A house fine decoration quality evaluation system and method

Abstract

The application relates to the field of building engineering quality detection, in particular to a house fine decoration quality evaluation system and method, which comprises a ground flatness analysis module, a hollow drum condition judgment module, a hollow drum analysis module, a joint quality analysis module, a closing process evaluation module and a comprehensive quality evaluation module. The system evaluates the comprehensive evaluation index of the house ground engineering according to the flatness of the house ground, the proportion of the hollow drum area to the total area, the joint edge quality of the house ground and the closing quality of the house ground, compares the evaluation grade, and feeds back to the system, so that the limitation of single index evaluation is avoided, the house ground engineering quality is comprehensively and systematically evaluated, and the actual quality level of the engineering can be more truly reflected.

Description

TECHNICAL FIELD

[0001] The application relates to the field of building engineering quality detection, in particular to a house fine decoration quality evaluation system and method. BACKGROUND

[0002] At present, with the rapid development of the real estate industry, fine decoration houses have become the mainstream trend of the market due to their advantages such as convenience, efficiency and controllable cost, and the quality evaluation of fine decoration is very important.

[0003] However, the existing quality detection technology still has defects, for example, the existing Chinese patent with the application number 202311838072.9 discloses a decoration quality detection method and system based on artificial intelligence, which collects wall coating images, carries out noise filtering, feature analysis and calculation, uses a segmentation neural network to distinguish brush marks and normal areas, and determines the wall coating quality according to the area ratio of the two.

[0004] However, the above patent has the following problems: 1. Only the wall coating image is analyzed and detected, and other key parts in the house decoration are not detected, so the overall condition related to the image in the decoration quality cannot be comprehensively reflected.

[0005] 2. The scheme only focuses on pixel processing and brush mark detection of the wall coating image, and does not combine other detection methods for comprehensive quality evaluation in multiple dimensions, so the evaluation of the decoration quality is not comprehensive and in-depth. SUMMARY

[0006] In order to overcome the defects in the background art, the embodiments of the present application provide a house fine decoration quality evaluation system and method, which can effectively solve the problems in the above background art.

[0007] The purpose of the present application can be achieved by the following technical scheme: the present application provides a house fine decoration quality evaluation system, which comprises: a ground flatness analysis module, which obtains three-dimensional coordinate information of each ground point, extracts elevation information, and evaluates the ground flatness of the house.

[0008] A hollow drum condition judgment module is used to collect sound wave signals of hollow drum sample areas and normal sample areas by knocking, and analyze and extract corresponding sound wave attenuation characteristics.

[0009] A hollow drum analysis module is used to construct a discrimination model according to the sound wave attenuation characteristics, screen the hollow drum area, and calculate the proportion of the hollow drum area to the total area by connecting corresponding boundary knocking points.

[0010] A joint quality analysis module is used to extract the joint edge through a threshold algorithm, form a gradient amplitude sequence, count the number of fluctuations to calculate the sawtooth degree of the joint edge, and further evaluate the joint edge quality of the ground of the house.

[0011] The closing process evaluation module is used to generate the actual three-dimensional model of each closing site, and evaluate the closing quality of the house floor by aligning and comparing with the standard model.

[0012] The comprehensive quality evaluation module is used to comprehensively evaluate the comprehensive evaluation index of the house floor engineering, compare the evaluation grade, and feed back to the system.

[0013] Preferably, the specific analysis method of the floor flatness analysis module is: selecting a plurality of floor points according to the set gap, using a laser scanning device to emit a laser beam to each floor point and receive the corresponding reflected signal to obtain the three-dimensional coordinate information of each floor point.

[0014] After preprocessing the three-dimensional coordinate information of each floor point, the elevation of each floor point is extracted, and a reference plane is determined, and the deviation between the elevation of each floor point and the reference plane is calculated.

[0015] The flatness of the house floor is obtained by calculating the mean square deviation between the elevation of each floor point and the reference plane.

[0016] Preferably, the specific operation method of the hollow condition judgment module is: A1. Selecting the existing floor area with hollow condition and the floor area meeting the standard specification as the hollow sample area and the normal sample area, respectively.

[0017] A2. The hollow sample area and the normal sample area are divided into a plurality of equal-area detection areas, and are sequentially numbered, and a plurality of knocking points are uniformly arranged in each detection area according to the set rule, and a selected knocking tool is used to knock each knocking point with a fixed force, and a sound wave signal acquisition device is started to record the sound wave signal generated by each knocking.

[0018] A3. The original sound wave signal collected is subjected to noise reduction processing, and on the time domain waveform of the sound wave signal after noise reduction processing, the knocking starting time and the signal duration time are determined for the signals of the hollow sample area and the normal sample area, respectively, the amplitude value of the sound wave signal in a specific time period after the knocking starting time is calculated, the initial amplitude information is obtained, and the time experienced by the signal amplitude from the initial value to a specific proportion is recorded, and the sound wave attenuation speed is calculated.

[0019] A4. Using Fourier transform algorithm, time domain sound wave signals from the hollow sample area and the normal sample area are converted into frequency domain signals respectively to generate respective frequency spectrum graphs. In the frequency spectrum graphs, by comparing the differences in the distribution of signal energy in different frequency bands in the frequency spectrum graphs of the hollow sample area and the normal sample area, a characteristic frequency range related to the hollow is identified. In the characteristic frequency range, initial amplitude information and sound wave attenuation speed of the signals of the hollow sample area and the normal sample area are extracted respectively.

[0020] A5. Integrating sound wave attenuation characteristic data obtained from time domain analysis of the hollow sample area and the normal sample area, sound wave attenuation characteristics of the hollow area and the normal area are constituted by comparing characteristic differences.

[0021] Preferably, the specific analysis method of the hollow analysis module is to construct a discrimination model according to the sound wave attenuation characteristics of the hollow area and the normal area.

[0022] The floor surface to be detected is divided into several equal-area floor areas, and several knocking points are uniformly arranged. Sound wave attenuation characteristics of each knocking point of each floor area are obtained and input into the discrimination model, which judges whether each floor area belongs to the hollow area.

[0023] Each knocking point determined as the hollow area is recorded as a boundary knocking point of the hollow area. The area of the hollow area is calculated by connecting the boundary knocking points of the hollow area to form a polygon, and then compared with the total area of the floor surface to obtain the proportion of the area of the hollow area in the total area.

[0024] Preferably, the setting rule is to draw an equal-area grid on the floor surface, take each grid intersection as a knocking point, and assign a unique coordinate number to each knocking point. When collecting the sound wave signal generated by each knocking, the coordinate number of the corresponding knocking point is recorded synchronously to establish a preliminary association between the sound wave signal and the position of the knocking point.

[0025] Once the discrimination model determines that a certain sound wave signal is from the hollow area, the position of the knocking point corresponding to the sound wave signal is determined according to the preliminary association between the sound wave signal and the position of the knocking point.

[0026] Preferably, the specific analysis method of the joint quality analysis module is to divide the joint area of the floor surface to be detected, use an industrial camera to obtain images of each joint area, obtain gradient amplitude values of each pixel point in the images of each joint area through grayscale processing and gradient calculation of the images of each joint area.

[0027] A gradient amplitude threshold is set, and each pixel point of each joint region image is traversed, if the gradient amplitude of a pixel point is greater than or equal to the gradient amplitude threshold, the pixel point is determined as an edge point, if the gradient amplitude of a pixel point is less than the gradient amplitude threshold, the pixel point is determined as a non-edge point, and the joint edge of each joint region image is obtained by connecting each edge point.

[0028] For each joint region image, the gradient amplitudes of the contour points and the pixel points in a certain neighborhood in the edge direction are extracted along the contour in sequence according to the direction of the joint edge, the extracted gradient amplitudes are arranged in the order of the contour points, and a sequence representing the change of the gradient amplitudes along the joint edge direction is formed.

[0029] A lower threshold and an upper threshold are set respectively, the change of the gradient amplitude in this range is determined as an effective fluctuation, the gradient amplitude sequence along the joint edge direction is traversed, the change between adjacent gradient amplitudes is checked in sequence from the starting point of the sequence, when the gradient amplitude changes from less than the lower threshold to greater than the lower threshold or from greater than the upper threshold to less than the upper threshold, it is considered that an effective fluctuation occurs, the count is increased by 1, and the fluctuation number of the joint edge under the current set threshold is obtained.

[0030] The fluctuation frequency per unit length is obtained by dividing the fluctuation number by the number of contour points, and the sawtooth degree of the joint edge is calculated based on the fluctuation number and the fluctuation frequency.

[0031] The sawtooth degrees of the joint edges of each joint region image are calculated by this method, and the joint edge quality of the house floor is obtained by taking the average.

[0032] Preferably, the specific analysis method of the closing process evaluation module is: dividing the closing region of the house floor to be detected, using a three-dimensional scanner to scan each closing part of the house floor, acquiring the angle point cloud data of each closing part of the house floor, reconstructing the surface according to the angle point cloud data of each closing part of the house floor, generating the actual three-dimensional model of each closing part of the house floor, aligning the actual three-dimensional model of each closing part of the house floor with the designed standard three-dimensional model in space position and direction, and overlapping by the iterative closest point algorithm.

[0033] The distance deviation of each group of corresponding points between the actual model and the standard model is calculated by comparing the overlapped three-dimensional models, and the closing quality of the house floor is evaluated.

[0034] Preferably, the specific analysis method of each type of deviation value of each corresponding position between the actual model and the standard model is as follows: a representative local area that can accurately reflect the local characteristics of the model is selected on the standard model, a center point of the local area is selected, and the local area is defined as a template; the defined template is searched and matched on the actual model, the most similar area to the template is found through a gray-based template matching algorithm, and a center point of the area is obtained; and each center point determined on the standard model and the actual model is corresponded one by one to form each group of corresponding points.

[0035] Preferably, the specific analysis method of the comprehensive quality evaluation module is as follows:

[0036] The flatness of the house floor, the proportion of the hollow area to the total area, the joint edge quality of the house floor, and the closing quality of the house floor are obtained respectively, weights are set for them in sequence, the products of each item and the corresponding weight are multiplied, and then the products are added to obtain the comprehensive evaluation index of the house floor engineering.

[0037] The threshold values of each evaluation grade are set, the comprehensive evaluation index of the house floor engineering is compared with the evaluation index range corresponding to each evaluation grade threshold value to obtain the evaluation grade corresponding to the comprehensive evaluation index of the house floor engineering, and the evaluation grade is fed back to the system.

[0038] Preferably, the present application provides a house fine decoration quality evaluation method, and the specific steps of the evaluation method are as follows: S1. three-dimensional coordinate information of each floor point is obtained, elevation information is extracted, and the flatness of the house floor is evaluated.

[0039] S2. sound wave signals of hollow drum sample areas and normal sample areas are collected by knocking, and corresponding sound wave attenuation characteristics are analyzed and extracted.

[0040] S3. a discrimination model is constructed according to the sound wave attenuation characteristics, hollow drum areas are screened, and the proportion of the area of the hollow drum area to the total area is calculated by connecting corresponding boundary knocking points.

[0041] S4. a threshold algorithm is used to extract a joint edge to form a gradient amplitude sequence, the sawtooth degree of the joint edge is calculated by counting the number of fluctuations, and then the joint edge quality of the house floor is evaluated.

[0042] S5. actual three-dimensional models of each closing part are generated, the closing quality of the house floor is evaluated by aligning and comparing with a standard model.

[0043] S6. the comprehensive evaluation index of the house floor engineering is comprehensively evaluated, the evaluation grade is obtained by comparison, and the evaluation grade is fed back to the system.

[0044] Compared with the prior art, the embodiments of the present application have at least the following advantages or beneficial effects: first, the present application can timely find the problems of unevenness, local protrusion or depression of the ground by obtaining the three-dimensional coordinate information of each ground point, extracting the elevation information and evaluating the flatness of the ground of the house.

[0045] Second, the present application can screen the hollow drum area by extracting the acoustic wave attenuation characteristics of the hollow drum sample area and the normal sample area, constructing a discrimination model, and connecting the corresponding boundary knocking points to calculate the proportion of the hollow drum area in the total area. This precise positioning method can clearly determine the specific position and shape of the hollow drum and intuitively reflect the severity of the hollow drum of the ground of the house.

[0046] Third, the present application can clearly understand the problems existing in the quality of the joint edge by extracting the joint edge through the threshold algorithm, constructing a gradient amplitude sequence, calculating the sawtooth degree of the joint edge by counting the fluctuation times, and evaluating the quality of the joint edge of the ground of the house, thereby guiding the improvement of the construction process.

[0047] Fourth, the present application can accurately find the difference between the actual closing-in part and the design requirement by generating the actual three-dimensional model of each closing-in part, aligning and comparing with the standard model, and evaluating the closing-in quality of the ground of the house.

[0048] Fifth, the present application can avoid the limitation of single index evaluation by comprehensively evaluating the comprehensive evaluation index of the house ground engineering, comparing to obtain the evaluation grade, and feeding back to the system, and can comprehensively and systematically evaluate the quality of the house ground engineering, which can more truly reflect the actual quality level of the engineering. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0050] Figure 1 It is a module connection diagram of a house fine decoration quality evaluation system.

[0051] Figure 2 It is Figure 1 Flow chart of hollow drum condition judgment module.

[0052] Figure 3 It is a flow chart of a house fine decoration quality evaluation method. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort belong to the scope of the present application.

[0054] Please refer to Figure 1 As shown in the figure, a house fine decoration quality evaluation system, the system includes ground flatness analysis module, hollow drum condition judgment module, hollow drum analysis module, joint quality analysis module, closing process evaluation module, comprehensive quality evaluation module.

[0055] The comprehensive quality evaluation module and ground flatness analysis module, hollow drum condition judgment module, hollow drum analysis module, joint quality analysis module, closing process evaluation module are connected, and the hollow drum condition judgment module and the hollow drum analysis module are connected.

[0056] The ground flatness analysis module extracts the elevation information by obtaining the three-dimensional coordinate information of each ground point, and evaluates the ground flatness of the house.

[0057] The specific analysis method of the ground flatness analysis module is: a plurality of ground points are selected according to a set gap, a laser scanning device is used to emit laser beams to each ground point and receive corresponding reflection signals to obtain the three-dimensional coordinate information of each ground point; the elevation data can be accurately extracted, which provides a reliable basis for the evaluation of ground flatness, and avoids the misjudgment of ground flatness caused by measurement error.

[0058] After preprocessing the three-dimensional coordinate information of each ground point, the elevation of each ground point is extracted, and a reference plane is determined, and the deviation between the elevation of each ground point and the reference plane is calculated.

[0059] The flatness of the house ground is obtained by calculating the mean square deviation of the deviation between the elevation of each ground point and the reference plane; accurate ground flatness data can help understand the ground condition, and in subsequent ground construction, such as laying tiles, pouring floor, etc. According to the flatness condition, corresponding measures can be taken to ensure the construction quality and reduce the cost of rework and repair.

[0060] It should be noted that the elevation of each ground point is the Z coordinate value extracted from the three-dimensional coordinate information of each ground point after preprocessing, and for each ground point, the actual elevation is compared with the elevation of the reference plane, and the deviation value between the elevation of the point and the reference plane is calculated by subtraction operation, and then the flatness of the house ground is obtained by calculating the mean square deviation.

[0061] The hollow drum condition judgment module is used for collecting sound wave signals of the hollow drum sample area and the normal sample area through knocking, and analyzing and extracting corresponding sound wave attenuation characteristics.

[0062] Please refer to Figure 2 As shown in the figure, the specific operation method of the hollow drum condition judgment module is: A1. Select the existing ground area with hollow drum condition and the ground area meeting the standard specification as the hollow drum sample area and the normal sample area respectively; By comparing the ground areas in different states, the characteristics related to hollow drum can be found more accurately, providing a reference for identifying hollow drum.

[0063] A2. The hollow drum sample area and the normal sample area are divided into several equal-area detection areas and numbered in turn, and the knocking points are uniformly set in each detection area according to the set rule. The selected knocking tool is used to knock the knocking points in turn with fixed force, and the sound wave signal acquisition device is started to record the sound wave signals generated by each knocking; Starting the sound wave signal acquisition device to record the sound wave signals generated by each knocking can completely capture the sound wave changes when knocking the ground. These sound wave signals contain rich ground structure information, which provides original data support for subsequent analysis of whether the ground has hollow drum and the characteristics of the hollow drum.

[0064] A3. The collected original sound wave signals are subjected to noise reduction processing. On the time domain waveform of the sound wave signals after noise reduction processing, the knocking starting time and signal duration time are determined for the signals of the hollow drum sample area and the normal sample area respectively. The amplitude value of the sound wave signal in a specific time period after the knocking starting time is calculated to obtain the initial amplitude information, and the time experienced by the signal amplitude from the initial value to a certain proportion is recorded to calculate the sound wave attenuation speed; It is helpful to more accurately analyze the characteristics of the sound wave signal, avoid the influence of noise on the subsequent analysis results, and make the extracted characteristic information more truly reflect the actual situation of the ground.

[0065] It should be noted that the specific analysis method of the signal duration time is: setting the upper and lower amplitude threshold values, when the amplitude of the sound wave signal after noise reduction exceeds the upper amplitude threshold value, it is considered that the knocking starts, and the signal amplitude is continuously monitored from the knocking starting time, when the signal amplitude decreases to the lower amplitude threshold value and remains for a fixed time, it is considered that the signal ends.

[0066] The specific analysis method of the initial amplitude information is: in the determined specific time period, the average value of the signal amplitude in the time period is calculated as the initial amplitude information.

[0067] The specific analysis method of the sound wave attenuation speed is: setting a specific proportion of signal amplitude attenuation, starting from the initial amplitude, monitoring the change of signal amplitude, when the signal amplitude attenuates to a specific proportion of the initial amplitude, recording the time experienced, calculating the sound wave attenuation speed according to the recorded attenuation time and initial amplitude information, and the calculation formula of the attenuation speed is: attenuation speed = (initial amplitude-attenuated amplitude) / attenuation time.

[0068] A4. Using the Fourier transform algorithm, the time-domain sound wave signals from the hollow sample area and the normal sample area are converted into frequency-domain signals respectively to generate respective frequency spectrum graphs. In the frequency spectrum graphs, by comparing the differences in the distribution of signal energy in different frequency bands in the frequency spectrum graphs of the hollow sample area and the normal sample area, the characteristic frequency range related to the hollow is identified. In the characteristic frequency range, the initial amplitude information and the sound wave attenuation speed of the signals of the hollow sample area and the normal sample area are extracted respectively. The characteristic information of different frequency bands may be more sensitive to the hollow. By extracting these characteristics in a specific frequency range, the differences between the hollow area and the normal area can be more accurately captured, and the accuracy and reliability of the hollow identification are improved.

[0069] It should be noted that the specific analysis method of the characteristic frequency range identification is: dividing the frequency range of the frequency spectrum graph into several frequency bands, calculating the signal energy of the frequency spectrum graph of the hollow sample area and the normal sample area in each frequency band by summing the square of the signal amplitude of each frequency point in the frequency band, comparing the energy distribution of the two sample areas in each frequency band, and finding the frequency band with significant energy difference. The frequency band with significant energy difference is determined as the characteristic frequency range related to the hollow.

[0070] The specific analysis method of extracting the initial amplitude information and the sound wave attenuation speed of the signals of the hollow sample area and the normal sample area is: in the determined characteristic frequency range, for the frequency-domain signals of the hollow sample area and the normal sample area, finding the frequency point with the maximum amplitude, taking the amplitude value of the frequency point as the initial amplitude information in the frequency domain, recording the corresponding frequency change when the frequency-domain signal amplitude attenuates to a specific proportion from the initial amplitude value, and combining the corresponding relationship between frequency and time to calculate the sound wave attenuation speed in the characteristic frequency range.

[0071] A5. Integrating the sound wave attenuation characteristic data obtained from the time-domain analysis of the hollow sample area and the normal sample area, and constructing the sound wave attenuation characteristics of the hollow area and the normal area by comparing the characteristic differences; the sound wave characteristic rules for distinguishing the hollow area and the normal area can be comprehensively and systematically summarized, which can be used as a basis for judging whether there is a hollow on the ground, and provides a scientific and effective method for hollow detection in actual engineering.

[0072] It should be noted that the feature difference includes the difference of initial amplitude and the ratio of attenuation speed, and the sound wave attenuation feature capable of distinguishing the hollow area and the normal area is constructed according to the obtained feature difference.

[0073] The hollow analysis module is configured to construct a discrimination model according to the sound wave attenuation feature, screen the hollow area, and calculate the proportion of the area of the hollow area to the total area by connecting the corresponding boundary knocking points.

[0074] The specific analysis method of the hollow analysis module is to construct a discrimination model according to the sound wave attenuation feature of the hollow area and the normal area.

[0075] It should be noted that the specific construction method of the discrimination model is to divide the pre-processed sound wave attenuation features of the hollow area and the normal area into a training set and a test set according to a certain proportion, input the divided training set data into the model, and the model learns according to the input feature data and the corresponding label (hollow area or normal area). The model predicts the training data according to the current parameters, compares it with the real label, calculates the value of the loss function, adjusts the parameters of the model by the stochastic gradient descent method, so that the value of the loss function gradually decreases. In each iteration, the algorithm updates the parameters according to the gradient of the loss function to adjust the model performance in the direction of better performance. The above input data, loss function calculation and parameter adjustment process is repeated until the loss function converges to a smaller value.

[0076] The ground of the house to be detected is divided into several equal-area ground areas, and several knocking points are uniformly set. The sound wave attenuation features of the knocking points of each ground area are obtained and input into the discrimination model, which judges whether each ground area belongs to the hollow area. By dividing the equal-area areas and uniformly setting the knocking points, the entire house ground can be covered comprehensively and systematically, avoiding omission of some areas, and more accurately capturing the sound wave attenuation features of different positions of the ground. Because the hollow conditions of different positions may be different, uniformly setting the knocking points helps to detect potential hollow areas more carefully and reduce the possibility of misjudgment and omission.

[0077] The knocking points determined as the hollow area are recorded as the boundary knocking points of the hollow area, and the area of the hollow area is calculated by connecting the boundary knocking points of the hollow area to form a polygon. Then, the proportion of the area of the hollow area to the total area is obtained by comparing the total area of the house ground. Calculating the proportion of the area of the hollow area to the total area can quantitatively evaluate the hollow degree of the house ground, which has important guiding significance for judging the quality of the house ground and whether further repair measures need to be taken.

[0078] It should be noted that the plane coordinates of the knocking points of the boundaries of the hollow area are obtained, the area of the hollow area polygon is calculated by using the polygon area calculation formula, the total ground area data of the building is obtained from the building drawing, the total ground area of the building is extracted, the hollow area is divided by the total ground area of the building, and then multiplied by 100%, to obtain the proportion of the hollow area to the total ground area of the building.

[0079] The setting rule is to draw an equal-area grid on the ground, take each grid intersection as a knocking point, and assign a unique coordinate number to each knocking point. When collecting the sound wave signal generated by each knocking, the coordinate number of the corresponding knocking point is recorded synchronously, so as to establish a preliminary association between the sound wave signal and the position of the knocking point. Once the discrimination model determines that a certain sound wave signal comes from the hollow area, the position of the knocking point corresponding to the sound wave signal is determined according to the preliminary association between the sound wave signal and the position of the knocking point. This helps to accurately locate the hollow area and facilitates subsequent maintenance and other work.

[0080] The joint quality analysis module is used to extract the joint edge by threshold algorithm, construct a gradient amplitude sequence, count the number of fluctuations to calculate the sawtooth degree of the joint edge, and further evaluate the joint edge quality of the building ground.

[0081] The specific analysis method of the joint quality analysis module is: dividing the joint area of the building ground to be detected, obtaining the image of each joint area by using an industrial camera, obtaining each joint area image after gray scale processing, calculating the gradient of each joint area image to obtain the gradient amplitude of each pixel point in each joint area image; the gradient amplitude of the pixel point at the joint edge usually changes obviously, and by calculating the gradient amplitude, the joint edge can be more clearly identified, which provides a basis for subsequent operations such as extracting the joint edge, constructing the gradient amplitude sequence, and counting the number of fluctuations to calculate the sawtooth degree of the joint edge, so as to accurately evaluate the joint edge quality of the building ground.

[0082] It should be noted that the gradient calculation adopts Sobel operator, and two 3x3 convolution kernels are used to perform convolution operation on each joint area image in horizontal and vertical directions respectively to obtain the gradient approximation value of each joint area image in horizontal and vertical directions The gradient amplitude is calculated by the formula , wherein represents the number of the th joint area image, .

[0083] A gradient amplitude threshold is set, and each pixel point of each joint region image is traversed. If the gradient amplitude of a certain pixel point is greater than or equal to the gradient amplitude threshold, the pixel point is determined to be an edge point. If the gradient amplitude of a certain pixel point is less than the gradient amplitude threshold, the pixel point is determined to be a non-edge point. The joint edge of each joint region image is obtained by connecting each edge point. In the actual image acquisition process, noise may be generated due to various factors. The noise may cause abnormal fluctuations in the gradient amplitude of some pixel points. By setting a suitable threshold, the noise points can be avoided from being misjudged as edge points, the stability and reliability of edge detection are improved, and the detection result can maintain good consistency under different image conditions.

[0084] For each joint region image, the gradient amplitudes of the contour points and the pixel points in a certain neighborhood in the edge direction are extracted along the contour in sequence according to the direction of the joint edge. The extracted gradient amplitudes are arranged in the order of the contour points to form a sequence representing the change of the gradient amplitude along the joint edge direction. The characteristics of the joint edge can be comprehensively described. Not only the pixel point information on the joint edge is considered, but also the gradient amplitude change of the pixel points in the neighborhood. In this way, the local characteristics and change trend of the joint edge can be more carefully reflected.

[0085] It should be noted that for the extracted joint edge contour, the order of the contour points is usually arranged in sequence according to the clockwise or counterclockwise direction. For each contour point, all pixel points in the neighborhood in the edge direction are found in the image according to the defined neighborhood range. The gradient amplitudes of the corresponding positions in the gradient amplitude image are obtained by traversing the pixel coordinates in the neighborhood.

[0086] A lower threshold and an upper threshold are set respectively. The change of the gradient amplitude in this range is determined to be an effective fluctuation. The gradient amplitude sequence along the joint edge direction is traversed. Starting from the starting point of the sequence, the change between adjacent gradient amplitudes is checked in sequence. When the gradient amplitude changes from less than the lower threshold to greater than the lower threshold or from greater than the upper threshold to less than the upper threshold, it is considered that an effective fluctuation occurs, and the count is increased by 1. The fluctuation number of the joint edge under the current set threshold is obtained. In this way, the misjudgment of subtle and unimportant fluctuations is avoided, and the actual change of the joint edge is more accurately reflected.

[0087] The frequency of the wave is obtained by dividing the number of waves by the number of profile points, and the sawtooth degree of the joint edge is calculated based on the number of waves and the frequency of the wave; the sawtooth degree is calculated by comprehensively considering the number of waves and the frequency of the wave, so that the quality condition of the joint edge can be more comprehensively reflected, the number of waves reflects the frequency of significant changes on the joint edge, and the frequency of the wave considers the distribution of the changes in unit length, and the combination of the two can more specifically describe the characteristics of the joint edge, so that more rich and comprehensive information is provided for judging the quality of the joint edge.

[0088] In a preferred embodiment of the present application, the specific analysis method of the sawtooth degree of the joint edge is as follows: the number of waves and the frequency of the wave are extracted, multiplied by the corresponding weights, and then the products are added to obtain the sawtooth degree of the joint edge.

[0089] For example, the weights corresponding to the number of waves and the frequency of the wave are .

[0090] The sawtooth degree of the joint edge of each joint area image is calculated by this method, and the quality of the joint edge of the house floor is obtained by taking the average value; the average value as a unified index facilitates the comparison of the quality of the joint edge between different houses or different parts of the same house, whether it is the quality monitoring of different batches of houses in the construction process or the regular inspection of the joint condition of different areas in the use process of the house, the comparison of the average value can quickly find the change trend and difference of the quality, and timely measures are taken for processing.

[0091] The closing process evaluation module is used to generate the actual three-dimensional model of each closing part, and the closing quality of the house floor is evaluated by aligning and comparing with the standard model.

[0092] The specific analysis method of the closing process evaluation module is as follows: the closing area of the house floor to be detected is divided, the three-dimensional scanner is used to scan each closing part of the house floor, the angle point cloud data of each closing part of the house floor is acquired, the surface reconstruction is performed according to the angle point cloud data of each closing part of the house floor, the actual three-dimensional model of each closing part of the house floor is generated, the actual three-dimensional model of each closing part of the house floor is aligned with the designed standard three-dimensional model in space position and direction, and the coincidence is performed through the iterative closest point algorithm; this comparison can quantitatively show the deviation of the closing part in size, shape and position, help to quickly locate the problem area, and clearly indicate the place that needs to be improved or adjusted, so as to provide an objective and accurate basis for the evaluation of the closing process, and help to improve the quality and compliance of the closing process.

[0093] It should be noted that the iterative closest point algorithm is used for accurate alignment, and the corresponding point pairs between the actual model and the standard model are found through continuous iteration, and the transformation parameters (translation, rotation and scaling) of the model are calculated according to these corresponding point pairs, so that the two models gradually coincide. In each iteration, the nearest point of the point on the actual model to the standard model is calculated, and the transformation parameters of the model are updated according to these nearest points until the set convergence condition is met.

[0094] The distance deviation of each group of corresponding points of the actual model and the standard model is calculated by comparing the superimposed three-dimensional models, and the closing quality of the house floor is evaluated.

[0095] It should be noted that the calculation formula of the distance deviation is , wherein The coordinate difference of two points in the x-axis, y-axis and z-axis directions is respectively represented.

[0096] The specific analysis method of each type of deviation value of the actual model and the standard model at each corresponding position is: selecting a representative local area on the standard model which can accurately reflect the local characteristics of the model, selecting the center point of the local area, and defining the local area as a template, searching and matching the defined template on the actual model, finding the most similar area to the template based on the gray-scale template matching algorithm, and obtaining the center point of the area. The center points determined on the standard model and the actual model are one-to-one corresponding, forming each group of corresponding points. This matching method is based on the gray-scale information of the model and does not depend on the specific geometric shape or other complex features of the model, which has strong robustness and adaptability, can effectively complete the matching task under different scanning conditions and model accuracy, and ensures the accuracy of the corresponding relationship.

[0097] It should be noted that for the local area of regular shape (such as rectangle, circle, etc.), the coordinates of the center point are directly calculated according to its geometric properties, and for the local area of irregular shape, the approximate position of the center point is obtained by calculating the average value of the coordinates of all points in the area.

[0098] The comprehensive quality evaluation module is used for comprehensive evaluation of the comprehensive evaluation index of the house floor engineering, comparison of the evaluation grade, and feedback to the system.

[0099] The specific analysis method of the comprehensive quality evaluation module is: the flatness of the house ground, the proportion of the hollow area to the total area, the joint edge quality of the house ground, and the closing quality of the house ground are obtained respectively, weights are set for them in turn, the products are multiplied by the corresponding weights, and then the products are added to obtain the comprehensive evaluation index of the house ground engineering; the indexes are multiplied by the corresponding weights and then added to obtain the comprehensive evaluation index, the quality of the house ground engineering is converted into a specific numerical value, and this quantitative evaluation method makes the quality evaluation more objective, accurate and comparable.

[0100] In a preferred embodiment of the present application, the specific analysis method of the comprehensive evaluation index of the house ground engineering is as follows: the flatness of the house ground corresponding to the house ground engineering, the proportion of the hollow area to the total area, the joint edge quality of the house ground, and the closing quality of the house ground are extracted, multiplied by the corresponding weights, and then the products are added to obtain the comprehensive evaluation index of the house ground engineering.

[0101] For example, the weights corresponding to the flatness of the house ground, the proportion of the hollow area to the total area, the joint edge quality of the house ground, and the closing quality of the house ground are .

[0102] The evaluation index corresponding to the comprehensive evaluation index of the house ground engineering is obtained by comparing the comprehensive evaluation index of the house ground engineering with the evaluation index range corresponding to each evaluation grade threshold value, and the evaluation grade corresponding to the comprehensive evaluation index of the house ground engineering is fed back to the system; clear grade division can quickly understand the level of the engineering quality, and measures can be taken in time to rectify the engineering with a lower quality grade, and the engineering with a higher quality grade can be summarized for promotion, which is helpful to improve the overall engineering quality level.

[0103] Please refer to Figure 3 In addition, the present application provides a house fine decoration quality evaluation method, and the specific steps of the method are as follows: S1. Obtain the three-dimensional coordinate information of each ground point, extract the elevation information, and evaluate the flatness of the house ground.

[0104] S2. Collect the sound wave signals of the hollow sample area and the normal sample area by knocking, analyze and extract the corresponding sound wave attenuation characteristics.

[0105] S3. Construct a discrimination model according to the sound wave attenuation characteristics, screen the hollow area, and calculate the proportion of the hollow area to the total area by connecting the corresponding boundary knocking points.

[0106] S4. Extract the joint edge by threshold algorithm, form a gradient amplitude sequence, calculate the sawtooth degree of the joint edge by counting the fluctuation times, and then evaluate the joint edge quality of the house ground.

[0107] S5. Generating actual three-dimensional model of each joint site, evaluating the quality of the joint of the house floor by comparing with the standard model.

[0108] S6. Comprehensive evaluation index of the house floor engineering is evaluated, the evaluation grade is obtained by comparison, and the system is fed back.

[0109] Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and modifications to the above-mentioned embodiments within the scope of the present application, which are still covered by the protection scope of the present application.

Claims

1. A quality evaluation system for interior decoration of houses, characterized in that, The system specifically comprises the following modules: The ground flatness analysis module extracts elevation information by obtaining three-dimensional coordinate information of each ground point, and evaluates the flatness of the house ground; The hollow condition judgment module is used for collecting sound wave signals of hollow sample areas and normal sample areas through knocking, and analyzing and extracting corresponding sound wave attenuation characteristics; The hollow analysis module is used for constructing a discrimination model according to the sound wave attenuation characteristics, screening hollow areas, and calculating the proportion of the hollow area to the total area by connecting corresponding boundary knocking points; The joint quality analysis module is used for extracting joint edges by threshold algorithm, forming a gradient amplitude sequence, and calculating the sawtooth degree of the joint edge by counting the fluctuation frequency, and then evaluating the joint edge quality of the house ground; The closing process evaluation module is used for generating an actual three-dimensional model of each closing part, and evaluating the closing quality of the house ground by aligning and comparing with the standard model; The comprehensive quality evaluation module is used for comprehensively evaluating the comprehensive evaluation index of the house ground engineering, comparing to obtain the evaluation grade, and feeding back to the system; The specific analysis method of the joint quality analysis module is as follows: The joint area of the house ground to be detected is divided, and the image of each joint area is obtained by using an industrial camera, and the joint edge of each joint area image is analyzed after gray processing. For each joint area image, the gradient amplitude of the pixel points in a certain neighborhood in the edge direction along the contour is extracted in sequence according to the direction of the joint edge, the extracted gradient amplitude is arranged in the order of the contour points, and a sequence representing the change of the gradient amplitude along the joint edge direction is formed. The lower threshold and the upper threshold are set respectively, the change of the gradient amplitude in this range is determined as an effective fluctuation, the gradient amplitude sequence along the joint edge direction is traversed, the change between adjacent gradient amplitudes is checked in sequence from the starting point of the sequence, when the gradient amplitude changes from less than the lower threshold to more than the lower threshold or from more than the upper threshold to less than the upper threshold, it is considered that an effective fluctuation occurs, the count is added by 1, and the fluctuation frequency per unit length is obtained by dividing the fluctuation frequency by the number of contour points. The sawtooth degree of the joint edge is calculated based on the fluctuation frequency and the fluctuation frequency. The sawtooth degree of the joint edge of each joint area image is calculated by this method, and the joint edge quality of the house ground is obtained by taking the average. The specific operation method of the hollow drum condition judgment module is: A1. Selecting the existing ground area with hollow drum condition and the ground area meeting the standard specification as the hollow drum sample area and the normal sample area respectively; A2. Dividing the hollow drum sample area and the normal sample area into several equal-area detection areas, and numbering them in turn, setting the knocking points in each detection area according to the set rule, and knocking the knocking points with the selected knocking tool with fixed force, while starting the sound wave signal collection device to record the sound wave signals generated by each knocking; A3. The original sound wave signals collected are subjected to noise reduction processing, and on the time domain waveform of the sound wave signals after noise reduction processing, the signal duration time and the knocking starting time are determined for the signals of the hollow drum sample area and the normal sample area respectively, the amplitude value of the sound wave signal in a specific time period after the knocking starting time is calculated to obtain the initial amplitude information, and the time experienced by the signal amplitude from the initial value to a certain proportion is recorded to calculate the sound wave attenuation speed; A4. Using Fourier transform algorithm, the time domain sound wave signals from the hollow drum sample area and the normal sample area are converted into frequency domain signals respectively to generate their respective frequency spectrum graphs, and in the frequency spectrum graphs, the characteristic frequency range related to the hollow drum is identified by comparing the differences in the distribution of signal energy in different frequency bands in the frequency spectrum graphs of the hollow drum sample area and the normal sample area, and the initial amplitude information and the sound wave attenuation speed of the signals of the hollow drum sample area and the normal sample area in the characteristic frequency range are extracted respectively; A5. Integrating the sound wave attenuation characteristic data obtained from the time domain analysis of the hollow drum sample area and the normal sample area, the sound wave attenuation characteristics of the hollow drum area and the normal area are constituted by comparing the characteristic differences.

2. The housing finishing quality evaluation system according to claim 1, characterized in that: The specific analysis method of the ground flatness analysis module is: Selecting several ground points according to the set gap, emitting laser beams to each ground point by using the laser scanning device and receiving the corresponding reflection signals to obtain the three-dimensional coordinate information of each ground point; After preprocessing the three-dimensional coordinate information of each ground point, the elevation of each ground point is extracted, and the reference plane is determined, and the deviation between the elevation of each ground point and the reference plane is calculated respectively; The flatness of the ground of the house is obtained by calculating the mean square deviation of the deviation between the elevations of each ground point and the reference plane.

3. The housing finishing quality evaluation system according to claim 1, characterized in that: The specific analysis method of the hollow drum analysis module is: Constructing a discrimination model according to the sound wave attenuation characteristics of the hollow drum area and the normal area; Dividing the house ground to be detected into several equal-area ground areas, and uniformly setting several knocking points, obtaining the sound wave attenuation characteristics of each knocking point of each ground area, and inputting them into the discrimination model, the discrimination model judges whether each ground area belongs to the hollow drum area; The knocking points determined as the hollow drum area are recorded as the boundary knocking points of the hollow drum area, and the area of the hollow drum area is calculated by connecting the boundary knocking points of the hollow drum area to form a polygon, and then compared with the total area of the house ground to obtain the proportion of the area of the hollow drum area to the total area.

4. The housing finishing quality evaluation system according to claim 3, characterized in that: The setting rule is to draw an equal-area grid on the ground, take each grid intersection as a knocking point, and assign a unique coordinate number to each knocking point. When collecting the sound wave signal generated by each knock, the coordinate number of the corresponding knocking point is recorded synchronously, thereby establishing a preliminary association between the sound wave signal and the position of the knocking point. Once the discrimination model determines that a certain sound wave signal comes from a hollow area, the position of the knocking point corresponding to the sound wave signal is determined according to the preliminary association between the sound wave signal and the position of the knocking point.

5. The housing finishing quality evaluation system according to claim 1, characterized in that: The specific analysis method of the joint quality analysis module further includes: Gradient calculation is performed on each joint area image to obtain the gradient amplitude of each pixel point in each joint area image. A gradient amplitude threshold is set, and each pixel point in each joint area image is traversed. If the gradient amplitude of a certain pixel point is greater than or equal to the gradient amplitude threshold, the pixel point is determined to be an edge point. If the gradient amplitude of a certain pixel point is less than the gradient amplitude threshold, the pixel point is determined to be a non-edge point. The joint edge of each joint area image is obtained by connecting each edge point.

6. The housing finishing quality evaluation system according to claim 1, characterized in that: The specific analysis method of the closing-in process evaluation module is: The closing-in area of the house ground to be detected is divided, a three-dimensional scanner is used to scan each closing-in part of the house ground, the angle point cloud data of each closing-in part of the house ground collected is obtained, surface reconstruction is performed according to the angle point cloud data of each closing-in part of the house ground, the actual three-dimensional model of each closing-in part of the house ground is generated, the actual three-dimensional model of each closing-in part of the house ground is aligned with the designed standard three-dimensional model in space position and direction, and the alignment is performed through an iterative closest point algorithm. The distance deviation of each group of corresponding points between the actual model and the standard model is calculated by comparing the three-dimensional models after the alignment, and the closing-in quality of the house ground is evaluated.

7. The housing finishing quality evaluation system according to claim 6, characterized in that: The specific analysis method of each type of deviation value between the actual model and the standard model at each corresponding position is: A representative local area that can accurately reflect the local characteristics of the model is selected on the standard model, the center point of the local area is selected, and the local area is defined as a template. The defined template is searched and matched on the actual model, the most similar area to the template is found through a gray-based template matching algorithm, and the center point of the area is obtained. Each center point determined on the standard model and the actual model is corresponded one by one to form each group of corresponding points.

8. The housing finishing quality evaluation system according to claim 1, characterized in that: The specific analysis method of the comprehensive quality evaluation module is: The flatness of the house ground, the proportion of the hollow area to the total area, the joint edge quality of the house ground, and the closing-in quality of the house ground are obtained, and weights are set for them in sequence. The products of each item and its corresponding weight are multiplied, and the products are added to obtain a comprehensive evaluation index of the house ground engineering. Each evaluation grade threshold is set, the comprehensive evaluation index of the house ground engineering is compared with the evaluation index range corresponding to each evaluation grade threshold, the evaluation grade corresponding to the comprehensive evaluation index of the house ground engineering is obtained, and the evaluation grade is fed back to the system.

9. A method for evaluating the quality of a fine finish of a house, characterized by, The specific steps of the quality evaluation method are as follows: S1. Obtain the three-dimensional coordinate information of each ground point, extract the elevation information, and evaluate the flatness of the house ground. S2. Collect the sound wave signals of the hollow drum sample area and the normal sample area by knocking, and analyze and extract the corresponding sound wave attenuation characteristics; S3. Construct a discrimination model according to the sound wave attenuation characteristics, screen the hollow drum area, and calculate the proportion of the hollow drum area in the total area by connecting the corresponding boundary knocking points; S4. Extract the joint edge through the threshold algorithm, form a gradient amplitude sequence, count the number of fluctuations to calculate the sawtooth degree of the joint edge, and then evaluate the quality of the joint edge of the house floor; S5. Generate the actual three-dimensional model of each closing part, compare it with the standard model, and evaluate the closing quality of the house floor; S6. Comprehensive evaluation index of house floor engineering, comparison of evaluation grade, and feedback to the system.

Images