1. The utility model provides a fire control facility detects and on-the-spot acceptance evaluation system based on intelligence AI and removal APP, detecting system includes data acquisition system, removes end system and backend server, remove end system with backend server connects, remove end system includes:

an account number login module is used for logging in the mobile phone,

the GPS module is used for positioning the positions of the checking and accepting personnel;

the account login module comprises: the registration module is used for applying registration and warehousing in the system by the fire-fighting detection mechanism and the acceptance mechanism; the identity authentication module is used for checking qualification information, personnel certificates and personnel identity card information of the fire-fighting detection mechanism and the acceptance mechanism and recording; the system comprises an account number distribution module and a login module;

it is characterized in that the mobile terminal system further comprises:

the data input module is used for inputting the acquired data of the data acquisition module to the mobile terminal system;

the data processing module is used for processing the acquired data;

the evaluation module is used for comparing the processed data with the standard data and evaluating to obtain an evaluation result;

and the wireless transmission module is used for uploading the evaluated data to the background server.

2. The intelligent AI and mobile APP based fire protection equipment detection and field acceptance assessment system of claim 1 wherein the data processing module processes the collected data by:

step 1, acquiring acquired data of a fire fighting facility, wherein the acquired data comprises an image and measurement data of an electronic fence area;

step 2, performing three-dimensional reconstruction on the image of the electronic fence area to obtain point cloud data of the fire fighting facility;

step 3, clustering the point cloud data to obtain a plurality of point clusters, and obtaining a minimum external surrounding frame of the point clusters;

calculating the probability density and the direction uniformity index of the point cluster according to the minimum external surrounding frame;

according to the probability density and the direction uniformity index of the point clusters, calculating the consistency index of the point clusters to obtain the average consistency index of the fire-fighting facilities;

step 4, calculating the difference value of the average consistency index and the standard consistency index of the fire-fighting facilities, and calculating the integrity index of each fire-fighting facility;

step 5, acquiring acceptance rules according to fire-fighting acceptance criteria, selecting corresponding key point clusters according to the acceptance rules, extracting mass centers of the key point clusters, acquiring K point clouds closest to the mass centers through a nearest neighbor search algorithm, acquiring mass centers of the point clusters and key point sets of the corresponding point clouds, acquiring key point cluster sets according to the key point sets, and calculating the distance between any two key point clusters;

calculating difference values of the distances and a set standard value respectively, and summing absolute values of the difference values to obtain a normative index of the fire-fighting facility;

and 6, calculating the difference evaluation index according to the integrity index and the normative index.

3. The intelligent AI and mobile APP based fire protection equipment detection and field acceptance assessment system of claim 1, wherein in step 2, the directional uniformity index is calculated as:

carrying out primary gridding processing on the minimum external enclosure frame, counting the number of vector points of each grid method, forming a grid enclosure frame according to the number of vector points of the grid method, and calculating the volume of the grid enclosure frame;

and calculating the directional uniformity index of each grid according to the number and the volume of the grid-method vector points to obtain the directional uniformity index of the point cluster.

4. The intelligent AI and mobile APP based fire fighting equipment detection and field acceptance assessment system according to claim 3, wherein in step 2, the probability density of the point clusters is obtained by performing a first gridding process on the minimum outside bounding box to obtain three-dimensional grids, calculating the probability density of each three-dimensional grid, and calculating the mean of the probability densities of all three-dimensional grids;

the calculation process of the directional uniformity index is as follows: carrying out second gridding processing on the grids obtained after the first gridding processing, counting the number of normal vector points of each second grid, forming a second grid surrounding frame according to the number of the normal vector points, and calculating the volume of the second grid surrounding frame;

and calculating the direction uniformity index of each second grid according to the number and the volume of the grid-method vector points to obtain the direction uniformity index of the point cluster.

5. The intelligent AI and mobile APP based fire protection equipment detection and field acceptance assessment system of claim 3 or 4, wherein in step 2, before the clustering process, further comprising meshing the point cloud data of the electric fence area.

6. The intelligent AI and mobile APP based fire fighting equipment detection and field acceptance assessment system according to claim 5, wherein the standard consistency index is obtained by clustering the point cloud data of a standard library according to pre-established point cloud data of the standard library to obtain a plurality of standard point clusters and obtaining a minimum external bounding box of the standard point clusters;

calculating the probability density and the direction uniformity index of the standard point cluster according to the minimum external enclosure frame;

calculating the consistency index of the standard point cluster according to the probability density and the direction uniformity index of the standard point cluster, and acquiring the standard consistency index of the fire-fighting facility;

the point clusters in step 2 correspond to the standard point clusters one to one.

7. The intelligent AI and mobile APP based fire protection equipment detection and field acceptance assessment system of claim 5, wherein the standard distance is calculated from the point cloud data of the standard library for the chamfer distance between any two standard key point clusters, summed.

8. The intelligent AI and mobile APP based fire protection equipment detection and field acceptance assessment system of claim 7 wherein the normative index is:

wherein the content of the first and second substances,is a set standard value corresponding to the a-th acceptance rule,the distance between any two key point clusters corresponding to the a-th acceptance rule is shown.

9. The intelligent AI and mobile APP based fire protection equipment detection and field acceptance assessment system of claim 8 wherein the normative index is:

wherein q isThe number of the acceptance rules, n1 is the number of the standard point clusters, n is the number of the point clusters,represents the average probability density of the jth standard point cluster,mean probability density of the ith point cluster, respectively.

10. The intelligent AI and mobile APP based fire fighting equipment detection and field acceptance assessment system according to claim 2 wherein said assessment module compares said discrepancy indicator with a set standard and said measurement data with a set standard data respectively, and when said discrepancy indicator is less than said set standard and said measurement data is similar to said set standard data, said fire fighting equipment is qualified and an assessment of said fire fighting equipment is obtained.

Background

The acceptance of fire-fighting facilities is one of the important links related to the life and property safety of people. Since the 20 th century 90 s, China begins to detect building fire-fighting facilities, the quality of construction and installation of automatic fire-fighting facilities in buildings is greatly improved all the time, the reliability of facility operation is obviously improved, the building fire-fighting facilities are installed to prevent accidents in the bud, and the life and property safety of people is better guaranteed.

However, at present, the fire-fighting facilities in a part of large buildings still have the situations that the fire-fighting facilities cannot control the disaster timely and effectively due to the fact that the fire-fighting facilities are not of the quality and are not maintained at the right place, so that the acceptance of the fire-fighting facilities is particularly important.

However, when fire fighting equipment is inspected in the prior art, the safety of the fire fighting equipment is evaluated mainly by observation and recording of technical service personnel, but as the state connection of the service personnel to fire fighting equipment in a building is not comprehensive, the whole safety condition of the building and the existing fire hazard cannot be identified, the technical service personnel cannot know fire fighting laws and regulations and technical standards, and are unfamiliar with fire fighting detection equipment instruments, the detection quality is reduced, some detection equipment instruments even use misalignment and failure detection equipment instruments, the detection quality and the service level are finally low, and the authenticity and the accuracy of detection results are lack of public confidence, so that the detection quality of the fire fighting equipment of the building is directly influenced.

Disclosure of Invention

The invention aims to provide a fire-fighting equipment detection and on-site acceptance evaluation system based on intelligent AI and mobile APP, which is used for solving the problem that the detection quality of the existing on-site acceptance evaluation system for building fire-fighting equipment is inaccurate due to human or instrument reasons and the like.

The invention provides a technical scheme of a fire fighting facility detection and on-site acceptance evaluation system based on intelligent AI and mobile APP, wherein the detection system comprises a data acquisition system, a mobile end system and a background server, the mobile end system is connected with the background server, and the mobile end system comprises:

an account number login module is used for logging in the mobile phone,

the GPS module is used for positioning the positions of the checking and accepting personnel;

the account login module comprises: the registration module is used for applying registration and warehousing in the system by the fire-fighting detection mechanism and the acceptance mechanism; the identity authentication module is used for checking qualification information, personnel certificates and personnel identity card information of the fire-fighting detection mechanism and the acceptance mechanism and recording; the system comprises an account number distribution module and a login module;

the mobile terminal system further includes:

the data input module is used for inputting the acquired data of the data acquisition module to the mobile terminal system;

the data processing module is used for processing the acquired data;

the evaluation module is used for comparing the processed data with the standard data and evaluating to obtain an evaluation result;

and the wireless transmission module is used for uploading the evaluated data to the background server.

Further, the process of processing the collected data by the data processing module is as follows:

step 1, acquiring acquired data of a fire fighting facility, wherein the acquired data comprises an image and measurement data of an electronic fence area;

step 2, performing three-dimensional reconstruction on the image of the electronic fence area to obtain point cloud data of the fire fighting facility;

step 3, clustering the point cloud data to obtain a plurality of point clusters, and obtaining a minimum external surrounding frame of the point clusters;

calculating the probability density and the direction uniformity index of the point cluster according to the minimum external surrounding frame;

according to the probability density and the direction uniformity index of the point clusters, calculating the consistency index of the point clusters to obtain the average consistency index of the fire-fighting facilities;

step 4, calculating the difference value of the average consistency index and the standard consistency index of the fire-fighting facilities, and calculating the integrity index of each fire-fighting facility;

step 5, acquiring acceptance rules according to fire-fighting acceptance criteria, selecting corresponding key point clusters according to the acceptance rules, extracting mass centers of the key point clusters, acquiring K point clouds closest to the mass centers through a nearest neighbor search algorithm, acquiring mass centers of the point clusters and key point sets of the corresponding point clouds, acquiring key point cluster sets according to the key point sets, and calculating the distance between any two key point clusters;

calculating difference values of the distances and a set standard value respectively, and summing absolute values of the difference values to obtain a normative index of the fire-fighting facility;

step 6, calculating the difference evaluation index according to the integrity index and the normative index; and when the difference index is smaller than a set standard and the measured data is similar to the set standard data, the fire-fighting equipment is qualified.

Further, in step 2, the calculation process of the directional uniformity index is as follows:

carrying out primary gridding processing on the minimum external enclosure frame, counting the number of vector points of each grid method, forming a grid enclosure frame according to the number of vector points of the grid method, and calculating the volume of the grid enclosure frame;

and calculating the directional uniformity index of each grid according to the number and the volume of the grid-method vector points to obtain the directional uniformity index of the point cluster.

Further, in the step 2, the probability density of the point cluster is obtained by performing a first gridding process on the minimum outside bounding box to obtain three-dimensional grids, calculating the probability density of each three-dimensional grid, and calculating the mean value of the probability densities of all the three-dimensional grids;

the calculation process of the directional uniformity index is as follows: carrying out second gridding processing on the grids obtained after the first gridding processing, counting the number of normal vector points of each second grid, forming a second grid surrounding frame according to the number of the normal vector points, and calculating the volume of the second grid surrounding frame;

and calculating the direction uniformity index of each second grid according to the number and the volume of the grid-method vector points to obtain the direction uniformity index of the point cluster.

Further, in step 2, before the clustering process, a meshing process is further performed on the point cloud data of the electronic fence area.

Further, the standard consistency index is to perform clustering processing on the point cloud data of the standard library according to the point cloud data of the pre-established standard library to obtain a plurality of standard point clusters, and obtain a minimum external enclosure frame of the standard point clusters;

calculating the probability density and the direction uniformity index of the standard point cluster according to the minimum external enclosure frame;

calculating the consistency index of the standard point cluster according to the probability density and the direction uniformity index of the standard point cluster, and acquiring the standard consistency index of the fire-fighting facility;

the point clusters in step 2 correspond to the standard point clusters one to one.

Further, the integrity indexes of the fire-fighting equipment are as follows: GM2 ═ GM-GM₀Wherein GM is the average consistency index of the point cluster, GM₀Is the standard consistency index of the fire-fighting facilities.

Further, the standard distance is obtained by calculating the chamfer angle distance between any two standard key point clusters according to the point cloud data of the standard library and then summing.

Further, the normative index is as follows:

wherein the content of the first and second substances,is a set standard value corresponding to the a-th acceptance rule,and the distance between any two key point clusters corresponding to the a-th acceptance rule is obtained.

Further, the normative index is as follows:

wherein q is the number of acceptance rules, n1 is the number of standard point clusters, n is the number of point clusters,represents the average probability density of the jth standard point cluster,mean probability density of the ith point cluster, respectively.

Further, the evaluation module compares the difference index with a set standard and compares the measurement data with the set standard data, and when the difference index is smaller than the set standard and the measurement data is close to the set standard data, the fire-fighting equipment is qualified, and an evaluation result of the fire-fighting equipment is obtained.

The invention has the beneficial effects that:

the data processing module arranged in the on-site acceptance evaluation system can process the acquired collected data, particularly the image of the electronic fence area, namely the consistency index is acquired by converting the image data into point cloud data, and the influence of human factors can be avoided during detection; simultaneously, starting from the integrality and the normative of the fire-fighting equipment, the detection of the fire-fighting equipment is carried out, and the detection accuracy is improved.

The invention respectively carries out twice gridding treatment on a plurality of point clusters, reflects the point cloud uniformity of the first grid by utilizing the direction uniformity index of the second grid, and can effectively and accurately reflect the spatial distribution of the point cloud in the grid from the perspective of local detail.

Drawings

In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings needed to be used in the embodiment will be briefly described as follows:

FIG. 1 is a block diagram of the fire fighting equipment detection and on-site acceptance evaluation system based on intelligent AI and mobile APP;

FIG. 2 is a block diagram of the architecture of the mobile end system of the intelligent AI and mobile APP based fire protection detection and on-site acceptance assessment system of the present invention;

FIG. 3 is a flowchart of a method for intelligent AI and mobile APP based fire fighting equipment detection and field acceptance assessment according to the present invention;

FIG. 4 is a flow chart of a data processing method of a data processing module in the intelligent AI and mobile APP based fire fighting equipment detection and field acceptance assessment system of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention.

Fig. 1 is a fire fighting equipment detection and on-site acceptance evaluation system based on intelligent AI and mobile APP according to the present invention, which includes a data acquisition system, a mobile end system and a background server.

The data acquisition system is artificial intelligence toolbox for gather fire-fighting equipment's data, and this artificial intelligence toolbox includes image acquisition device, illuminometer, sound level meter, distancer, tape measure, anemograph, digital micro-manometer etc. and specific other relevant detection instrument, and here no longer too much introduction refers to the detection instrument that has now disclosed.

The image acquisition device mainly acquires the image information of the fire fighting equipment, analyzes the fire fighting facilities, can quickly process a plurality of images, and improves the efficiency of acquiring the image information. While other illuminometers, sound level meters, distance meters, tape measures, etc. are used to measure measurable data of the degree to which an object is illuminated, the noise of a fire fighting installation, the distance of a fire fighting installation, the height, width, length, area, thickness, etc., respectively.

It should be noted that the artificial intelligence toolbox is an independent hardware device, for the image acquisition device, it can transmit the acquired image data to the mobile terminal system, the transmission can be a wired transmission, it can also directly set the image acquisition device on the mobile terminal system, and when the mobile terminal system is a mobile phone system, the image acquisition device can be a camera carried by the mobile phone; data collected by tools such as an illuminometer, a sound level meter and the like can be directly and manually recorded into a mobile end system.

In addition, an arrow in the structural block diagram of fig. 1 indicates a relationship of data transmission among the data acquisition system, the mobile end system, and the background server, and does not represent a connection relationship among the three; in fact, the data acquisition system is not connected with the mobile end system, but after the data acquisition system acquires data, the acceptance personnel records the acquired related data into the mobile end system; and the mobile terminal system is wirelessly connected with the background server.

As shown in fig. 2, the mobile terminal system includes an account login module, a GPS module, a data entry module, a data processing module, an evaluation module, and a wireless transmission module;

the account login module comprises:

the registration module is used for applying registration and warehousing in the system by the fire-fighting detection mechanism and the acceptance mechanism;

the identity authentication module is used for checking qualification information, personnel certificates and personnel identity card information of the fire-fighting detection mechanism and the acceptance mechanism and recording;

the account number distribution module is used for randomly establishing a temporary account number and distributing the temporary account number to corresponding tasks of technical service personnel after filing; it should be noted that the assigned account may be automatically established as an authorized account through a program based on uuid, and a specific method is the prior art and is not described here in detail.

And the login module is used for logging in by using the temporary account.

And the GPS module is used for positioning the positions of the check personnel, judging whether the check personnel reach the nearby fire-fighting facilities for checking and accepting, and reminding the check personnel to log in the APP.

And the data entry module is used for inputting the data acquired by the data acquisition system into the mobile terminal system.

And the data processing module is used for processing the acquired data acquired by the data acquisition system, such as performing three-dimensional reconstruction on the acquired image, acquiring image characteristics and performing acceptance analysis on the fire-fighting equipment.

And the evaluation module is used for comparing the processed data with the standard data and evaluating.

And the wireless transmission module is used for uploading the evaluated data to a background server, and can be a GPRS module or a WIFI module.

The handheld terminal system is basic acceptance equipment and is carried by an acceptance person, and when the acceptance person reaches the position of the fire-fighting equipment needing acceptance, the data of the acceptance can be recorded into the handheld terminal system.

The handheld terminal system in this embodiment may be a mobile terminal App.

And the background server forms an acceptance evaluation table according to the received evaluated data and stores the related data.

It should be noted that the acceptance personnel (detection institution, acceptance institution) need to apply for registration and warehousing, and perform operations such as qualification information, personnel certificate, personnel identification card filing, etc. on the platform.

When the fire-fighting equipment is accepted, the fire-fighting equipment detection mechanism and the fire-fighting acceptance evaluation mechanism need to carry out contract record; after the recording, the platform returns a temporary account number of the institution company, and the temporary account number is distributed to the task of the acceptance staff and establishes the temporary account number of the acceptance staff.

Position information wherein is whether be located near the fire control facility region that needs the acceptance for the acceptance personnel, and after the acceptance personnel got into corresponding region, usable interim account number login APP brushed face real name authentication began to utilize removal end APP and artificial intelligence terminal toolbox to carry out fire control and detects/assess.

Based on the fire-fighting equipment on-site acceptance evaluation system, the invention provides an embodiment of a data processing process of a data processing module suitable for the evaluation system, and the mobile APP is taken as an example and is specifically introduced by combining a specific application scene.

As shown in fig. 3, the acceptance evaluation method of the fire fighting equipment detection and on-site acceptance evaluation system based on the intelligent AI and the mobile APP of the present invention comprises:

1) acquiring position information of an acceptance person, judging whether the position information is in an electronic fence area, if so, acquiring identity information of the acceptance person, and logging in a mobile terminal; the identity information comprises a login account number and a corresponding login password of the acceptance checking personnel; the position information is that the checking and accepting personnel are located in the fire fighting facility detection area;

2) and the mobile end system acquires the uploaded acquired data of the fire-fighting equipment in real time, processes the acquired data, compares the processed acquired data with the set standard data, generates an acceptance result according to the comparison result, and performs acceptance evaluation on the fire-fighting equipment according to the acceptance result.

The electronic fence area in the step 2) is an area which needs to be subjected to fire fighting facility detection and fire fighting acceptance assessment in a working site.

As another embodiment, the mobile terminal APP background further comprises detection of technical service personnel, behavior detection is performed on personnel in the electronic fence area by using a field intelligent camera, and when the behavior is abnormal, a request is fed back to the terminal to perform an authentication requirement.

In this embodiment, the data processing module processes the acquired data, and the processing procedure, as shown in fig. 4, includes the following steps:

step 1, acquiring acquired data of a fire fighting facility, wherein the acquired data comprises an image and measurement data of an electronic fence area;

the data collected by the fire fighting equipment in the step comprises two groups of data, wherein one group of data is measurable data and the other group of data is detectable data; the measurable indexes are measured on site by corresponding measuring tools in the artificial intelligent terminal toolbox on the measurable indexes of distance, height, width, length, area, thickness, air speed and the like of the fire-fighting equipment. Since the measurable index is data measured directly by the corresponding tool, the present invention will not be discussed in detail.

The detectable index is the detection of image data through the image acquisition device, and mainly realizes the recording of the measurable index of the fire-fighting equipment and the acquisition of the image of the fire-fighting equipment, so as to obtain the detected/evaluated data.

Step 2, performing three-dimensional reconstruction on the image of the electronic fence area to obtain point cloud data of the fire fighting facility;

the image of the electronic fence area in this embodiment is measured on site through an artificial intelligence terminal toolbox, and specifically, the artificial intelligence terminal toolbox is used to scan the electronic fence area to obtain a three-dimensional representation of the area, wherein a sensor device is embedded in the artificial intelligence terminal toolbox, and a depth camera is preferred. A three-dimensional representation of the region, preferably a point cloud representation. The image collected by the depth camera is a depth image. And then, carrying out three-dimensional reconstruction on the scanned image of the electronic fence area to obtain point cloud data.

Therein, the three-dimensional reconstruction typically comprises the following steps:

a) extracting image features (such as SIFT, SURF and the like);

b) calculating feature matching between the images by using the features;

c) performing sparse reconstruction based on the matched features to obtain the camera pose of each image and a sparse feature point cloud (SfM);

d) and performing dense reconstruction based on the camera pose to obtain dense point cloud (PMVS/CMVS).

Since three-dimensional reconstruction is mature in the field of computer vision and its application, specific details of three-dimensional reconstruction are not described herein.

Furthermore, fire-fighting equipment target identification is carried out based on point cloud data, a three-dimensional object detection technology based on deep learning is adopted in the preferred technology, a three-dimensional object detector takes point clouds of a scene as input, a directional three-dimensional boundary box is generated around each detection object, a specific method comprises a method based on Reion Proposal and a method based on Single Shot, and finally three-dimensional representation of each fire-fighting equipment is obtained.

Step 3, clustering the point cloud data to obtain a plurality of point clusters, and obtaining a minimum external surrounding frame of the point clusters; calculating the probability density and the direction uniformity index of the point cluster according to the minimum external surrounding frame;

according to the probability density and the direction uniformity index of the point clusters, calculating the consistency index of the point clusters to obtain the average consistency index of the fire-fighting facilities;

the method for calculating the probability density of the point cluster in the above embodiment is as follows:

counting the point clouds in the point clusters, calculating the mean value and covariance of the point cloud geometry, and calculating the probability density of each point cluster;

density of the point cloud:

mean value of point cloud:

covariance of the point cloud:

the probability density of the point cluster is then:

wherein p is_iIs the density of the ith point cloud in the point cluster, n is the number of the point clouds in the point cluster,mean probability density of the ith point cluster, respectively.

In another embodiment, in order to facilitate the clustering of the fire fighting equipment components, the present application performs a gridding process before performing the clustering process on the point cloud data.

The calculation process of the directional uniformity index in the above embodiment is as follows:

carrying out primary gridding processing on the minimum external enclosure frame, counting the number of vector points of each grid method, forming a grid enclosure frame according to the number of vector points of the grid method, and calculating the volume of the grid enclosure frame;

and calculating the directional uniformity index of each grid according to the number and the volume of the grid-method vector points to obtain the directional uniformity index of the point cluster.

The directional uniformity indexes of each grid are as follows:

wherein the content of the first and second substances,gridN_brespectively representing the average number of grid normal vector points and the number of normal vector points in the b-th grid;gridV_bthe average volume of the 3D minimum bounding box of the grid normal vector points and the volume of the 3D minimum bounding box of the normal vector points in the B-th grid are respectively represented, B is the number of the grids, and B in this embodiment is 8.

In this embodiment, UD_bLarger values indicate a more uneven directional distribution.

The point cloud consistency index of each point cluster is as follows:

wherein, w1 and w2 are corresponding weight mapping values, respectively, 4 and 0.2.

The average consistency index of the point cluster is:

wherein n is the number of the point clusters, M_lAnd the point cloud consistency index of the first point cluster is obtained.

It should be noted that different fire-fighting equipment components may exist in the invention, so that firstly, DBSCAN density clustering is performed on the point clouds in each grid in the standard point cloud to obtain a plurality of point clusters, wherein one point cluster is considered as one component, and the DBSCAN density clustering generally needs manual parameter adjustment; there should generally be a threshold G1 for clusters of points, ensuring that each cluster of points has at least a G1 point cloud composition, with an empirical value of 50.

In this embodiment, after the probability density of the point clusters is calculated, statistics of point clouds of the point clusters in directions need to be considered, so that a normal vector of each point cloud in the grid after the gridding processing is estimated, where the normal vector estimation method includes K neighbor estimation, radius neighbor estimation, hybrid search estimation, and the like, and specifically, the adopted method may be selected according to actual situations, and finally, a normal vector point of each point cloud is obtained.

According to the invention, the spatial distribution of the point clouds in the grids can be effectively reflected through the average probability density distribution and the direction uniformity index of each grid, so that the point clouds are used for detecting the integrity of facilities, and meanwhile, the problem of accuracy reduction caused by the inconsistency of the point clouds (the point clouds are three-dimensionally reconstructed and the reconstructed point clouds may be inconsistent even if the same equipment is at the same position) is further eliminated by considering the volume and the number of the minimum external enclosure frames formed by each second grid.

Step 4, calculating the difference value of the average consistency index and the standard consistency index of the electronic fence area, and calculating the integrity index of each fire-fighting facility of the electronic fence area;

the integrity indexes of the fire-fighting equipment in the embodiment are as follows:

GM2＝|GM-GM₀|

wherein GM is the average consistency index of the point cluster, GM₀Is a standard consistency index.

The standard consistency index in the embodiment is to perform clustering processing on point cloud data of a standard library according to point cloud data of the standard library established in advance to obtain a plurality of standard point clusters, and obtain a minimum external enclosure frame of the standard point clusters; calculating the probability density and the direction uniformity index of the standard point cluster according to the minimum external surrounding frame;

calculating the consistency index of the standard point cluster according to the probability density and the direction uniformity index of the standard point cluster, namely obtaining the standard consistency index of the fire-fighting facility;

it should be noted that, in the embodiment, the average consistency index and the standard consistency index of the fire fighting equipment are obtained by mapping three-dimensionally reconstructed point clusters and standard point clusters one by one, that is, a point cluster set of a reconstructed three-dimensional point cloud is obtained by mapping a point cluster set of a standard point cloud, a minimum external bounding box of each point cluster in the standard point cloud is obtained first, and then a point cloud in the minimum external bounding box at the same position in the reconstructed point cloud is obtained.

Step 5, acquiring acceptance rules according to fire-fighting acceptance criteria, selecting corresponding key point clusters according to the acceptance rules, extracting mass centers of the key point clusters, acquiring K point clouds closest to the mass centers through a nearest neighbor search algorithm, acquiring mass centers of the point clusters and key point sets of the corresponding point clouds, acquiring key point cluster sets according to the key point sets, and calculating the distance between any two key point clusters;

calculating difference values of the distances and a set standard value respectively, and summing absolute values of the difference values to obtain a normative index of the fire-fighting facility;

wherein, the calculated chamfer distance of all the key point clusters is as follows:

in the formula, S_k、S_tRespectively a key point set of the kth key point cluster and the t-th key point cluster; x is S_kAny coordinate in the set, y being S_tAny coordinate in the set; the first term represents S_kTo any point of S_tThe second term represents S_tTo any point of S_kAverage minimum distance of (c).

The normative indexes are as follows:

wherein the content of the first and second substances,is a set standard value corresponding to the a-th acceptance rule,the distance between any two key point clusters corresponding to the a-th acceptance rule is shown.

The set standard value in this embodiment can be obtained by calculating the distance between any two standard key point clusters according to the point cloud data of the standard library, and then summing the calculated distances; it should be clear that the set standard value at this time corresponds to the calculated distance between any two key point clusters one by one. That is, the corresponding standard values set under different acceptance rules are different, and may be set directly according to actual conditions.

In the above embodiment, the inconsistency of the point clouds at the spatial positions is considered, so that the 3D point cloud chamfering distance is adopted to estimate the distance of the nearest neighbor point cloud set at the same position, thereby more accurately estimating the normalization of the facility, and avoiding the inconsistency of the point clouds at the spatial positions and the distance error caused by the three-dimensional reconstruction error of the point clouds.

As a further embodiment, the present invention also considers the influence of the probability density of the point cluster on the normative index, and the formula is as follows:

wherein q is the number of acceptance rules, n1 is the number of standard point clusters, n is the number of point clusters,represents the average probability density of the jth standard point cluster,mean probability density of the ith point cluster, respectively.

In this embodiment, the larger the difference of GM5, the less standardized the facility is; the term is the difference between the standard point cluster probability density sum and the detection point cluster probability density sum, and the larger the difference is, the larger the change of the average probability density of the point clusters is, namely, the more or larger impurities exist or parts are missing.

It should be noted that the acceptance rule in this embodiment is formulated according to the fire-fighting acceptance standard, different acceptance standards correspond to different acceptance rules, the key point clusters corresponding to different acceptance rules are different, that is, the meaning of the calculated chamfer distance is also different. The number of the specific acceptance rules may be one or more, and the set of the key point clusters corresponding to each acceptance rule is also different.

For example, the fire-fighting acceptance standard of the fire hydrant cabinet is as follows:

1) the center of the valve is 140mm away from the side surface of the box, the center of the valve is 100mm away from the rear inner surface of the box, and the allowable deviation is +/-5 mm;

2) the verticality allowable deviation of the fire hydrant box body installation is 3 mm.

And according to the acceptance rule, the positions of the fire hydrant valve center and the side surface of the box are all key points, and the center point cluster and the point cluster on the side surface of the box are correspondingly selected to calculate the chamfering distance.

And 6, calculating the difference evaluation index according to the integrity index and the normative index.

Wherein the difference evaluation indexes are as follows:

Diff＝GM2*GM5

the above formula is the most preferable mode of the present invention, and it is needless to say that the Diff may be calculated by multiplying GM2 by GM 4; for Diff, the larger the value, the more different the device is considered to be from standard library devices.

Therefore, the scheme of the invention can solve the problems of inaccurate and unreliable acceptance evaluation caused by inconsistent standards and strong subjectivity of technical service personnel in some inspection items in fire-fighting acceptance and evaluation.

The setting criteria in this embodiment are; since different fire-fighting facilities are different, standards of different fire-fighting facilities are different, and therefore, the standards can be set according to actual conditions.

And according to the difference evaluation indexes obtained in the steps, the evaluation module compares the difference evaluation indexes with a set standard and the measurement data with the set standard data, and when the difference indexes are smaller than the set standard and the measurement data are similar to the set standard data, the fire-fighting equipment is qualified, and the acceptance result of the fire-fighting equipment is obtained.

The acceptance result can be the result of whether the fire-fighting equipment is qualified or not, and can also be a directly generated evaluation table; of course the rating table may also be generated in the background server.

The mobile APP uploads the obtained acceptance result to the background server for storage through the wireless communication module.

As another embodiment, the present invention provides another example of the data processing method of the data processing module.

Specifically, in this embodiment, the meshing process is performed on the plurality of point clusters twice, the point clusters are firstly subjected to the meshing process to obtain different first three-dimensional grids, and the probability density of each first three-dimensional grid is calculated; performing secondary gridding processing on each three-dimensional grid for the second time, calculating a direction uniformity index of the second three-dimensional grid, and reflecting the uniformity of the first three-dimensional grid by using the direction uniformity index of the second three-dimensional grid;

specifically, the method for calculating the consistency index of the point cluster according to the embodiment includes the following steps:

1. calculating a probability density of the first three-dimensional mesh:

1) carrying out gridding treatment on the plurality of point clusters to obtain a three-dimensional grid;

the empirical size of the grid division in this embodiment is:

l'＝L/4

w'＝W/2

h'＝H/6

wherein L, W, H is respectively the length, width and height of the minimum external enclosure frame of the fire-fighting equipment,

h' is the length, width and height of the grid respectively.

Taking the fire hydrant as an example, the fire hydrant can be divided into 48 three-dimensional grids according to the grids.

2) Counting the point clouds in the three-dimensional grid, calculating the mean value and covariance of the point cloud geometry, and

calculating the probability density of the three-dimensional grid;

wherein p is_IThe probability density of the I point cloud in the grid is shown, and N is the number of the grids in the point cluster.

Since a specific method has been given in the first embodiment, details of the calculation of the probability density of the point cloud in the above formula are not described here.

2. Calculating a directional uniformity index:

1) carrying out further grid division on the first three-dimensional grids, uniformly dividing each first three-dimensional grid into h small grids, wherein the grids are called second three-dimensional grids, and the length, width and height of each second three-dimensional grid are half of those of the original grids;

2) counting the number of normal vector points of each second three-dimensional grid, constructing a 3D minimum external enclosure frame according to the number, calculating the volume of the enclosure frame, and calculating the directional uniformity index of each second three-dimensional grid according to the number and the volume;

wherein the directional uniformity index of each grid is as follows:

wherein the content of the first and second substances,gridN_rrespectively representing the average number of the normal vector points of the second three-dimensional grid and the number of the normal vector points in the r-th second three-dimensional grid,gridV_rand respectively representing the average volume of the 3D minimum external bounding box of the normal vector point of the second three-dimensional grid and the volume of the 3D minimum external bounding box of the normal vector point in the r-th second grid.

It should be noted that, in this embodiment, only the statistical number can only know the number difference in each large direction, and cannot know the directional distribution in each large direction (i.e., each quadrant), so here, the directional distribution is reflected by the volume of the 3D minimum bounding box of the normal vector point, and a larger value means a more uniform directional distribution. And extracting a 3D minimum external bounding box for the normal vector points in each second grid, and then reflecting the directional uniformity distribution of the second grid by using the volume of the minimum external bounding box.

3. And calculating the consistency index of the point cloud in each first three-dimensional grid, and further acquiring the average consistency index of the point clusters.

In this embodiment, the spatial distribution of the point clouds in the grids can be effectively reflected by the average probability density distribution and the direction uniformity index of each grid, and then the spatial distribution is used for detecting the integrity of the facility, and meanwhile, the volume and the number of the minimum external bounding boxes formed by each second grid are considered, so that the problem of accuracy reduction caused by the inconsistency of the point clouds (the point clouds are three-dimensionally reconstructed point clouds, and the reconstructed point clouds may not be consistent even if the same equipment is at the same position) is further solved.

In this embodiment, since the number of reconstructed mesh point clouds is not always consistent with the point clouds in the standard library, the average probability density of the computed mesh represents the probability that a position in the mesh is occupied, and the distribution of the point clouds in the mesh can be represented.

The statistics can represent the spatial distribution of the point clouds in the grid, and the statistics of the point clouds in the direction is lacked, so that the normal vector of each point cloud in the grid is further estimated, and an implementer can freely select the normal vector by using a normal vector estimation method such as K neighbor estimation, radius neighbor estimation, mixed search estimation and the like, so that the normal vector point of each point cloud is finally obtained, and the normal vector point also needs to be in the grid.

It should be noted that while the preferred embodiments of the present invention have been described, additional variations and modifications to these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.