A measuring device and method for building construction

Abstract

The invention discloses a measuring device for building construction and a method thereof, and relates to the field of measurement technology. The measuring device for building construction comprises a visual module and a control system. The visual module is used to collect a target surface image; the control system is used to obtain the target surface image, select a target feature of the target surface image, extract the target feature edge and a crack skeleton line, calculate a skeleton line normal vector according to the crack skeleton line, and calculate the crack width based on the skeleton line normal vector and the target feature edge; the invention realizes the automatic measurement of the crack width of a wall, a concrete component or other plane of a building structure, and the operation is simple and the measurement data is accurate.

Description

Technical Field

[0001] The present invention relates to the field of measurement technology, and in particular to a measurement device and a method for building construction. Background Art

[0002] my country is a country with a large number of buildings, many of which are cracked. Cracks have a vital impact on the safety and durability of buildings. Building cracks mainly exist in concrete. Concrete cracks can be divided into micro cracks and macro cracks. Micro cracks have little effect on the structure itself. Most concrete works with tiny cracks. Concrete macro cracks are concrete cracks. If their width exceeds the calculation range of crack control in the specification, they can be regarded as harmful cracks, which will bring safety hazards to the structure and may cause safety accidents. Therefore, crack detection has always been widely concerned by domestic scholars.

[0003] At present, the following methods are mainly used to detect the width of concrete cracks: first, measurement with a feeler gauge or crack width comparison card; second, measurement using a crack microscope equipped with a light source; third, use a crack width tester, display the crack image on the screen and measure it by manually reading the width.

[0004] The existing crack width detection methods have the following disadvantages: first, the related instruments are expensive and uneconomical, and the operation of the instruments is complicated and requires professional personnel to operate; second, the above detection methods all require manual operation, and manual observation of concrete structure cracks has a certain degree of subjectivity, including instrument accuracy problems, resulting in low accuracy of crack detection data. Summary of the invention

[0005] Therefore, in order to solve the above-mentioned shortcomings, on the one hand, the present invention provides a measurement device for construction, comprising:

[0006] A visual module, used to collect target surface images;

[0007] The control system acquires a target surface image, selects a target feature of the target surface image, extracts the target feature edge and the crack skeleton line, calculates the skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton line normal vector and the target feature edge;

[0008] The control system comprises:

[0009] An information acquisition unit, used to acquire a target surface image acquired by a visual module;

[0010] An image processing unit is used to select a target feature in the target surface image, form a target feature selection area, and extract the target feature edge and crack skeleton line in the target feature selection area;

[0011] The width calculation unit calculates the skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton normal vector and the target feature edge.

[0012] The present invention obtains a target surface image collected by a visual module; selects a target feature in the target surface image to form a target feature selection area, extracts the target feature edge and the crack skeleton line in the target feature selection area; calculates a skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton normal vector and the target feature edge, thereby realizing automatic measurement of the crack width of a wall, concrete component or other building structure plane, with simple operation and accurate measurement data.

[0013] Furthermore, the measuring device also includes a walking device, and the control system and the visual module are mounted on the walking device, and the whole measuring device is driven to move on the ground by the walking device.

[0014] The whole device is driven by the walking device to move, and after the measurement work in the current area is completed, it can automatically move to the next measurement area, thus realizing the automation of the measurement work.

[0015] Furthermore, the measuring device further comprises a displacement assembly, which is integrally mounted on the walking device, and the visual module is mounted on the displacement assembly, and the visual module is driven to move linearly by the displacement assembly;

[0016] The displacement assembly includes a lifting structure, through which the visual module is driven to move vertically linearly;

[0017] The lifting structure comprises:

[0018] Second power unit;

[0019] A fixed column, the fixed column is fixedly installed on the walking device;

[0020] A movable column, which is wrapped around the outside of the fixed column and driven to move vertically by a second power device;

[0021] The mounting platform is mounted on a movable column and moves vertically linearly with the movable column. The visual module is mounted on the mounting platform.

[0022] The second power device drives the entire installation platform to move vertically, thereby driving the visual module to move vertically linearly, thereby increasing the upper limit of the effective height of the visual module.

[0023] Furthermore, a first sensor is provided on one side of the installation platform, and the first sensor is used to detect a first distance from both ends of the side surface of the installation platform to the target surface;

[0024] The measuring device also includes a posture control unit connected to the information collection unit, which performs parallel judgment according to the first distance to obtain a parallel judgment result. If not parallel, the walking device is controlled to move to adjust the parallelism between the device and the target surface.

[0025] Through the posture control unit, a parallel judgment is made according to the first distance to obtain a parallel judgment result. If not parallel, the movement of the walking device is controlled to adjust the parallelism between the device and the target surface, thereby keeping the side of the installation platform parallel to the target surface, thereby improving the measurement accuracy.

[0026] Furthermore, the lifting structure also includes a first power device, which drives the installation platform to move vertically linearly along the movable column.

[0027] The first power device drives the installation platform to move linearly relative to the movable column, thereby increasing the lower limit of the visual module's operating height. In conjunction with the above-mentioned second power device, the visual module's operating height range is increased.

[0028] Furthermore, the displacement assembly further comprises a traveling device, the traveling device is mounted on the mounting platform, and the visual module is mounted on the traveling device;

[0029] The traveling device comprises:

[0030] a fifth power device, through which the vision module is driven to slide vertically relative to the mounting platform;

[0031] A fourth power device, through which the visual module is driven to perform horizontal lateral linear movement;

[0032] A third power device is used to drive the visual module to move horizontally and longitudinally linearly.

[0033] Furthermore, a second sensor is provided on the above-mentioned visual module, and a second distance between the visual module and the target surface is acquired by the second sensor;

[0034] The measuring device also includes a working control unit connected to the information collection unit, and configured to perform a spacing judgment according to the second distance, and control the traveling device to perform an action according to the spacing judgment result.

[0035] The visual module is driven by the traveling device to approach the target surface. During this process, the second sensor obtains the second distance between the visual module and the target surface; the working control unit makes a spacing judgment based on the second distance. If the second distance is equal to the preset distance, the third power device is controlled to stop working, and then the fourth power device is controlled to drive the visual module to move laterally. After moving to the last stroke, the fifth power device drives the visual module to rise to a preset height, and the fourth power device drives the visual module to move in the opposite direction, and so on and so forth until it rises to the limit height or the maximum height of the target surface.

[0036] On the other hand, the present invention also provides a construction measurement method, the measurement method using the above construction measurement device, the measurement method comprising:

[0037] Control the device to move to the target initial measurement position by remote control, pushing or remote control;

[0038] Performing parallel judgment according to the first distance to obtain a parallel judgment result, and if not parallel, controlling the movement of the walking device to adjust the parallelism between the device and the target surface;

[0039] Specifically, the information collection unit obtains the first distance collected by the first sensors at both ends of the side of the installation platform respectively. l 1 , l 2 Upload to the attitude control unit, the attitude control unit l 1 , l 2 to judge l 1 Is it equal to l 2 If it is not equal, the walking device is controlled to move and the angle is adjusted until l 1 , l 2 equal;

[0040] If they are parallel, the walking device is controlled to move in a direction parallel to the target surface and walk into the target surface to obtain the target measurement working length;

[0041] The method is: by light quantity discrimination, when there is a target at a preset distance position of the third sensor, the light quantity is reflected and received, and begins to accumulate until the target is not detected or the fourth sensor detects that the distance between the obstacle and the device reaches a dangerous distance;

[0042] The target measurement working length is calculated as follows:

[0043]

[0044] in, L To measure the working length, the unit is m ; V It is the walking speed of the device when measuring the working length of the target, in units of m / s ; T It is the time taken by the device from the initial position to the point where the target cannot be detected or the fourth sensor detects that the distance between the obstacle and the device reaches the dangerous distance, in units of s ; a The dangerous distance between the obstacle and the device. When there is no obstacle, a Take 0, the unit is m ;

[0045] After the measurement is completed, the device returns to the initial position, and the traveling device drives the visual module to approach the target surface. During this process, the second sensor obtains a second distance between the visual module and the target surface;

[0046] The working control unit determines the distance according to the second distance, and controls the third power device to stop working if the second distance is equal to the preset distance;

[0047] Then, the fourth power device is controlled to drive the visual module to move horizontally. After the movement reaches the final stroke, the fifth power device drives the visual module to rise to a preset height, and the fourth power device drives the visual module to move in the opposite direction, and the process is repeated until the limit height or the maximum height of the target surface is reached. During this process, the visual module performs crack measurement on the target surface.

[0048] The crack measurement work includes:

[0049] Acquire the target surface image acquired by the vision module;

[0050] Select the target feature in the target surface image to form a target feature selection area, and extract the target feature edge and crack skeleton line in the target feature selection area;

[0051] Calculate the skeleton line normal vector according to the crack skeleton line, and calculate the crack width based on the skeleton normal vector and the target feature edge;

[0052] The step of extracting target feature edges and crack skeleton lines in the target feature selection area includes:

[0053] Faster R-CNN and YOLO V5(x) target detection network models are used to perform initial identification and extraction of bridge cracks;

[0054] The target detection network model Faster R-CNN with Resnet50 as the backbone feature extraction network and the target detection network model YOLO V5(x) with CSPDarknet53 as the backbone feature extraction network were used to perform secondary recognition and extraction of bridge cracks under the TensorFlow2 framework;

[0055] In the PyTorch framework, the U-Net semantic segmentation network model with VGG16 as the backbone feature extraction network is used to segment and extract cracks;

[0056] The crack edge is extracted by the Alpha Shapes algorithm to form the crack edge line; the crack skeleton line is extracted by iterative calculation of the medial axis transformation;

[0057] The calculation of the crack width based on the skeleton normal vector and the target feature edge includes:

[0058] Determine the intersection point between the skeleton normal vector and the target feature edge line;

[0059] Obtaining the intersection coordinates, calculating the crack width based on the intersection coordinates, and obtaining the crack width calculation result;

[0060] The calculation of the crack width based on the skeleton normal vector and the target feature edge includes:

[0061] Determine the intersection point between the skeleton normal vector and the target feature edge line;

[0062] Obtaining the intersection coordinates, calculating the crack width based on the intersection coordinates, and obtaining the crack width calculation result;

[0063] The crack width is calculated as follows:

[0064]

[0065] in, H The crack width of the preset point on the crack skeleton line, in units of mm ; X 1 , X 2 is the horizontal coordinate of the intersection point between the skeleton normal vector and the target feature edge line, Y 1 , Y 2 is the ordinate of the intersection point between the skeleton normal vector and the target feature edge line;

[0066] The skeleton normal vector is calculated by estimating the normal vector of the skeleton line using SVD singular decomposition, thereby obtaining the skeleton normal vector;

[0067] In the above measurement process, the information acquisition unit obtains the measured working length, the work control unit calculates the remaining measured working length according to the measured working length, and controls the travel of the walking device and the displacement component based on the remaining measured working length;

[0068] The remaining measuring working length is calculated as follows:

[0069]

[0070] in, C is the remaining measuring working length in units of m ; L To measure the working length, the unit is m ; n is the number of times the device walks during the measurement process; X is the length of each walk, in units of m ;

[0071] When the measurement work of the current area is completed, if C Greater than X , then the walking distance of the walking device of the control device is X Then continue the measurement work, the displacement component drives the visual module to move in the horizontal direction parallel to the target feature X Length; if C Less than X , then the walking distance of the walking device of the control device is X At the same time, during measurement, the displacement component drives the visual module to move in a horizontal direction parallel to the target feature. C If there is an obstacle at the end of the target feature, that is, when the fourth sensor at the end of the walking device detects the obstacle, the working control unit controls the walking device to walk. C Length, control displacement component drives the vision module to move in the horizontal direction parallel to the target feature X - C length, so that it moves to the initial position of the current area ( X - C , 0), and then start measuring. After each rise to the preset height, the horizontal coordinate of the point X - C is the initial or final horizontal coordinate.

[0072] The present invention has the following advantages:

[0073] The present invention obtains a target surface image collected by a visual module; selects a target feature in the target surface image to form a target feature selection area, extracts the target feature edge and the crack skeleton line in the target feature selection area; calculates a skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton normal vector and the target feature edge, thereby realizing automatic measurement of the crack width of a wall, concrete component or other building structure plane, with simple operation and accurate measurement data.

[0074] The whole device is driven by the walking device to move, and after the measurement work in the current area is completed, it can automatically move to the next measurement area, thus realizing the automation of the measurement work.

[0075] The second power device drives the entire mounting platform to move vertically, thereby driving the visual module to move vertically linearly, thereby increasing the upper limit of the visual module's action height. The first power device drives the mounting platform to move linearly relative to the movable column, thereby increasing the lower limit of the visual module's action height. In conjunction with the second power device, the visual module's action height range is increased.

[0076] Through the posture control unit, a parallel judgment is made according to the first distance to obtain a parallel judgment result. If not parallel, the movement of the walking device is controlled to adjust the parallelism between the device and the target surface, thereby keeping the side of the installation platform parallel to the target surface, thereby improving the measurement accuracy.

[0077] The visual module is driven by the traveling device to approach the target surface. During this process, the second sensor obtains the second distance between the visual module and the target surface; the working control unit makes a spacing judgment based on the second distance. If the second distance is equal to the preset distance, the third power device is controlled to stop working, and then the fourth power device is controlled to drive the visual module to move laterally. After moving to the last stroke, the fifth power device drives the visual module to rise to a preset height, and the fourth power device drives the visual module to move in the opposite direction, and so on and so forth until it rises to the limit height or the maximum height of the target surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a control logic diagram of a measuring device used in building construction;

[0079] Figure 2 yes Figure 1 The logical diagram of the control system in the measuring device for building construction shown;

[0080] Figure 3 It is a schematic diagram of the structure of a measuring device used in building construction;

[0081] Figure 4 yes Figure 3 A front view of the measuring device for building construction shown;

[0082] Figure 5 yes Figure 3 A side view of the measuring device for building construction shown;

[0083] Figure 6 yes Figure 5 A first structural schematic diagram of a lifting structure in a measuring device for building construction is shown;

[0084] Figure 7 yes Figure 6 A second structural schematic diagram of the lifting structure shown;

[0085] Figure 8 yes Figure 5 A side view of the lifting structure shown;

[0086] Fig. 9 yes Figure 6 A schematic diagram of the internal structure of the fixed column and the movable column in the lifting structure shown;

[0087] Fig.10 yes Figure 5 The schematic diagram of the structure of the traveling device in the measuring device for building construction shown;

[0088] In the figure:

[0089] 100. Walking device;

[0090] 200, displacement assembly; 210, lifting structure; 211, installation platform; 212, fixed column; 213, movable column; 214, first power device; 215, rack; 216, second power device; 217, gear box; 218, sprocket mechanism; 219, guide structure; 220, travel device; 221, third power device; 222, first transmission mechanism; 223, fourth power device; 224, second transmission mechanism; 225, fifth power device; 226, third transmission mechanism;

[0091] 300. Data Center;

[0092] 400, control system; 410, information acquisition unit; 420, image processing unit; 430, posture control unit; 440, work control unit; 450, width calculation unit; 460, communication unit;

[0093] 500, visual module;

[0094] 600, first sensor;

[0095] 700. Second sensor. DETAILED DESCRIPTION

[0096] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0097] In this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises", or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0098] As described in the background technology, the existing crack width detection methods have the following shortcomings: first, the related instruments are expensive and uneconomical, and the operation of the instruments is complicated and requires professional personnel to operate; second, the above detection methods all require manual operation, and manual observation of concrete structure cracks has a certain degree of subjectivity, including instrument accuracy problems, resulting in low accuracy of crack detection data.

[0099] Embodiment 1:

[0100] Therefore, in order to solve the above technical problems existing in the prior art, this embodiment provides a measurement device for construction, such as Figure 1 As shown, the measuring device for building construction comprises:

[0101] A visual module 500 is used to collect target surface images;

[0102] Control system 400, acquires a target surface image, selects a target feature of the target surface image, extracts the target feature edge and the crack skeleton line, calculates the skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton line normal vector and the target feature edge;

[0103] like Figure 2 As shown, the control system includes:

[0104] An information acquisition unit 410 is used to acquire a target surface image acquired by a visual module;

[0105] The image processing unit 420 selects the target feature in the target surface image to form a target feature selection area, and extracts the target feature edge and crack skeleton line in the target feature selection area;

[0106] The width calculation unit 450 calculates a skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton normal vector and the target feature edge.

[0107] In this embodiment, the information acquisition unit obtains the target surface image captured by the visual module, the image processing unit selects the target features in the target surface image to form an M*N target feature selection area, extracts the target feature edge and the crack skeleton line in the target feature selection area, and the width calculation unit calculates the skeleton line normal vector according to the crack skeleton line, the skeleton line normal vector includes the normal vector of each point of the skeleton line, and calculates the crack width based on the skeleton normal vector and the target feature edge.

[0108] Among them, extracting the target feature edge and crack skeleton line in the target feature selection area includes:

[0109] Faster R-CNN and YOLO V5(x) target detection network models are used to perform initial identification and extraction of bridge cracks;

[0110] The target detection network model Faster R-CNN with Resnet50 as the backbone feature extraction network and the target detection network model YOLO V5(x) with CSPDarknet53 as the backbone feature extraction network were used to perform secondary recognition and extraction of bridge cracks under the TensorFlow2 framework;

[0111] In the PyTorch framework, the U-Net semantic segmentation network model with VGG16 as the backbone feature extraction network is used to segment and extract cracks;

[0112] The crack edge is extracted by the Alpha Shapes algorithm to form the crack edge line; the crack skeleton line is extracted by iterative calculation of the medial axis transformation;

[0113] By identifying and segmenting the cracks in the above manner, the crack features of the target surface can be accurately extracted, and the accurate crack skeleton lines and edge lines can be obtained.

[0114] The calculation of the crack width based on the skeleton normal vector and the target feature edge includes:

[0115] Determine the intersection point between the skeleton normal vector and the target feature edge line;

[0116] The intersection coordinates are obtained, the crack width is calculated based on the intersection coordinates, and the crack width calculation result is obtained.

[0117] The present invention realizes the automatic measurement of the width of cracks on a wall, concrete component or other planes of a building structure, with simple operation and accurate measurement data.

[0118] In this embodiment, the crack width is calculated as follows:

[0119]

[0120] in, H The crack width of the preset point on the crack skeleton line, in units of mm ; X 1 , X 2 is the horizontal coordinate of the intersection point between the skeleton normal vector and the target feature edge line, Y 1 , Y 2 It is the ordinate of the intersection point between the skeleton normal vector and the target feature edge line.

[0121] The skeleton normal vector is calculated by using SVD singular decomposition to estimate the normal vector of the skeleton line, thereby obtaining the skeleton normal vector.

[0122] In this embodiment, if Figure 2 As shown, the control system further includes a communication unit 460 , through which the data calculated by the width calculation unit can be uploaded to the data center 300 .

[0123] Embodiment 2:

[0124] In order to realize automatic scanning and detection of target surface, such as Figure 1 As shown, the measuring device in this embodiment may further include a walking device 100, on which the control system and the vision module are mounted, and the whole measuring device is driven to move on the ground by the walking device.

[0125] The whole device is driven by the walking device to move, and after the measurement work in the current area is completed, it can automatically move to the next measurement area, thus realizing the automation of the measurement work.

[0126] In this embodiment, if Figures 3 to 5 As shown, the above-mentioned walking device can be selected from an electric-driven vehicle, an internal combustion-driven vehicle or other vehicles that have their own power and can automatically walk on the ground.

[0127] like Figure 1 As shown, the measuring device may further include a displacement component 200, such as Figures 3 to 5 As shown, the displacement assembly is integrally mounted on the walking device, and the above-mentioned visual module is mounted on the displacement assembly 200, and the visual module is driven to move linearly by the displacement assembly;

[0128] like Figure 5 As shown, the displacement assembly includes a lifting structure 210, through which the visual module is driven to move vertically linearly;

[0129] like Figure 6 As shown, the lifting structure includes:

[0130] Second power unit 216;

[0131] A fixed column 212, the fixed column is fixedly mounted on the walking device;

[0132] A movable column 213, which is wrapped around the outside of the fixed column 212 and driven by a second power device to move vertically;

[0133] The mounting platform 211 is mounted on a movable column and moves vertically linearly with the movable column. The vision module is mounted on the mounting platform 211 .

[0134] The entire installation platform 211 is driven to move vertically by the second power device, thereby driving the visual module to move vertically linearly, thereby increasing the upper limit of the effective height of the visual module.

[0135] In this embodiment, if Fig. 9 As shown, a sprocket mechanism 218 is installed in the movable column, and the sprocket mechanism is driven to move by the second power device. The chain of the sprocket mechanism is connected to the upper part of the fixed column, and the sprocket mechanism is driven to move by the second power device, thereby realizing the linear movement of the movable column; in addition, a pulley mechanism, a gear rack mechanism, a screw transmission mechanism, etc. can also be used.

[0136] In this embodiment, if Fig. 9 As shown, a guide structure 219 can also be provided in the above-mentioned fixed column and movable column. The movement of the movable column is guided by the guide structure.

[0137] Embodiment 3:

[0138] In this embodiment, if Figure 5 As shown, a first sensor 600 may also be provided on one side of the installation platform, and the first sensor is used to detect a first distance from both ends of the side surface of the installation platform to the target surface;

[0139] The control system also includes a posture control unit 430, which performs parallel judgment based on the first distance to obtain a parallel judgment result. If they are not parallel, the walking device is controlled to move to adjust the parallelism between the device and the target surface.

[0140] Specifically, the information collection unit obtains the first distance collected by the first sensors at both ends of the side of the installation platform respectively. l 1 , l 2 Upload to the attitude control unit, the attitude control unit l1 , l 2 to judge l 1 Is it equal to l 2 If it is not equal, the walking device is controlled to move and the angle is adjusted until l 1 , l 2 equal.

[0141] Through the posture control unit, a parallel judgment is made according to the first distance to obtain a parallel judgment result. If not parallel, the movement of the walking device is controlled to adjust the parallelism between the device and the target surface, thereby keeping the side of the installation platform parallel to the target surface, thereby improving the measurement accuracy.

[0142] Embodiment 4:

[0143] In order to increase the scope of the visual module, such as Figure 6 , Figure 7 , Figure 8 As shown, the lifting structure in this embodiment may further include a first power device 214, which drives the installation platform to move vertically linearly along the movable column.

[0144] In this embodiment, if Figure 6 , Figure 8 As shown, a rack 215 can be installed on the outer surface of the movable column, and a gear box 217 can be installed on the lower surface of the mounting platform. The gear box 217 cooperates with the rack and is connected to the second power device. The second power device drives the gear in the gear box to rotate, thereby realizing the linear movement of the mounting platform. Of course, in addition to the gear rack structure, a sprocket transmission mechanism, a pulley transmission, a screw transmission mechanism, etc. can also be used.

[0145] In this embodiment, if Figure 7 As shown, a guide structure may also be installed on the side of the movable column, through which the linear movement of the mounting platform is guided.

[0146] The first power device drives the installation platform to move linearly relative to the movable column, thereby increasing the lower limit of the visual module's action height. In conjunction with the second power device, the visual module's action height range is increased.

[0147] Embodiment 5:

[0148] In order to increase the range of the visual module, reduce the walking frequency of the walking device, such as Figure 5 As shown, the displacement assembly further includes a traveling device 220, which is mounted on a mounting platform, and the visual module is mounted on the traveling device;

[0149] like Fig.10 As shown, the traveling device comprises:

[0150] A fifth power device 225, through which the vision module is driven to slide vertically relative to the mounting platform;

[0151] A fourth power device 223, through which the visual module is driven to perform horizontal transverse linear movement;

[0152] The third power device 221 drives the visual module to move horizontally and longitudinally linearly.

[0153] In this embodiment, a second sensor 700 is provided on the above-mentioned visual module, and a second distance between the visual module and the target surface is acquired through the second sensor;

[0154] The control system further includes a working control unit 440, which is used to perform a spacing judgment according to the second distance, and control the traveling device to perform an action according to the spacing judgment result.

[0155] The visual module is driven by the traveling device to approach the target surface. During this process, the second sensor obtains the second distance between the visual module and the target surface; the working control unit 440 makes a spacing judgment based on the second distance. If the second distance is equal to the preset distance, the third power device is controlled to stop working, and then the fourth power device is controlled to drive the visual module to move laterally. After moving to the last stroke, the fifth power device drives the visual module to rise to a preset height, and the fourth power device drives the visual module to move in the opposite direction, and so on and so forth until it rises to the limit height or the maximum height of the target surface.

[0156] In this embodiment, if Fig.10 As shown, the traveling device also includes:

[0157] A first transmission mechanism 222, which is fixed on the upper surface of the mounting platform and driven to move by a third power device;

[0158] A second transmission mechanism 224, which is connected to the first transmission mechanism and moves linearly under the action of the first transmission mechanism. The second transmission mechanism 224 is vertically arranged with the first transmission mechanism and is driven to move by a fourth power device;

[0159] A third transmission mechanism 226, which is connected to the second transmission mechanism, and is vertically arranged as a whole, and drives the third transmission mechanism to move linearly under the action of the second transmission mechanism, and drives the third transmission mechanism to move through the fifth power device;

[0160] The visual module is installed on the third transmission mechanism.

[0161] In this embodiment, the first, second and third transmission mechanisms can all be lead screws, and can all be arranged inside the guide rails. The transmission mechanisms can slide relative to each other through the connection between the slider and the guide rail.

[0162] Embodiment 7:

[0163] In this embodiment, a third sensor may be provided on both sides of the walking device, and the measured working length of the target is collected by the third sensor;

[0164] The information acquisition unit obtains the measured working length, the working control unit calculates the remaining measured working length according to the measured working length, and controls the travel device and the displacement component stroke based on the remaining measured working length;

[0165] The remaining measuring working length is calculated as follows:

[0166]

[0167] in, C is the remaining measuring working length in units of m ; L To measure the working length, the unit is m ; n is the number of times the device walks during the measurement process; X is the length of each walk, in units of m ;

[0168] When the measurement work of the current area is completed, if C Greater than X , then the walking distance of the walking device of the control device is X Then continue the measurement work, the displacement component drives the visual module to move in the horizontal direction parallel to the target feature X Length; if C Less than X , then the walking distance of the walking device of the control device is X At the same time, during measurement, the displacement component drives the visual module to move in a horizontal direction parallel to the target feature. C If there is an obstacle at the end of the target feature, that is, when the fourth sensor at the end of the walking device detects the obstacle, the working control unit controls the walking device to walk. C Length, control displacement component drives the vision module to move in the horizontal direction parallel to the target feature X - C length, so that it moves to the initial position of the current area ( X - C , 0), and then start measuring. After each rise to the preset height, the horizontal coordinate of the pointX - C is the initial horizontal coordinate or the end horizontal coordinate. For example, when the visual module is from the coordinate ( X - C ,0) to ( X , 0) coordinates, the fifth power device drives the visual module to rise to a higher height H , then the terminal position of the horizontal motion stroke of this end is ( X - C , H ), and so on.

[0169] Through the above method, it is possible to realize automatic planning of the measurement work, thereby reducing human participation and realizing the automation of the measurement work.

[0170] Embodiment 8:

[0171] This embodiment provides a construction measurement method, which uses the construction measurement device described above. The measurement method includes:

[0172] Control the device to move to the target initial measurement position by remote control, pushing or remote control;

[0173] A parallel judgment is performed according to the first distance to obtain a parallel judgment result. If they are not parallel, the walking device is controlled to move to adjust the parallelism between the device and the target surface.

[0174] Specifically, the information collection unit obtains the first distance collected by the first sensors at both ends of the side of the installation platform respectively. l 1 , l 2 Upload to the attitude control unit, the attitude control unit l 1 , l 2 to judge l 1 Is it equal to l 2 If it is not equal, the walking device is controlled to move and the angle is adjusted until l 1 , l 2 equal.

[0175] If they are parallel, the walking device is controlled to move in a direction parallel to the target surface and walk into the target surface to obtain the target measurement working length;

[0176] The principle is: by means of light quantity discrimination, when there is a target at a preset distance position of the third sensor, the light quantity is reflected and received, and begins to accumulate until the target is no longer detected or the fourth sensor detects that the distance between the obstacle and the device reaches a dangerous distance;

[0177] The target measurement working length is calculated as follows:

[0178]

[0179] in, L To measure the working length, the unit is m ; V It is the walking speed of the device when measuring the working length of the target, in units of m / s ; T It is the time taken by the device from the initial position to the point where the target cannot be detected or the fourth sensor detects that the distance between the obstacle and the device reaches the dangerous distance, in units of s ; a The dangerous distance between the obstacle and the device. When there is no obstacle, a Take 0, the unit is m .

[0180] After the measurement is completed, the device returns to the initial position, and the traveling device drives the visual module to approach the target surface. During this process, the second sensor obtains a second distance between the visual module and the target surface;

[0181] The working control unit determines the distance according to the second distance, and controls the third power device to stop working if the second distance is equal to the preset distance;

[0182] Then the fourth power device is controlled to drive the visual module to move horizontally. After moving to the final stroke, the fifth power device drives the visual module to rise to a preset height, and the fourth power device drives the visual module to move in the opposite direction, and repeats this process until it rises to the limit height or the maximum height of the target surface. During this process, the visual module measures cracks on the target surface.

[0183] The crack measurement work includes:

[0184] Acquire the target surface image acquired by the vision module;

[0185] Select the target feature in the target surface image to form a target feature selection area, and extract the target feature edge and crack skeleton line in the target feature selection area;

[0186] Calculate the skeleton line normal vector according to the crack skeleton line, and calculate the crack width based on the skeleton normal vector and the target feature edge;

[0187] The step of extracting target feature edges and crack skeleton lines in the target feature selection area includes:

[0188] Faster R-CNN and YOLO V5(x) target detection network models are used to perform initial identification and extraction of bridge cracks;

[0189] The target detection network model Faster R-CNN with Resnet50 as the backbone feature extraction network and the target detection network model YOLO V5(x) with CSPDarknet53 as the backbone feature extraction network were used to perform secondary recognition and extraction of bridge cracks under the TensorFlow2 framework;

[0190] In the PyTorch framework, the U-Net semantic segmentation network model with VGG16 as the backbone feature extraction network is used to segment and extract cracks;

[0191] The crack edge is extracted by the Alpha Shapes algorithm to form the crack edge line; the crack skeleton line is extracted by iterative calculation of the medial axis transformation;

[0192] The calculation of the crack width based on the skeleton normal vector and the target feature edge includes:

[0193] Determine the intersection point between the skeleton normal vector and the target feature edge line;

[0194] The intersection coordinates are obtained, the crack width is calculated based on the intersection coordinates, and the crack width calculation result is obtained.

[0195] The calculation of the crack width based on the skeleton normal vector and the target feature edge includes:

[0196] Determine the intersection point between the skeleton normal vector and the target feature edge line;

[0197] The intersection coordinates are obtained, the crack width is calculated based on the intersection coordinates, and the crack width calculation result is obtained.

[0198] In this embodiment, the crack width is calculated as follows:

[0199]

[0200] in, H The crack width of the preset point on the crack skeleton line, in units of mm ; X 1 , X 2 is the horizontal coordinate of the intersection point between the skeleton normal vector and the target feature edge line, Y 1 , Y 2 It is the ordinate of the intersection point between the skeleton normal vector and the target feature edge line.

[0201] The skeleton normal vector is calculated by using SVD singular decomposition to estimate the normal vector of the skeleton line, thereby obtaining the skeleton normal vector.

[0202] In the above measurement process, the information acquisition unit obtains the measured working length, the work control unit calculates the remaining measured working length according to the measured working length, and controls the travel of the walking device and the displacement component based on the remaining measured working length;

[0203] The remaining measuring working length is calculated as follows:

[0204]

[0205] in, C is the remaining measuring working length in units of m ; L To measure the working length, the unit is m ; n is the number of times the device walks during the measurement process; X is the length of each walk, in units of m ;

[0206] When the measurement work of the current area is completed, if C Greater than X , then the walking distance of the walking device of the control device is X Then continue the measurement work, the displacement component drives the visual module to move in the horizontal direction parallel to the target feature X Length; if C Less than X , then the walking distance of the walking device of the control device is X At the same time, during measurement, the displacement component drives the visual module to move in a horizontal direction parallel to the target feature. C If there is an obstacle at the end of the target feature, that is, when the fourth sensor at the end of the walking device detects the obstacle, the working control unit controls the walking device to walk. C Length, control displacement component drives the vision module to move in the horizontal direction parallel to the target feature X - C length, so that it moves to the initial position of the current area ( X - C , 0), and then start measuring. After each rise to the preset height, the horizontal coordinate of the point X - C is the initial horizontal coordinate or the end horizontal coordinate. For example, when the visual module is from the coordinate ( X - C ,0) to ( X,0) coordinates, the fifth power device drives the visual module to rise to a height of H, then the terminal position of the horizontal motion stroke of this end is ( X - C , H ), and so on.

[0207] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A measuring device for construction, characterized in that: The measuring device comprises: A visual module, used to collect target surface images; The control system acquires a target surface image, selects a target feature of the target surface image, extracts the target feature edge and the crack skeleton line, calculates the skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton line normal vector and the target feature edge; The control system comprises: An information acquisition unit, used to acquire a target surface image acquired by a visual module; An image processing unit is used to select a target feature in the target surface image, form a target feature selection area, and extract the target feature edge and crack skeleton line in the target feature selection area; A width calculation unit calculates a skeleton line normal vector according to the crack skeleton line, and calculates the crack width based on the skeleton normal vector and the target feature edge; The measuring device also includes a walking device, the control system and the visual module are mounted on the walking device, and the walking device drives the entire measuring device to move on the ground; The measuring device also includes a displacement component, which is integrally mounted on the walking device, and the visual module is mounted on the displacement component, and the visual module is driven to move linearly by the displacement component; The displacement assembly includes a lifting structure, through which the visual module is driven to move vertically linearly; The lifting structure comprises: Second power unit; A fixed column, the fixed column is fixedly installed on the walking device; A movable column, which is wrapped around the outside of the fixed column and driven to move vertically by a second power device; A mounting platform, the mounting platform is mounted on a movable column and moves vertically linearly with the movable column, and the visual module is mounted on the mounting platform; A first sensor is provided on one side of the installation platform, and the first sensor is used to detect a first distance from both ends of the side surface of the installation platform to the target surface; The measuring device also includes a posture control unit connected to the information collection unit, which performs parallel judgment according to the first distance to obtain a parallel judgment result. If not parallel, the walking device is controlled to move to adjust the parallelism between the device and the target surface.

2. A construction measurement device according to claim 1, characterized in that: The lifting structure also includes a first power device, which drives the installation platform to move vertically linearly along the movable column.

3. A construction measurement device according to claim 1, characterized in that: The displacement assembly further comprises a traveling device, which is mounted on a mounting platform, and the visual module is mounted on the traveling device; The traveling device comprises: a fifth power device, through which the vision module is driven to slide vertically relative to the mounting platform; A fourth power device, through which the visual module is driven to perform horizontal lateral linear movement; A third power device is used to drive the visual module to move horizontally and longitudinally linearly.

4. A construction measurement device according to claim 3, characterized in that: A second sensor is provided on the visual module, and a second distance between the visual module and the target surface is acquired by the second sensor; The measuring device also includes a working control unit connected to the information collection unit, and configured to perform a spacing judgment according to the second distance, and control the traveling device to perform an action according to the spacing judgment result.

5. A measurement method for building construction, characterized in that: The measuring method uses the building construction measuring device according to any one of claims 1 to 4, and the measuring method comprises: Control the device to move to the target initial measurement position by remote control, pushing or remote control; Performing parallel judgment according to the first distance to obtain a parallel judgment result, and if not parallel, controlling the movement of the walking device to adjust the parallelism between the device and the target surface; Specifically, the information collection unit obtains the first distance collected by the first sensors at both ends of the side of the installation platform respectively. l 1 , l 2 Upload to the attitude control unit, the attitude control unit l 1 , l 2 to judge l 1 Is it equal to l 2 If it is not equal, the walking device is controlled to move and the angle is adjusted until l 1 , l 2 equal; If they are parallel, the walking device is controlled to move in a direction parallel to the target surface and walk into the target surface to obtain the target measurement working length; The method is: by light quantity discrimination, when there is a target at a preset distance position of the third sensor, the light quantity is reflected and received, and begins to accumulate until the target is not detected or the fourth sensor detects that the distance between the obstacle and the device reaches a dangerous distance; The target measurement working length is calculated as follows: ； in, L To measure the working length, the unit is m ; V It is the walking speed of the device when measuring the working length of the target, in units of m / s ; T It is the time taken by the device from the initial position to the point where the target cannot be detected or the fourth sensor detects that the distance between the obstacle and the device reaches the dangerous distance, in units of s ; a The dangerous distance between the obstacle and the device. When there is no obstacle, a Take 0, the unit is m ; After the measurement is completed, the device returns to the initial position, and the traveling device drives the visual module to approach the target surface. During this process, the second sensor obtains a second distance between the visual module and the target surface; The working control unit determines the distance according to the second distance, and controls the third power device to stop working if the second distance is equal to the preset distance; Then, the fourth power device is controlled to drive the visual module to move horizontally. After the movement reaches the final stroke, the fifth power device drives the visual module to rise to a preset height, and the fourth power device drives the visual module to move in the opposite direction, and the process is repeated until the limit height or the maximum height of the target surface is reached. During this process, the visual module performs crack measurement on the target surface. The crack measurement work includes: Acquire the target surface image acquired by the vision module; Select the target feature in the target surface image to form a target feature selection area, and extract the target feature edge and crack skeleton line in the target feature selection area; Calculate the skeleton line normal vector according to the crack skeleton line, and calculate the crack width based on the skeleton normal vector and the target feature edge; The step of extracting target feature edges and crack skeleton lines in the target feature selection area includes: Faster R-CNN and YOLO V5(x) target detection network models are used to perform initial identification and extraction of bridge cracks; The target detection network model Faster R-CNN with Resnet50 as the backbone feature extraction network and the target detection network model YOLO V5(x) with CSPDarknet53 as the backbone feature extraction network were used to perform secondary recognition and extraction of bridge cracks under the TensorFlow2 framework; In the PyTorch framework, the U-Net semantic segmentation network model with VGG16 as the backbone feature extraction network is used to segment and extract cracks; The crack edge is extracted by the Alpha Shapes algorithm to form the crack edge line; the crack skeleton line is extracted by iterative calculation of the medial axis transformation; The calculation of the crack width based on the skeleton normal vector and the target feature edge includes: Determine the intersection point between the skeleton normal vector and the target feature edge line; Obtaining the intersection coordinates, calculating the crack width based on the intersection coordinates, and obtaining the crack width calculation result; The calculation of the crack width based on the skeleton normal vector and the target feature edge includes: Determine the intersection point between the skeleton normal vector and the target feature edge line; Obtaining the intersection coordinates, calculating the crack width based on the intersection coordinates, and obtaining the crack width calculation result; The crack width is calculated as follows: ； in, H The crack width of the preset point on the crack skeleton line, in units of mm ; X 1 , X 2 is the horizontal coordinate of the intersection point between the skeleton normal vector and the target feature edge line, Y 1 , Y 2 is the ordinate of the intersection point between the skeleton normal vector and the target feature edge line; The skeleton normal vector is calculated by estimating the normal vector of the skeleton line using SVD singular decomposition, thereby obtaining the skeleton normal vector; In the above measurement process, the information acquisition unit obtains the measured working length, the work control unit calculates the remaining measured working length according to the measured working length, and controls the travel of the walking device and the displacement component based on the remaining measured working length; The remaining measuring working length is calculated as follows: ； in, C is the remaining measuring working length in units of m ; L To measure the working length, the unit is m ; n is the number of times the device walks during the measurement process; X is the length of each walk, in units of m ; When the measurement work of the current area is completed, if C Greater than X , then the walking distance of the walking device of the control device is X Then continue the measurement work, the displacement component drives the visual module to move in the horizontal direction parallel to the target feature X Length; if C Less than X , then the walking distance of the walking device of the control device is X At the same time, during measurement, the displacement component drives the visual module to move in a horizontal direction parallel to the target feature. C If there is an obstacle at the end of the target feature, that is, when the fourth sensor at the end of the walking device detects the obstacle, the working control unit controls the walking device to walk. C Length, control displacement component drives the vision module to move in the horizontal direction parallel to the target feature X - C length, so that it moves to the initial position of the current area ( X - C , 0), and then start measuring. After each rise to the preset height, the horizontal coordinate of the point X - C is the initial or final horizontal coordinate.

Images