Building floor hollowing detection device

Abstract

The present invention discloses a device for detecting hollow drums on building ground, comprising: a driving platform, a knocking unit and a detection unit. The steps of detecting hollow drums include: selecting the knocking force of a pendulum according to the size of a paving board, a sound recorder recording the sound generated by the pendulum knocking the paving board; a processor calculating the probability of hollow drums existing at corresponding positions on the paving board, adjusting the probability of hollow drums based on the flatness of the paving board to obtain a detection probability, and determining whether hollow drums exist at corresponding positions on the paving board according to the detection probability. By adopting the above technical solution, flatness and sound are combined, hollow drum detection is comprehensively performed, the knocking method and sound are further optimized and processed, and the accuracy of the detection result is significantly improved.

Description

Technical Field

[0001] The invention relates to the technical field of construction detection, and in particular to a device for detecting hollowing of building ground. Background Art

[0002] Modern buildings generally use floor tiles, marble, ceramic tiles or wood and other materials to pave the floor. The fit (or bonding) between the floor surface and the next layer of the base surface should be firm and without hollows.

[0003] The existing hollow drum detection scheme is mainly for the inspector to use a small hammer to knock on the ground and make an empirical judgment based on the sound produced. The main problem with the existing scheme is that it judges whether there is a hollow drum based on a single factor of the knocking sound, and generally does not further optimize and process the knocking method and sound. The detection result obtained in this way is poor in accuracy. The reason is that the sound is easily interfered by many external factors during the generation and propagation process, and is also affected by the material of the paving board itself. Summary of the invention

[0004] Purpose of the invention: The present invention provides a device for detecting hollowing on the floor of a building, which aims to conduct comprehensive hollowing detection by introducing the flatness of the paving board and combining it with the sound factor, so as to avoid inaccurate detection results due to external interference in the sound; further, corresponding grid division is performed according to the size of the paving board, and the force of knocking on the grid area is determined according to the size of the paving board. For the recorded sound, the high-frequency part is screened out according to the material of the paving board for detection, so as to improve the accuracy of the detection result.

[0005] Technical solution: The present invention provides a device for detecting hollow drums on the floor of a building, comprising: a driving platform, a knocking unit and a detection unit, wherein: the driving platform is used to drive the detection device to travel on the floor of the building; the knocking unit is arranged on the driving platform, and comprises a driving machine, a connecting rod and a pendulum, and the driving machine drives the pendulum to knock on the floor of the building through the connecting rod; the detection unit comprises a sound recorder, a distance measuring sensor, an image collector and a processor; the steps of hollow drum detection include: the image collector collects the surface image of the paving board that the driving platform is about to travel to, the distance measuring sensor measures multiple points on the surface of the paving board actual distance; using the surface image and the measured actual distance, calculate the plane size of the paving board and the spatial position of the point on the surface of the paving board, and obtain the three-dimensional model of the paving board based on the spatial position fitting of the point; selecting the striking force of the pendulum according to the size of the paving board, and a sound recorder records the sound produced by the pendulum hitting the paving board; the processor calculates the probability of the existence of hollow drums at the corresponding position of the paving board, calculates the flatness of the paving board according to the three-dimensional model, adjusts the probability of hollow drums based on the flatness to obtain the detection probability, and determines whether there are hollow drums at the corresponding position of the paving board according to the detection probability.

[0006] Specifically, the larger the size of the paving board, the smaller the impact force of the pendulum.

[0007] Specifically, the paving board is divided into grids according to the size of the paving board. The larger the size of the paving board is, the larger the corresponding grid size is set, and the pendulum strikes at the center of the grid.

[0008] Specifically, when the pendulum strikes the grid at the edge of the paving board, the striking force is increased.

[0009] Specifically, according to the three-dimensional model of the paving board, the average flatness of each grid area on the paving board is calculated, and the flatness is calculated as the distance between each point on the paving board and the horizontal plane of the paving board.

[0010] Specifically, the corresponding adjustment coefficient is determined according to the average flatness of the grid area, and the probability of hollowing is adjusted using the adjustment coefficient to obtain the detection probability. If the detection probability is higher than the standard threshold, it is determined that there is hollowing in the corresponding grid area; the worse the average flatness, the lower the corresponding adjustment coefficient.

[0011] Specifically, a corresponding high-frequency threshold and sound recognition model are selected according to the material of the paving board, and the processor processes the recorded sound, filters the high-frequency part whose frequency is higher than the high-frequency threshold, and inputs the corresponding sound recognition model to calculate the probability of the existence of hollow drums.

[0012] Specifically, the sound recognition model corresponding to each material is trained using the high-frequency part of the material that is higher than the high-frequency threshold as a sound training set.

[0013] Specifically, the knocking unit also includes a telescopic rod, one end of which is connected to the driving machine, and the other end of which is connected to the connecting rod.

[0014] Specifically, the striking unit further includes a spring connected to the pendulum, and the driving machine drives the pendulum to strike the building ground through a connecting rod and a spring.

[0015] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: it combines flatness and sound, conducts comprehensive hollow drum detection, further optimizes and processes the knocking method and sound, and significantly improves the accuracy of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the steps of hollow drum detection provided by the present invention;

[0017] Figure 2 A schematic diagram of the structure of the hollow drum detection device provided by the present invention;

[0018] 1-travel platform; 2-processor; 3-drive machine; 4-pendulum; 5-spring; 6-connecting rod; 7-telescopic rod; 8-drive machine support; 9-sound recorder; 10-distance measurement sensor; 11-paving board. DETAILED DESCRIPTION

[0019] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0020] See also Figure 1 and Figure 2 , which are respectively a schematic diagram of the steps of hollow drum detection and a schematic diagram of the structure of the hollow drum detection device provided by the present invention.

[0021] The present invention provides a building floor hollowing detection device, comprising: a traveling platform 1, a knocking unit and a detection unit, wherein: the traveling platform 1 is used to drive the detection device to travel on the building floor; the knocking unit is arranged on the traveling platform 1, and comprises a driving machine 3, a connecting rod 6 and a pendulum 4, and the driving machine 3 drives the pendulum 4 to knock the building floor through the connecting rod 6; the detection unit comprises a sound recorder 9, a distance measurement sensor 10, an image collector and a processor 2.

[0022] In a specific implementation, the distance measuring sensor 10 can be a radar or a depth camera, etc., which is used to measure the distance. During the hollow drum detection process, the distance measuring sensor 10 can detect the distance between a certain point on the paving board 11 and the distance measuring sensor 10.

[0023] In the embodiment of the present invention, the striking unit further comprises a telescopic rod 7 , one end of which is connected to the driving machine 3 , and the other end of which is connected to the connecting rod 6 . The driving machine 3 is disposed on a driving machine support 8 .

[0024] In the embodiment of the present invention, the striking unit includes a spring 5 connected to the pendulum 4, and the driving machine 3 drives the pendulum 4 to strike the building ground through the connecting rod 6 and the spring 5.

[0025] In a specific implementation, the travel platform 1 travels by driving wheels through a motor or other driving device.

[0026] In a specific implementation, the driving machine 3 can control the pendulum 4 to strike the paving board at different frequencies and amplitudes. The amplitude and the mass of the pendulum 4 determine the striking force. By setting the spring 5, the force of the pendulum 4 striking the floor can be increased within a limited amplitude stroke, which is suitable for situations where it is necessary to increase the striking force of the pendulum 4.

[0027] In a specific implementation, the detection unit can usually be arranged on the travel platform 1 .

[0028] In hollow drum detection step 1 , the image collector collects a surface image of the paving board 11 to which the driving platform 1 is about to drive, and the distance measurement sensor 10 measures actual distances to multiple points on the surface of the paving board 11 .

[0029] In a specific implementation, the traveling platform 1 drives the hollow drum detection device as a whole to travel on the paving board. The paving board to which the traveling platform 1 is about to travel specifically refers to the paving board to which the traveling platform 1 is about to travel and perform knocking and hollow drum detection.

[0030] In specific implementations, paving boards usually refer to boards made of floor tiles, marble, wood or ceramic tiles, which are laid on the base ground.

[0031] In hollowing detection step 2, the plane size of the paving board and the spatial position of the points on the surface of the paving board are calculated using the surface image and the actual distance measured, and a three-dimensional model of the paving board is obtained by fitting based on the spatial position of the points.

[0032] In a specific implementation, the surface image of the paving board and the points on the surface of the paving board used to measure the distance are selected based on the standard of calculating the spatial position of the points on the surface and fitting the three-dimensional model. In some embodiments, a more accurate three-dimensional spatial position detection device such as radar can be used.

[0033] In a specific implementation, the flatness of each point on the surface of the paving board can be calculated more accurately through the three-dimensional model of the paving board.

[0034] In hollow drum detection step 3, the striking force of the pendulum 4 is selected according to the size of the paving board, and the sound recorder 9 records the sound generated by the pendulum 4 striking the paving board.

[0035] In a specific implementation, the size of the paving board refers to the length and width of the paving board. The size can be calculated based on the three-dimensional model or calculated based on the surface image of the paving board and the distance between the measured points.

[0036] In the embodiment of the present invention, the larger the size of the paving board is, the smaller the striking force of the pendulum 4 is.

[0037] In specific implementation, due to different sizes, the sound of knocking feedback will also be different. Specifically, since each paving board is connected to other paving boards at the edge, the connection can be supported by other paving boards. Therefore, when the paving board size is relatively large, relatively speaking, the force and range of support from other paving boards are smaller, and it is easier to produce sound feedback when knocked. If the knocking force is too large, the sound generated will be too loud, and it will be difficult to perform hollow drum detection. When the paving board size is relatively small, the force and range of support from other paving boards are larger, and it is less likely to produce sound feedback when knocked. Therefore, the knocking force needs to be appropriately increased, and the sound generated will be more reasonable, which makes hollow drum detection easier.

[0038] In specific implementations, when the area of ​​the paving board is less than 0.08 square meters, the knocking force can usually be set to 20 to 40N, when it is 0.08 to 0.1 square meters, the knocking force can usually be set to 40 to 60N, and when it is more than 0.1 square meters, the knocking force can usually be set to more than 60N. Corresponding settings can be made in specific application scenarios.

[0039] In a specific implementation, different knocking forces can be set for different paving board materials. For harder or brittle materials such as marble or tiles, the knocking force can be appropriately reduced according to the above. For softer or tougher materials such as wood, the knocking force can be appropriately increased according to the above.

[0040] In the embodiment of the present invention, the paving board is divided into grids according to the size of the paving board. The larger the size of the paving board is, the larger the corresponding grid size is set, and the pendulum 4 strikes at the center of the grid.

[0041] In the specific implementation, in order to perform hollowing detection reasonably and accurately, the paving board is divided into grids and the center of the grid is tapped. As mentioned above, for larger paving boards, the larger the grid size is set, and for smaller paving boards, the smaller the grid size is set. Usually, the grid size is one eighth to one tenth of the paving board size.

[0042] In the embodiment of the present invention, when the pendulum 4 strikes the grid at the edge of the paving board, the striking force is increased.

[0043] In specific implementation, since the edge of the paving board is supported not only by the base ground but also by other paving boards, it is necessary to increase the knocking force to enhance the sound feedback. The enhanced force varies for different materials. For harder or brittle materials such as bricks, marble or ceramic tiles, the enhanced force is generally lower, about 10N. For softer or tougher materials such as wood, the enhanced force is generally higher, about 20N.

[0044] In an embodiment of the present invention, the average flatness of each grid area on the paving board is calculated based on the three-dimensional model of the paving board, and the flatness is calculated as the distance between each point on the paving board and the horizontal plane of the paving board.

[0045] In a specific implementation, the average flatness of a grid area refers to the average flatness of each point in the grid area. In actual calculations, since it is difficult to actually calculate the flatness of each point, the flatness of multiple points can be uniformly selected in the grid area for calculation.

[0046] In specific implementation, the horizontal plane of the paving board usually refers to the horizontal plane where the surface of the paving board is located. Even if the paving board is set at an inclination, the corresponding horizontal plane can be determined after eliminating the inclination angle. The flatness is calculated based on the distance of the point on the paving board above the horizontal plane of the paving board, that is, the greater the distance, the worse the flatness, and the smaller the distance, the better the flatness, which is reflected in the subsequent adjustment of the adjustment coefficient.

[0047] In hollowing detection step 4, the processor 2 calculates the probability of hollowing at the corresponding position on the paving board, calculates the flatness of the paving board according to the three-dimensional model, adjusts the probability of hollowing based on the flatness to obtain the detection probability, and determines whether hollowing exists at the corresponding position on the paving board according to the detection probability.

[0048] In an embodiment of the present invention, a corresponding adjustment coefficient is determined based on the average flatness of the grid area, and the probability of hollowing is adjusted using the adjustment coefficient to obtain a detection probability. If the detection probability is higher than a standard threshold, it is determined that hollowing exists in the corresponding grid area; the worse the average flatness, the lower the corresponding adjustment coefficient.

[0049] In a specific implementation, the probability of hollow drum is adjusted using the adjustment coefficient to obtain the detection probability, usually by multiplying the calculated probability of hollow drum by the adjustment coefficient. The standard threshold can be set accordingly according to the actual application scenario, usually 80%.

[0050] In an embodiment of the present invention, a corresponding high-frequency threshold and a sound recognition model are selected according to the material of the paving board, and the processor 2 processes the recorded sound, filters the high-frequency part whose frequency is higher than the high-frequency threshold, and inputs the corresponding sound recognition model to calculate the probability of the existence of hollow drums.

[0051] In the embodiment of the present invention, the sound recognition model corresponding to each material is trained using the high-frequency part of the material that is higher than the high-frequency threshold as a sound training set.

[0052] In the specific implementation, since the sound recognition models of different materials are trained separately, the corresponding sound recognition model is selected according to the material, one model corresponds to one sound, and the detection accuracy of the model will be improved accordingly. The training set includes the sounds produced by adjusting multiple factors such as different knocking forces, different knocking positions, different base surfaces, different materials, different hollow drum sizes, and different degrees of fit with the base surface.

[0053] In specific implementation, due to different materials, the feedback sound is obviously different. Generally speaking, for harder or brittle materials such as marble or tiles, the feedback sound frequency is generally higher, and for softer or tougher materials such as wood, the feedback sound frequency is generally lower. Therefore, for harder or brittle materials, the high-frequency threshold is higher, and for softer or tougher materials, the high-frequency threshold is lower. The high-frequency part obtained by high-frequency threshold screening can be more convenient for hollow drum detection.

[0054] In specific implementation, since in the presence of hollowing, the paving boards have no support from the underlying ground and are more likely to vibrate, we can focus on the high-frequency part of the sound. The probability of the existence of hollowing can be calculated from the high-frequency part, and other interference factors in the sound can be eliminated. Especially in indoor situations, the interference of sound is usually medium and low frequency, and the interference can be almost completely eliminated, significantly improving the accuracy of the hollowing detection results.

[0055] It should be noted that in the high-frequency part, when the frequency reaches a certain limit threshold, it indicates that the vibration of the pavement is relatively strong. Especially when the striking force of the pendulum 4 has been reasonably set according to the size and material of the pavement board, it can be directly determined that the corresponding grid has hollows.

Claims

1. A building floor hollowing detection device, It is characterized in that include: Traveling platform, striking unit and detection unit, wherein: The travel platform is used to drive the detection device to travel on the building ground; The knocking unit is arranged on the travel platform and comprises a driving machine, a connecting rod and a pendulum, wherein the driving machine drives the pendulum to knock the building ground through the connecting rod; The detection unit includes a sound recorder, a distance measurement sensor, an image collector and a processor; The steps of hollow drum detection include: The image collector collects the surface image of the paving board to which the driving platform is about to drive, and the distance measurement sensor measures the actual distances to multiple points on the surface of the paving board; Using the surface image and the actual distance obtained by measurement, the plane size of the paving board and the spatial position of the point on the surface of the paving board are calculated, and a three-dimensional model of the paving board is obtained by fitting based on the spatial position of the point; The striking force of the pendulum is selected according to the size of the paving board, and the sound recorder records the sound generated by the pendulum striking the paving board; The processor selects a corresponding high-frequency threshold and a sound recognition model according to the material of the paving board. The processor processes the recorded sound, filters the high-frequency part whose frequency is higher than the high-frequency threshold, and inputs the corresponding sound recognition model to calculate the probability of hollow drums. According to the three-dimensional model of the paving board, the average flatness of each grid area on the paving board is calculated, and the flatness is calculated as the distance between each point on the paving board and the horizontal plane of the paving board. According to the average flatness of the grid area, the corresponding adjustment coefficient is determined, and the adjustment coefficient is used to adjust the probability of hollow drums to obtain the detection probability. If the detection probability is higher than the standard threshold, it is determined that hollow drums exist in the corresponding grid area.

2. The building floor hollowing detection device according to claim 1, It is characterized in that The step of selecting the striking force of the pendulum according to the size of the paving board comprises: The larger the size of the paving board, the less force the pendulum will strike.

3. The building floor hollowing detection device according to claim 1, It is characterized in that The paving board is divided into grids according to the size of the paving board. The larger the size of the paving board, the larger the corresponding grid size is set, and the pendulum strikes at the center of the grid.

4. The building floor hollowing detection device according to claim 3, It is characterized in that When the pendulum strikes the grid at the edge of the paving board, the striking force is increased.

5. The building floor hollowing detection device according to claim 3, It is characterized in that The worse the average flatness, the lower the corresponding adjustment coefficient.

6. The building floor hollowing detection device according to claim 5, It is characterized in that The sound recognition model corresponding to each material is trained using the high-frequency part of the material that is higher than the high-frequency threshold as a sound training set.

7. The building floor hollowing detection device according to claim 1, It is characterized in that The knocking unit also includes a telescopic rod, one end of which is connected to the driving machine, and the other end of which is connected to the connecting rod.

8. The building floor hollowing detection device according to claim 1, It is characterized in that The knocking unit also includes a spring connected to the pendulum, and the driving machine drives the pendulum to knock the building ground through the connecting rod and the spring.

Images