two moments;

[0009] Based on the corner positions at the at least two moments, determining the settlement rate of the building.

[0010] According to the method for building settlement monitoring provided by the present invention, the determining the settlement rate of the building based on the corner positions at the at least two moments includes:

[0011] Based on the conversion relationship between the pixel coordinate system and the world coordinate system, and the corner positions in the pixel coordinate system at the at least two moments, determining the settlement rate of the building in the world coordinate system;

[0012] The conversion relationship is determined based on the shooting parameters of the monitoring device that shoots the monitoring images, or based on the pixel settlement information of the building in the pixel coordinate system and the true settlement information of the building in the world coordinate system.

[0013] A building settlement monitoring method provided by the present invention, the determining step of the conversion relationship includes:

[0014] Perform corner detection on the reference images of the building at at least two reference times to obtain the reference corner positions of the building in the reference images at the at least two reference times;

[0015] Based on the reference corner positions at the at least two reference times, determine the pixel settlement information in the pixel coordinate system;

[0016] Obtain the true settlement information in the world coordinate system detected at the at least two reference times;

[0017] Determine the conversion relationship based on the pixel settlement information and the true settlement information.

[0018] A building settlement monitoring method provided by the present invention, the determining the conversion relationship based on the pixel settlement information and the true settlement information includes:

[0019] Calculate a first error between the pixel settlement information and the true settlement information;

[0020] When the first error is less than a first threshold, determine the conversion relationship based on the pixel settlement information and the true settlement information.

[0021] A building settlement monitoring method provided by the present invention, the calculating the first error between the pixel settlement information and the true settlement information includes:

[0022] Determine the pixel settlement rates and the true settlement rates at two reference times from the pixel settlement information and the true settlement information;

[0023] Based on the ratio between the pixel settlement rate and the true settlement rate at two reference times, determine the first error.

[0024] A building settlement monitoring method provided by the present invention, the determining the conversion relationship based on the pixel settlement information and the true settlement information further includes:

[0025] When the first error is greater than or equal to the first threshold, adjust at least one of the shooting parameters of the monitoring device for shooting the reference image, the algorithm parameters for performing corner detection on the reference image, and the interval duration of the reference times until the newly determined first error is less than the first threshold;

[0026] When the first error is less than the first threshold, determine the shooting parameters of the monitoring device for shooting the monitoring image based on the shooting parameters of the monitoring device for shooting the reference image, determine the time for shooting the monitoring image based on the interval duration of the reference time, and determine the algorithm parameters for corner detection of the monitoring image based on the algorithm parameters for corner detection of the reference image.

[0027] According to a building settlement monitoring method provided by the present invention, before obtaining the monitoring images of the building at at least two times, it further includes:

[0028] Perform corner detection on the reference images of the building at at least two reference times to obtain the reference corner positions of the building in the reference images at the at least two reference times;

[0029] Based on the reference corner positions at the at least two reference times and the shooting parameters of the monitoring device for shooting the reference image, determine the monitoring settlement information in the world coordinate system;

[0030] Obtain the true settlement information in the world coordinate system detected at the at least two reference times;

[0031] Calculate the second error between the monitoring settlement information and the true settlement information. When the second error is less than the second threshold, determine the shooting parameters of the monitoring device for shooting the monitoring image based on the shooting parameters of the monitoring device for shooting the reference image, determine the time for shooting the monitoring image based on the interval duration of the reference time, and determine the algorithm parameters for corner detection of the monitoring image based on the algorithm parameters for corner detection of the reference image.

[0032] The present invention also provides a building settlement monitoring device, including:

[0033] An image acquisition unit for acquiring monitoring images of a building at at least two times;

[0034] A corner detection unit for performing corner detection on the monitoring images at the at least two times to obtain the corner positions of the building in the monitoring images at the at least two times;

[0035] A settlement monitoring unit for determining the settlement rate of the building based on the corner positions at the at least two times.

[0036] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the building settlement monitoring method as described in any one of the above.

[0037] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the building settlement monitoring method described in any one of the above is implemented.

[0038] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, the building settlement monitoring method described in any one of the above is implemented.

[0039] The building settlement monitoring method, device, electronic device and storage medium provided by the present invention can detect and locate the corner positions of the building in the monitoring image through corner detection, so as to realize the settlement monitoring of the building based on the corner positions at at least two moments. In this process, there is no need to specifically set observation points for the building, nor will it affect the appearance of the building. Only by using the monitoring devices around the building to capture the monitoring images containing the building can the settlement monitoring be realized. Without adding additional hardware, accurate settlement monitoring with low cost and sustainable observation is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0041] Figure 1 is a schematic flow chart of the building settlement monitoring method provided by the present invention;

[0042] Figure 2 is a schematic flow chart of the determination step of the conversion relationship provided by the present invention;

[0043] Figure 3 is a schematic flow chart of the parameter verification method provided by the present invention;

[0044] Figure 4 is a schematic flow chart of the method for obtaining the monitoring parameters of the building settlement monitoring provided by the present invention;

[0045] Figure 5 is a schematic structural diagram of the building settlement monitoring device provided by the present invention;

[0046] Figure 6 is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] Currently, the methods for realizing building settlement monitoring mainly include the following three types:

[0049] One is to paint the building surface or add special observation points, and use monitoring equipment to aim at the observation points to continuously observe the building settlement rate. This method requires adding additional observation points on the building surface, which will affect the appearance of the building. And if high-precision settlement monitoring is desired, laser equipment usually needs to be equipped, which will bring additional high costs.

[0050] Another is to use the satellite positioning system and dynamic coordinate equipment to obtain the longitude, latitude coordinates and altitude coordinates of the building in real time, so as to judge the building settlement rate. This method also requires adding observation points on the outer surface of the building, and the cost of calling the satellite system is extremely high, and it is difficult to popularize.

[0051] There is also one that installs laser emitters and targets at different positions inside the building, and judges whether there is uneven settlement of the building by judging the change of the laser punctuation position. This method requires adding professional observation equipment inside the building, which will also result in high costs. And the high cost is not conducive to the realization of long-term monitoring.

[0052] In view of the above problems, an embodiment of the present invention provides a building settlement monitoring method. Figure 1 It is a schematic flowchart of the building settlement monitoring method provided by the present invention. As Figure 1 shown, the method includes:

[0053] Step 110, obtaining monitoring images of the building at at least two moments.

[0054] The building here is the building that needs to be monitored for settlement. The monitoring images including the building can be collected by selecting monitoring equipment that can capture the building around the outside of the building. The monitoring equipment here can be cameras installed on the roadside, or cameras for security monitoring in the park, etc. The embodiments of the present invention do not make specific limitations on this.

[0055] Moreover, in order to realize the monitoring of the building settlement, periodic image collection needs to be carried out when collecting the monitoring images. For example, image collection can be performed once every preset time interval, such as collecting monitoring images once every day or two.

[0056] It is understandable that the monitoring images obtained at at least two moments are images of the same building taken at different moments under the same viewing angle with the same monitoring device parameters. The images of the same building at different moments can prominently reflect the settlement situation of the building over time.

[0057] Step 120: Perform corner detection on the monitoring images at the at least two moments to obtain the corner positions of the building in the monitoring images at the at least two moments.

[0058] Specifically, after obtaining the monitoring images at at least two moments, corner detection can be performed separately for each monitoring image. The corner detection here is used to detect the positions with obvious edge changes in the monitoring images. It is understandable that most buildings themselves have obvious edges and corners. For example, the edge endpoints of a square building and the top of a dome-shaped building can all be regarded as the corners in the monitoring images. Through corner detection, the corner positions in the monitoring images can be located.

[0059] Furthermore, in the embodiments of the present invention, corner detection can be implemented by the Harris corner detection algorithm, or can also be implemented by algorithms such as Shi-Tomasi corner detection, oravec corner detection, Nobel corner detection, etc. The embodiments of the present invention do not make specific limitations in this regard. Among them, the Harris corner detection algorithm, as a method widely used in computer vision systems to obtain image features, can calculate the gray difference between adjacent pixel points and compare with the threshold parameter X to obtain the feature type of the pixel points, so as to distinguish corners, edges, and flat areas. The threshold parameter X referred to here generally takes values of 0.04 - 0.06, and the larger its value, the fewer the corners. The number of detected corners in the monitoring image can be adjusted by adjusting the size of the threshold parameter X.

[0060] Considering that the monitoring images at at least two moments contain the same building under the same viewing angle, the corner positions of the building in different monitoring images obtained by separately performing corner detection on the monitoring images are the positions of the same corners at different moments in different images. That is, the corners of the building in the monitoring images obtained through corner detection are natural observation points that do not require artificial setting. By comparing the corner positions in different monitoring images, the settlement situation of the building can be determined.

[0061] Step 130: Determine the settlement rate of the building based on the corner positions at the at least two moments.

[0062] Specifically, after obtaining the corner positions at at least two moments, the settlement change situation of the building over time can be calculated based on the corner positions at at least two moments, so as to obtain the settlement rate of the building.

[0063] Furthermore, there are various ways to calculate the settlement rate. For example, the pixel difference in the vertical direction of the corner positions in the monitoring images at every two adjacent moments can be calculated, and then based on the shooting parameters of the monitoring device that captures the monitoring images, the pixel difference is converted into the height in the real world, and the height is divided by the time difference between two adjacent moments as the settlement rate. Another example is to calculate the pixel difference in the vertical direction of the corner positions in the monitoring images at every two adjacent moments, and then based on the pre-stored conversion relationship between the pixel distance and the distance in the real world, the pixel difference is converted into the height in the real world, and the height is divided by the time difference between two adjacent moments as the settlement rate. The embodiments of the present invention do not make specific limitations on this.

[0064] The method provided by the embodiments of the present invention can detect the corner positions of the building in the monitoring images through corner detection, so as to realize the settlement monitoring of the building based on the corner positions at at least two moments. In this process, there is no need to specifically set observation points for the building, nor will it affect the appearance of the building. Only by using the monitoring devices around the building to capture the monitoring images containing the building can the settlement monitoring be realized. Without adding additional hardware, accurate settlement monitoring with low cost and sustainable observation is achieved.

[0065] Based on the above embodiments, step 130 includes:

[0066] Based on the conversion relationship between the pixel coordinate system and the world coordinate system, and the corner positions in the pixel coordinate system at the at least two moments, determine the settlement rate of the building in the world coordinate system;

[0067] The conversion relationship is determined based on the shooting parameters of the monitoring device that captures the monitoring images, or based on the pixel settlement information of the building in the pixel coordinate system and the real settlement information of the building in the world coordinate system.

[0068] Specifically, the corner positions obtained through corner detection belong to the position information in the coordinate system constructed by the monitoring images. Here, the coordinate system constructed by the monitoring images is denoted as the pixel coordinate system, and the corner positions are the corner positions in the pixel coordinate system.

[0069] For the settlement monitoring of the building, what is concerned is the settlement situation of the building in the real world. Here, the coordinate system constructed based on the real world is denoted as the world coordinate system.

[0070] After obtaining the corner positions, the comparison of the corner positions at different times is carried out in the pixel coordinate system and cannot reflect the settlement situation in the real world. Therefore, based on the conversion relationship between the pixel coordinate system and the world coordinate system, the settlement situation reflected by the corner positions in the pixel coordinate system needs to be converted to the world coordinate system, so as to obtain the settlement rate in the world coordinate system.

[0071] The coordinate conversion here can be to first convert the corner positions to the world coordinate system based on the conversion relationship, and then calculate the settlement rate based on the corner positions in the world coordinate system; it can also be to complete the settlement rate calculation in the pixel coordinate system first, and then convert the calculated settlement rate to the world coordinate system. The embodiments of the present invention do not make specific limitations on this.

[0072] Here, the conversion relationship can be preset. The determination of the conversion relationship can be realized depending on the shooting parameters of the monitoring device that shoots the monitoring image. The shooting parameters here can include the distance between the monitoring device and the building, and can also include the device height and lens angle of the monitoring device when shooting the building. It can also include the resolution of the monitoring image, the focal length when shooting, etc. The embodiments of the present invention do not make specific limitations on this. For example, the scaling ratio between the image and the physical object can be determined based on the focal length, and this scaling ratio can be used as the conversion relationship.

[0073] Alternatively, the determination of the conversion relationship can also be obtained by fitting the conversion relationship based on the pixel settlement information of the building in the pixel coordinate system obtained by image shooting historically and the real settlement information measured in the world coordinate system. It can be understood that the pixel settlement information and the real settlement information here refer to the settlement situation of the same building in the same time period. The difference between the two is that the pixel settlement information is in the pixel coordinate system and is obtained by analyzing the captured image, while the real settlement information is in the world coordinate system and is obtained by on-site measurement.

[0074] Furthermore, the pixel settlement information and the real settlement information are of the same data type. For example, both the pixel settlement information and the real settlement information are settlement displacements, or both the pixel settlement information and the real settlement information are settlement rates.

[0075] The method provided by the embodiments of the present invention realizes the rapid calculation of the settlement rate in the world coordinate system through the pre-calculated conversion relationship between the pixel coordinate system and the world coordinate system, can improve the settlement monitoring efficiency, and reduce the calculation amount required for settlement monitoring.

[0076] Based on any of the above embodiments, Figure 2 is a schematic flowchart of the step for determining the conversion relationship provided by the present invention. As Figure 2 shown, the step for determining the conversion relationship may include:

[0077] Step 210: Detect corner points of the reference images of the building at at least two reference times to obtain the reference corner point positions of the building in the reference images at the at least two reference times.

[0078] Specifically, for a building, images can be taken at at least two reference times. Here, the images containing the building obtained by taking pictures are denoted as reference images. It can be understood that the shooting parameters of the monitoring device for taking pictures of the building here are the same as those of the monitoring device for taking pictures of the building in step 110. The shooting time is denoted as the reference time, and the images obtained by taking pictures are denoted as reference images in order to distinguish them from the monitoring images in step 110.

[0079] For the reference images at at least two reference times, corner point detection is performed respectively to obtain the corner point positions of the building at each reference time, which are denoted as reference corner point positions here.

[0080] Step 220: Determine the pixel settlement information in the pixel coordinate system based on the reference corner point positions at the at least two reference times.

[0081] Specifically, after obtaining the reference corner point positions at each reference time, the pixel settlement information can be determined based on the reference corner point positions at every two adjacent reference times. The pixel settlement information here can be the pixel displacement of the reference corner point positions in the vertical direction within a period of time, or the rate of the reference corner point in the vertical direction. The embodiments of the present invention do not make specific limitations on this.

[0082] Step 230: Obtain the true settlement information in the world coordinate system detected at the at least two reference times.

[0083] Specifically, at the same reference time, the actual settlement situation of the building can be manually collected, that is, the true settlement information in the world coordinate system is obtained. It should be noted that the true settlement information here and the pixel settlement information in step 220 refer to the settlement situation of the same building within the same period of time.

[0084] Step 240: Determine the conversion relationship based on the pixel settlement information and the true settlement information.

[0085] Specifically, after obtaining the pixel settlement information and the true settlement information of the same building within the same period of time, the conversion relationship required to convert from the pixel settlement information to the true settlement information, that is, the conversion relationship from the pixel coordinate system to the world coordinate system, can be fitted.

[0086] The method provided by the embodiment of the present invention determines the conversion relationship by combining the pixel settlement information and the true settlement information at the same reference time, so that when subsequent building settlement monitoring is carried out, it can be directly mapped from the pixel coordinate system to the world coordinate system, which helps to improve the calculation efficiency of building settlement monitoring.

[0087] Based on any of the above embodiments, step 240 includes:

[0088] Calculate a first error between the pixel settlement information and the true settlement information;

[0089] When the first error is less than a first threshold, determine the conversion relationship based on the pixel settlement information and the true settlement information.

[0090] Specifically, before determining the conversion relationship based on the pixel settlement information and the true settlement information, the error in the settlement situation between the pixel settlement information and the true settlement information can be calculated first.

[0091] Considering that the true settlement information is obtained by manual collection and has high reliability, by comparing the error between the pixel settlement information and the true settlement information, the credibility of the pixel settlement information itself can be measured. Considering that the pixel settlement information and the true settlement information are values in different coordinate systems, when comparing, the pixel settlement information can be converted to the world coordinate system and then compared with the true settlement information; or the true settlement information can be converted to the pixel coordinate system and then compared with the pixel settlement information. The embodiment of the present invention does not make specific limitations on this. Here, the error between the pixel settlement information and the true settlement information is denoted as the first error. If the first error between the two is less than the first threshold, it is considered that the information reflected by the pixel settlement information and the true settlement information in the settlement situation is close, and the pixel settlement information itself is credible, and the conversion relationship can be determined by applying the pixel settlement information and the true settlement information; otherwise, it is considered that the information reflected by the pixel settlement information and the true settlement information in the settlement situation is inconsistent, and the pixel settlement information itself is not credible, and the conversion relationship is not determined based on the pixel settlement information and the true settlement information.

[0092] The method provided by the embodiment of the present invention determines the conversion relationship when the first error between the pixel settlement information and the true settlement information is less than the first threshold, effectively ensuring the reliability of the conversion relationship, and further ensuring the reliability of the implementation of the settlement monitoring method.

[0093] Based on any of the above embodiments, in step 240, the calculation of the first error between the pixel settlement information and the true settlement information includes:

[0094] Determine the pixel settlement rate and the true settlement rate at two reference times from the pixel settlement information and the true settlement information;

[0095] Determine the first error based on the ratio between the pixel settlement rate and the true settlement rate at two reference times.

[0096] Specifically, the pixel settlement information and the true settlement information may include the pixel settlement rate and the true settlement rate at at least two reference times. The pixel settlement rate and the true settlement rate at two reference times can be taken therefrom.

[0097] That is, the pixel settlement rate and the true settlement rate at two reference times are obtained here, where each reference time corresponds to a pixel settlement rate and a true settlement rate. For a reference time, the ratio of the pixel settlement rate corresponding to the reference time to the true settlement rate can be calculated, and the ratio here can reflect the difference between the pixel settlement rate and the true settlement rate at the same time.

[0098] Thus, the ratios between the pixel settlement rate and the true settlement rate at two reference times can be compared, that is, the change magnitude of the difference between the pixel settlement rate and the true settlement rate at different times is analyzed. If the change in the difference between the pixel settlement rate and the true settlement rate at different times is small, it is considered that the difference between the pixel settlement rate and the true settlement rate is stable, and the first error between the two is small; otherwise, it is considered that the difference between the pixel settlement rate and the true settlement rate is unstable, and the first error between the two is large.

[0099] Further, the first error y 1 can be calculated based on the following formula:

[0100]

[0101] In the formula, V 1 , V 2 are the true settlement rates at two reference times, is the pixel settlement rate at two reference times.

[0102] For example, the true settlement rate V 1 of a building is 0.01 mm / day, V 2 is 0.01 mm / day, the first threshold Y is 100%, if the pixel settlement rate is 2 pt / day, is 5 pt / day, calculate y 1 *100% = 150%, that is, y 1 > Y, not meeting the error range. If the pixel settlement rate is 2 pt / day, is 3.5 pt / day, calculate y 1 *100% = 75%, that is, y1 <Y satisfies the error range.

[0103] Based on any of the above embodiments, in step 240, determining the conversion relationship based on the pixel settlement information and the true settlement information further includes:

[0104] When the first error is greater than or equal to the first threshold, adjust at least one of the shooting parameters of the monitoring device for shooting the reference image, the algorithm parameters for corner detection of the reference image, and the interval duration of the reference time until the newly determined first error is less than the first threshold.

[0105] Specifically, for the case where the first error is greater than or equal to the first threshold, that is, the case where the pixel settlement information itself is considered untrustworthy, it is necessary to adjust the shooting parameters of the monitoring device for shooting the reference image and / or the algorithm parameters for corner detection of the reference image.

[0106] It can be understood that the shooting parameters of the monitoring device may include at least one of the distance between the device and the building, the device height, the lens angle, the resolution of the monitored image, and the focal length during shooting.

[0107] Among the above parameters, the distance will affect the size of the building in the image. To avoid the building being too small to distinguish the building settlement in the image, for example, the distance can be adjusted within 10m;

[0108] The resolution and focal length will affect the clarity of the building in the captured image. The resolution can be adjusted to the highest resolution within the device's capabilities.

[0109] The algorithm parameters for corner detection will affect the number of angles detected in the image. Taking the Harris corner detection algorithm as an example, the initial value of the threshold parameter X in the Harris corner detection algorithm can be set as large as possible and gradually decreased in steps of 0.001 during the adjustment process to meet the requirement of at least identifying 10 corners.

[0110] The interval duration of the reference time will affect the size of the building settlement situation in the image. It can be understood that the larger the interval duration, the more obvious the settlement situation that can be reflected. The initial value of the interval duration can be set as small as possible and gradually increased in steps of 1 day during the adjustment process to satisfy that the displacement of the same corner point in two adjacent images is greater than one pixel point.

[0111] By adjusting at least one of the above various types of data and applying the adjusted various types of data to re-collect the reference corner point positions of the reference image, the pixel settlement information is updated accordingly. Then, based on the new pixel settlement information and the true settlement information, the first error is calculated. When the first error is greater than or equal to the first threshold, the adjustment of at least one of the above various types of data is returned. When the first error is less than the first threshold, a conversion relationship is fitted based on the new pixel settlement information and the true settlement information, and the shooting parameters of the latest monitoring device, the algorithm parameters for corner detection of the reference image, and the interval duration of the reference time can be used as the parameters for subsequent shooting of the monitoring images.

[0112] The method provided by the embodiments of the present invention reduces the error of settlement monitoring based on images by adjusting at least one of the shooting parameters of the monitoring device for shooting the reference image, the algorithm parameters for corner detection of the reference image, and the interval duration of the reference time, thereby improving the reliability of subsequent settlement monitoring.

[0113] Based on any of the above embodiments, step 240 further includes:

[0114] When the first error is less than the first threshold, the shooting parameters of the monitoring device for shooting the reference image, the algorithm parameters for corner detection of the reference image, and the interval duration of the reference time are determined as the parameters for shooting the monitoring image and for corner detection of the monitoring image.

[0115] The method provided by the embodiments of the present invention applies the error-verified data to subsequent building settlement monitoring, which can ensure the reliability of subsequent settlement monitoring.

[0116] Based on any of the above embodiments, Figure 3 is a schematic flowchart of the parameter verification method provided by the present invention. As Figure 3 shown, before performing step 110, the method further includes:

[0117] Step 310, perform corner detection on the reference images of the building at at least two reference times to obtain the reference corner point positions of the building in the reference images at the at least two reference times.

[0118] Specifically, for a building, images can be taken at at least two reference times. Here, the taken images containing the building are denoted as reference images. It can be understood that the shooting parameters of the monitoring device for shooting the building here are the same as those of the monitoring device for shooting the building in step 110. Denoting the shooting time as the reference time and the taken image as the reference image is to distinguish it from the monitoring image in step 110.

[0119] For the reference images at at least two reference times, corner detection is performed separately to obtain the corner positions of the building at each reference time, which are denoted as reference corner positions here.

[0120] Step 320: Based on the reference corner positions at the at least two reference times and the shooting parameters of the monitoring device that shoots the reference images, determine the monitored settlement information in the world coordinate system.

[0121] Specifically, after obtaining the reference corner positions at each reference time, based on the reference corner positions at every two adjacent reference times and the conversion relationship from the pixel coordinate system where the reference corner positions are located to the world coordinate system of the real world determined based on the shooting parameters, determine the monitored settlement information in the world coordinate system obtained through the reference images.

[0122] Here, specifically, the reference corner positions can be first converted to the world coordinate system based on the conversion relationship, and then the settlement rate can be calculated based on the corner positions in the world coordinate system as the monitored settlement information; it can also be that the settlement rate is calculated in the pixel coordinate system first, and then the calculated settlement rate is converted to the world coordinate system as the monitored settlement information. The embodiments of the present invention do not make specific limitations on this.

[0123] Step 330: Obtain the real settlement information in the world coordinate system detected at the at least two reference times.

[0124] Specifically, at the same reference time, the actual settlement situation of the building can be manually collected, that is, the real settlement information in the world coordinate system is obtained. It should be noted that the real settlement information here and the monitored settlement information in step 320 refer to the settlement situation of the same building in the same time period, and both the real settlement information and the monitored settlement information are in the world coordinate system.

[0125] Step 340: Calculate the second error between the monitored settlement information and the real settlement information. When the second error is less than the second threshold, determine the shooting parameters of the monitoring device that shoots the monitoring images based on the shooting parameters of the monitoring device that shoots the reference images, determine the time to shoot the monitoring images based on the time interval between the reference times, and determine the algorithm parameters for corner detection of the monitoring images based on the algorithm parameters for corner detection of the reference images.

[0126] Specifically, after obtaining the real settlement information and the monitored settlement information, the two can be compared to determine the error between the real settlement information and the monitored settlement information, which is denoted as the second error here. The second error can reflect the credibility of the monitored settlement information itself.

[0127] If the second error between the two is less than the second threshold, it is considered that the information reflected by the monitored settlement information and the true settlement information in terms of the settlement situation is close, and the monitored settlement information itself is credible. The various parameters used to obtain the monitored settlement information, including the shooting parameters of the monitoring device for shooting the reference image, the interval duration of the reference time, and the algorithm parameters for corner detection of the reference image, can be continued to be used in subsequent building settlement monitoring;

[0128] Otherwise, it is considered that the information reflected by the monitored settlement information and the true settlement information in terms of the settlement situation is inconsistent, and the monitored settlement information itself is not credible. It is necessary to adjust the various parameters used to obtain the monitored settlement information, and re-collect the reference corner positions of the reference image by applying the adjusted various parameters, thereby updating the monitored settlement information, and calculating the second error based on the new monitored settlement information and the true settlement information. In the case where the second error is greater than or equal to the second threshold, return the adjustment of at least one of the above various parameters. In the case where the second error is less than the second threshold, the new various parameters are continued to be used in subsequent building settlement monitoring.

[0129] Based on any of the above embodiments, Figure 4 is a schematic flowchart of a method for obtaining monitoring parameters for building settlement monitoring provided by the present invention. As Figure 4 shown, before performing the above building settlement monitoring method, it is necessary to first determine the monitoring parameters for performing building settlement monitoring. The monitoring parameters here can include the shooting parameters of the monitoring device for shooting the reference image, the interval duration of the reference time, and the algorithm parameters for corner detection of the reference image.

[0130] The acquisition of the monitoring parameters can be achieved based on the following steps:

[0131] Step 410, preliminarily determine the monitoring parameters:

[0132] In this step, the shooting parameters of the monitoring device for shooting the reference image, the interval duration of the reference time, and the algorithm parameters for corner detection of the reference image can be preliminarily determined.

[0133] Among them, for the shooting parameters, surveillance devices that can be used near the building and can capture the building can be selected, and the number of surveillance devices is not limited; and the shooting position A where the surveillance device can capture the building is selected, including the device height, lens angle, etc. The number of shooting positions here is not limited, and policy-based changes are supported. Multiple fixed options or fixed variables can be set according to the actual situation; in addition, the specific parameter a that the surveillance device can capture the building also needs to be selected, which can include resolution, focal length, etc. The number of selectable parameters is not limited, and policy-based changes are supported. Multiple fixed options or fixed variables can be set according to the actual situation. Also, the number of images N specified for shooting by the surveillance device to reduce errors can be selected, with a default of 5, which can be changed according to the actual situation and continuous learning.

[0134] For the interval duration T, it can be default set to 1 day, and it is required that the actual displacement corresponding to the same corner point of adjacent two images must be at most 0.01 mm. The selection of the interval duration supports policy-based changes, and multiple fixed options or fixed variables can be set according to the actual situation.

[0135] For the algorithm parameters for corner detection, the threshold parameter X of the Harris corner detection algorithm can be selected. For example, it can be default set to 0.04 and can be gradually increased to 0.06 according to the policy.

[0136] In addition, in order to implement subsequent error calculation, a threshold Y related to the error can be preset. For example, the threshold range can be default set to 100%, and the threshold range can be changed according to the actual situation and continuous learning;

[0137] Also, the threshold parameter Z for setting the safety requirement of the building settlement rate can be selected, with a default of 0.02 mm / day, which can be changed according to the actual situation and continuous learning. Here, setting the threshold parameter Z is to be able to immediately trigger a settlement anomaly alarm when a settlement rate greater than Z is monitored.

[0138] Step 420, obtain a reference image based on the monitoring parameters and perform corner detection on the reference image:

[0139] Here, based on the shooting parameters and interval duration in the monitoring parameters preliminarily determined in the previous step, a reference image containing the building can be collected. Specifically, the surveillance device can be controlled to capture the external images of the building at the capture position A using the parameter a at the interval duration T for the specified number N, that is, the reference images P1 - PN.

[0140] Also, the surveillance device or the affiliated cloud server can be controlled to use the Harris corner detection algorithm with the threshold parameter X to analyze the reference images P1 - PN to obtain the corner positions of the building in the pixel coordinate system in adjacent two reference images.

[0141] Meanwhile, the true settlement information collected manually can be obtained, and the true settlement information here can specifically be the true settlement rate V.

[0142] Step 430, determine whether direct calculation of the monitored settlement information is supported:

[0143] Here, whether direct calculation of the monitored settlement information is supported depends on the processing capacity and computing resources of the monitoring device itself. If the monitoring device has strong processing capacity and computing resources, the monitored settlement information can be directly calculated, and step 441 is executed; otherwise, it is determined that the monitoring device does not directly calculate the monitored settlement information, and step 442 is executed.

[0144] Step 441, obtain the monitored settlement information of the building:

[0145] Specifically, based on the shooting parameters in the monitoring parameters, through the geometric and equal-picture calculation formula, the monitored settlement rate v in the world coordinate system can be directly calculated on the basis of the corner point position as the monitored settlement information, and then step 451 is executed.

[0146] Step 451, determine whether the second error is less than the second threshold:

[0147] The second error here is the error obtained based on the monitored settlement rate v and the true settlement rate V, and can specifically be expressed as judging where v 1 and V 1 respectively represent the monitored settlement rate and the true settlement rate at the same moment. is the second error, and Y 2 is the second threshold.

[0148] Here, if Step 480 is executed; otherwise, step 460 is executed. And if the specified number N is greater than 2, that is, there are multiple groups of monitored settlement rates and true settlement rates, it is necessary that the second error of each group is less than the second threshold to execute step 480.

[0149] Step 442, obtain the pixel settlement information of the building:

[0150] Limited by the computing power of the monitoring device, the settlement information can be calculated only in the pixel coordinate system, that is, the displacement speed in the vertical direction of the same corner point in the pixel coordinate system in adjacent images is calculated, that is, the pixel settlement rate is obtained as the pixel settlement information, and then step 452 is executed.

[0151] Step 452, determine whether the first error is less than the first threshold:

[0152] The first error here is based on the pixel settlement rate The error obtained from the true settlement rate V can be specifically expressed as a judgment where V 1 and V 2 are the true settlement rates at two reference times, is the pixel settlement rate at two reference times, is the first error, and Y 1 is the first threshold.

[0153] Here, if Step 480 is executed; otherwise, Step 460 is executed. Moreover, if the specified number N is greater than 2, that is, there are multiple sets of pixel settlement rates and true settlement rates, it is necessary that the first error of each group is less than the first threshold to execute Step 480.

[0154] Step 460, adjust the monitoring parameters:

[0155] At this time, the monitoring parameters are regarded as errors to be optimized, and the monitoring parameters need to be dynamically adjusted, and then Step 470 is executed.

[0156] For example, if the true settlement rate V of a building is 0.01 mm / day and the interval duration T is 1 day, that is, it is required that the Harris corner detection algorithm can identify the actual displacement corresponding to at most 0.01 mm between adjacent two pixel points in adjacent two images. If the current monitoring equipment and observation parameters can only identify the actual displacement corresponding to 0.1 mm between adjacent two pixel points in the same picture, the interval duration T can be extended to 10 days, so that the actual displacement rate corresponding to adjacent two pixel points in adjacent two pictures can be 0.01 mm / day.

[0157] Step 470, re-verify based on the adjusted monitoring parameters until the error is less than the threshold:

[0158] Here, based on the adjusted monitoring parameters, Step 420 can be executed to update the error and re-judge whether the updated error is less than the threshold. If it is less, Step 480 is executed.

[0159] Step 480, confirm that the current monitoring parameters are correct and available.

[0160] Step 490, solidify the monitoring parameters for subsequent continuous building settlement monitoring.

[0161] It should be noted that in the above method, in order to reduce the manual workload in the optimization and adjustment link of Step 460, in the link of initially determining the monitoring parameters in Step 410, multiple sets of monitoring parameters can be set to avoid the situation of multiple manual reworks.

[0162] In addition, after solidifying the monitoring parameters, if there is a situation of manually collecting the true settlement rate or other customized requirements, the verification and optimization of the monitoring parameters can be re-triggered.

[0163] The building settlement monitoring device provided by the present invention will be described below. The building settlement monitoring device described below can be correspondingly referred to the building settlement monitoring method described above.

[0164] Figure 5 is a schematic structural diagram of the building settlement monitoring device provided by the present invention, as Figure 5 shown, the device includes:

[0165] An image acquisition unit 510, configured to acquire monitoring images of the building at at least two moments;

[0166] A corner point detection unit 520, configured to perform corner point detection on the monitoring images at the at least two moments to obtain the corner point positions of the building in the monitoring images at the at least two moments;

[0167] A settlement monitoring unit 530, configured to determine the settlement rate of the building based on the corner point positions at the at least two moments.

[0168] The device provided by the embodiment of the present invention can locate the corner point positions of the building in the monitoring images through corner point detection, so as to realize the settlement monitoring of the building based on the corner point positions at at least two moments. In this process, there is no need to specifically set observation points for the building, nor will it affect the appearance of the building. Only by using the monitoring devices around the building to capture the monitoring images containing the building can the settlement monitoring be realized. Without adding additional hardware, accurate settlement monitoring with low cost and sustainable observation is achieved.

[0169] Based on any of the above embodiments, the settlement monitoring unit is configured to:

[0170] Determine the settlement rate of the building in the world coordinate system based on the conversion relationship between the pixel coordinate system and the world coordinate system, and the corner point positions in the pixel coordinate system at the at least two moments;

[0171] The conversion relationship is determined based on the shooting parameters of the monitoring device that shoots the monitoring images, or based on the pixel settlement information of the building in the pixel coordinate system and the true settlement information of the building in the world coordinate system.

[0172] Based on any of the above embodiments, the device further includes a conversion determination unit, configured to:

[0173] Perform corner detection on the reference images of the building at at least two reference times to obtain the reference corner positions of the building in the reference images at the at least two reference times;

[0174] Determine the pixel settlement information in the pixel coordinate system based on the reference corner positions at the at least two reference times;

[0175] Obtain the true settlement information in the world coordinate system detected at the at least two reference times;

[0176] Determine the conversion relationship based on the pixel settlement information and the true settlement information.

[0177] Based on any of the above embodiments, the conversion determination unit is configured to:

[0178] Calculate a first error between the pixel settlement information and the true settlement information;

[0179] In the case where the first error is less than a first threshold, determine the conversion relationship based on the pixel settlement information and the true settlement information.

[0180] Based on any of the above embodiments, the conversion determination unit is configured to:

[0181] Determine the pixel settlement rates and true settlement rates at two reference times from the pixel settlement information and the true settlement information;

[0182] Determine the first error based on the ratio between the pixel settlement rate and the true settlement rate at two reference times.

[0183] Based on any of the above embodiments, the conversion determination unit is further configured to:

[0184] In the case where the first error is greater than or equal to the first threshold, adjust at least one of the shooting parameters of the monitoring device for shooting the reference image, the algorithm parameters for performing corner detection on the reference image, and the interval duration of the reference times until the re-determined first error is less than the first threshold;

[0185] In the case where the first error is less than the first threshold, determine the shooting parameters of the monitoring device for shooting the monitoring image based on the shooting parameters of the monitoring device for shooting the reference image, determine the time for shooting the monitoring image based on the interval duration of the reference times, and determine the algorithm parameters for performing corner detection on the monitoring image based on the algorithm parameters for performing corner detection on the reference image.

[0186] Based on any of the above embodiments, the device further includes a parameter determination unit, configured to:

[0187] Perform corner detection on the reference images of the building at at least two reference times to obtain the reference corner positions of the building in the reference images at the at least two reference times;

[0188] Based on the reference corner positions at the at least two reference times and the shooting parameters of the monitoring device that shoots the reference images, determine the monitored settlement information in the world coordinate system;

[0189] Obtain the true settlement information in the world coordinate system detected at the at least two reference times;

[0190] Calculate the second error between the monitored settlement information and the true settlement information. When the second error is less than the second threshold, determine the shooting parameters of the monitoring device that shoots the monitoring image based on the shooting parameters of the monitoring device that shoots the reference image, determine the time to shoot the monitoring image based on the interval duration of the reference times, and determine the algorithm parameters for performing corner detection on the monitoring image based on the algorithm parameters for performing corner detection on the reference image.

[0191] Figure 6 An example of a physical structure diagram of an electronic device is shown as Figure 6 shown. The electronic device may include: a processor 610, a communications interface 620, a memory 630, and a communication bus 640. Among them, the processor 610, the communications interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logical instructions in the memory 630 to execute the building settlement monitoring method, and this method includes:

[0192] Obtain the monitoring images of the building at at least two times;

[0193] Perform corner detection on the monitoring images at the at least two times to obtain the corner positions of the building in the monitoring images at the at least two times;

[0194] Based on the corner positions at the at least two times, determine the settlement rate of the building.

[0195] In addition, when the logical instructions in the above-mentioned memory 630 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0196] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the building settlement monitoring method provided by the above-mentioned various methods. The method includes:

[0197] Obtain monitoring images of the building at at least two moments;

[0198] Perform corner detection on the monitoring images at the at least two moments to obtain the corner positions of the building in the monitoring images at the at least two moments;

[0199] Based on the corner positions at the at least two moments, determine the settlement rate of the building.

[0200] On yet another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the building settlement monitoring method provided by the above-mentioned various methods. The method includes:

[0201] Obtain monitoring images of the building at at least two moments;

[0202] Perform corner detection on the monitoring images at the at least two moments to obtain the corner positions of the building in the monitoring images at the at least two moments;

[0203] Based on the corner positions at the at least two moments, determine the settlement rate of the building.

[0204] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0205] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring building settlement, characterized in that, it includes: Obtaining monitoring images of the building at at least two moments; Performing corner detection on the monitoring images at the at least two moments to obtain the corner positions of the building in the monitoring images at the at least two moments; Based on the corner positions at the at least two moments, determining the settlement rate of the building.

2. The method for monitoring building settlement according to claim 1, characterized in that, the determining the settlement rate of the building based on the corner positions at the at least two moments includes: Based on the conversion relationship between the pixel coordinate system and the world coordinate system, and the corner positions in the pixel coordinate system at the at least two moments, determining the settlement rate of the building in the world coordinate system; The conversion relationship is determined based on the shooting parameters of the monitoring device that shoots the monitoring images, or based on the pixel settlement information of the building in the pixel coordinate system and the true settlement information of the building in the world coordinate system.

3. The method for monitoring building settlement according to claim 2, characterized in that, the determining steps of the conversion relationship include: Performing corner detection on the reference images of the building at at least two reference moments to obtain the reference corner positions of the building in the reference images at the at least two reference moments; Based on the reference corner positions at the at least two reference moments, determining the pixel settlement information in the pixel coordinate system; Obtaining the true settlement information in the world coordinate system detected at the at least two reference moments; Based on the pixel settlement information and the true settlement information, determining the conversion relationship.

4. The method for monitoring building settlement according to claim 3, characterized in that, the determining the conversion relationship based on the pixel settlement information and the true settlement information includes: Calculating a first error between the pixel settlement information and the true settlement information; When the first error is less than a first threshold, determining the conversion relationship based on the pixel settlement information and the true settlement information.

5. The method for monitoring building settlement according to claim 4, characterized in that, the calculating the first error between the pixel settlement information and the true settlement information includes: Determining the pixel settlement rates and the true settlement rates at two reference moments from the pixel settlement information and the true settlement information; Based on the ratio between the pixel settlement rate and the true settlement rate at two reference moments, determining the first error.

6. The method for monitoring building settlement according to claim 4, characterized in that, the determining the conversion relationship based on the pixel settlement information and the true settlement information further includes: When the first error is greater than or equal to the first threshold, adjusting at least one of the shooting parameters of the monitoring device that shoots the reference images, the algorithm parameters for performing corner detection on the reference images, and the interval duration between the reference moments until the re-determined first error is less than the first threshold; When the first error is less than the first threshold, determine the shooting parameters of the monitoring device for shooting the monitoring image based on the shooting parameters of the monitoring device for shooting the reference image, determine the time for shooting the monitoring image based on the interval duration at the reference time, and determine the algorithm parameters for corner detection of the monitoring image based on the algorithm parameters for corner detection of the reference image.

7. The building settlement monitoring method according to claim 1 or 2, characterized in that, before obtaining the monitoring images of the building at at least two times, further includes: performing corner detection on the reference images of the building at at least two reference times to obtain the reference corner positions of the building in the reference images at the at least two reference times; determine the monitored settlement information in the world coordinate system based on the reference corner positions at the at least two reference times and the shooting parameters of the monitoring device for shooting the reference images; obtain the true settlement information in the world coordinate system detected at the at least two reference times; calculate a second error between the monitored settlement information and the true settlement information. When the second error is less than the second threshold, determine the shooting parameters of the monitoring device for shooting the monitoring image based on the shooting parameters of the monitoring device for shooting the reference image, determine the time for shooting the monitoring image based on the interval duration at the reference time, and determine the algorithm parameters for corner detection of the monitoring image based on the algorithm parameters for corner detection of the reference image.

8. A building settlement monitoring device, characterized in that, comprises: an image acquisition unit for acquiring monitoring images of a building at at least two times; a corner detection unit for performing corner detection on the monitoring images at the at least two times to obtain the corner positions of the building in the monitoring images at the at least two times; a settlement monitoring unit for determining the settlement rate of the building based on the corner positions at the at least two times.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the building settlement monitoring method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, it implements the building settlement monitoring method according to any one of claims 1 to 7.

Images