A patrol inspection method based on a fire protection pipe network self-service patrol inspection system
By combining hardware and software technology in the self-service inspection system of the fire protection pipeline network, rapid automatic detection of abnormal conditions such as cracks on the surface of the fire protection pipeline network is achieved, and the problems of low manual detection efficiency and safety hazards in the existing technology are solved, and the problems of high-quality manual detection are improved. Inspection quality and efficiency.
Patent Information
- Application Number
- CN202111369790.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-18
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2041-11-18
AI Technical Summary
The prior art lacks mature rapid automatic detection equipment for surface cracks of fire-fighting pipe networks, resulting in the detection reliance on manual labor, low efficiency and safety risks.
A self-service inspection system based on fire protection pipeline network is designed, using hardware equipment such as dual-track flaw detection car platform, surface array camera and RFID positioning technology, combined with software modules such as image debugging, puzzle and intelligent identification module, and using YOLO target detection and bale Si fusion and BP neural network realize automatic identification of abnormal conditions.
It improves the detection efficiency and quality, overcomes the problem of insufficient manual recording, reduces the difficulty of detection, and enhances the monitoring ability of abnormal situations in the fire protection pipeline network.
Smart Images

Figure CN114120227B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of fire protection pipe network inspection, and in particular to an inspection method based on a fire protection pipe network self-service inspection system. Background Art
[0002] Fire-fighting pipes refer to pipe materials used for fire-fighting, connecting fire-fighting equipment and instruments, and transporting water, gas or other media for fire-fighting. Due to special needs, there are special requirements for the thickness and material of fire-fighting pipes, and they are sprayed with red paint for transporting water for fire-fighting.
[0003] Rapid and automatic detection of cracks on the surface of fire protection pipe networks is a current technical challenge. There is a lack of relatively mature detection equipment on the market, and it currently relies mainly on manual identification. With the construction of smart cities, pipeline surface cracks, water leakage, block falling and other defects are increasingly becoming hidden dangers to safe operations. However, the current detection of pipeline defects still mainly relies on manual inspections for judgment, which is inefficient and the detection results are greatly affected by human factors, posing a great safety hazard. Therefore, it is necessary to design an inspection method based on a self-service inspection system for fire protection pipe networks. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and to better and effectively solve the problem of rapid and automatic detection of surface cracks in fire protection pipe networks. A patrol method based on a self-service patrol system for fire protection pipe networks is provided, which can intelligently patrol the pipe network and automatically identify abnormal conditions in the pipeline environment, effectively improving the efficiency and quality of the detection work.
[0005] In order to achieve the above object, the technical solution adopted by the present invention is:
[0006] A patrol method based on a fire protection pipe network self-service patrol system comprises the following steps:
[0007] Step (A), designing and purchasing hardware equipment of the fire pipe network self-service inspection system, and assembling the fire pipe network self-service inspection system;
[0008] Step (B), based on the assembled fire pipe network self-service inspection system, design the software module of the system;
[0009] Step (C), constructing the intelligent algorithm flow of the system according to the software modules of the system;
[0010] Step (D), designing the system usage process based on the system's intelligent algorithm process;
[0011] Step (E), complete the intelligent inspection of the fire protection pipe network according to the system usage process.
[0012] The aforementioned inspection method based on the fire pipe network self-service inspection system, step (A), designing and purchasing the hardware equipment of the fire pipe network self-service inspection system, assembling it into a fire pipe network self-service inspection system, wherein the fire pipe network self-service inspection system includes a control unit, a detection unit, a positioning unit, an operating unit and a vehicle platform, the control unit includes but is not limited to a host, a circuit board, a light source controller and an encoder, the detection unit includes but is not limited to an array camera and a stroboscopic light source, the positioning unit uses an RFID (precision identification technology) positioning tag, the control unit uses a touch screen, and the vehicle-mounted platform uses a double-track flaw detection trolley.
[0013] The aforementioned inspection method based on the fire pipe network self-service inspection system, step (B), based on the assembled fire pipe network self-service inspection system, designs the system's software modules, wherein the system's software modules include an image debugging acquisition module, an image playback module, a puzzle module and an intelligent recognition module, the image debugging acquisition module is used to receive and display the collected data in real time, the image playback module is used to operate historical data, the puzzle module is used to piece together the images collected by six cameras into a complete tunnel surface, and the intelligent recognition module is used to identify the defect type and produce a report.
[0014] The aforementioned inspection method based on the fire pipe network self-service inspection system, step (C), constructs the system's intelligent algorithm process according to the system's software module, wherein the system's intelligent algorithm process includes the following steps:
[0015] Step (C1), collecting and inputting tunnel images, using YOLO (object detection technology) to detect foreign objects, and binarizing the bottom ten lines of each image, and comparing the binarized image information with the set threshold;
[0016] Step (C2), if it is higher than the threshold, the current image is determined to be an abnormal image, and then the splicing operation is performed, and the fastener position is located by YOLO, and then the algorithm is used to locate the abnormal image, and the remaining part is subdivided into defects;
[0017] Step (C3), the types of defect detection include crack detection and water leakage detection, and the detection results are output to the defect database.
[0018] The aforementioned inspection method based on the fire pipe network self-service inspection system, step (C1), collects and inputs tunnel images, uses YOLO to detect foreign objects, and performs a binarization operation on the bottom ten lines of each image, and compares the binarized image information with the set threshold by judging, wherein information fusion is used in collecting and inputting tunnel images, and the information fusion adopts Bayesian fusion and BP neural network, and the specific steps are as follows,
[0019] Step (C11), obtain Bayesian fusion. In the process of using multiple sensors to measure a certain parameter, there are two random variables, one is the measured parameter μ, and the other is the measurement result x of each sensor. i ,i=1,2,…,m, and both variables are considered to obey normal distribution, as shown in formula (1),
[0020]
[0021] In this way, the conditional probability density of each measurement value is shown in formula (2):
[0022]
[0023] make Obviously, α is a constant and has nothing to do with the measured parameter μ, as shown in formula (3),
[0024]
[0025] From P = (μ|x 1 ,x 2 ,..,x k ) The derivation results show that it still satisfies the normal distribution form. Here, we assume that but Then μ N The value of is shown in formula (4),
[0026]
[0027] The optimal Bayesian fusion of the measured parameters is shown in formula (5):
[0028]
[0029] Step (C11), constructing a BP neural network, wherein the BP neural network is a multi-layer neural network based on an error back propagation algorithm, having a structure of three layers or more, and comprising an input layer, a hidden layer, and an output layer, and each layer is composed of a number of neurons, and there is no connection relationship between the neurons, and the layers are fully connected, and the weights between each connection are different.
[0030] The aforementioned inspection method based on the fire protection pipe network self-service inspection system, step (D), is based on the system's intelligent algorithm process design system usage process, wherein the system usage process includes mechanical assembly, electrical assembly debugging, system commissioning, real-time monitoring interface, image playback interface, intelligent recognition and statistical reporting and detection completion, the specific steps are as follows,
[0031] Step (D1), mechanical assembly, after assembling the inspection robot, install it on the vehicle body platform;
[0032] Step (D2), electrical assembly and debugging, connecting various communication interfaces and power interfaces of the equipment;
[0033] Step (D3), system commissioning, open the camera commissioning interface, adjust the camera parameters and check the shooting status, and save the collected data after confirming that the status is good;
[0034] Step (D4), real-time monitoring interface, input basic operation information, start the detection function and start collecting data, check the mileage speed and other information during the operation, and complete the operation within the skylight time;
[0035] Step (D5), image playback interface, view historical images and perform image stitching, then view defective marked images, and compare the two sets of historical data, record and save the differences;
[0036] Step (D6), intelligent identification and statistical report, select defect type and detect defect images, complete the detection and filter out the images that pass the inspection, then process the defect images and generate statistical reports with one click;
[0037] Step (D7), detection is completed, close the software and equipment, and receive the detection.
[0038] The aforementioned inspection method based on the fire pipe network self-service inspection system, step (E), completes the intelligent inspection of the fire pipe network according to the system usage process, and the specific steps are as follows:
[0039] Step (E1), using high-definition camera shooting technology to quickly inspect the pipeline network, and the camera shooting position and the light source position cover the tunnel lining in a circumferential direction of not less than 270°, and can complete clear photography of the pipeline network at different viewing distances;
[0040] Step (E2), the system records the location information of the damage and reflects it in the image information, and the location information includes but is not limited to the track direction side and circumferential area position, and multiple longitudinal and circumferential images are synthesized to form a continuous picture;
[0041] Step (E3), the camera must be able to clearly capture water leakage, missing corners, falling blocks, misalignment, cracks of the lining, and the falling off of pipelines in the tunnel. Water leakage includes wet marks, water seepage, dripping, and mud leakage, and can identify faults through system software;
[0042] Step (E4), the system has image display, storage and playback functions, and can compare any two images based on a common mileage benchmark, and the mileage of the image can be adjusted;
[0043] Step (E5), the system playback software has the functions of selecting the mileage range for viewing the continuous images, adjusting the graphics and zooming the images, and can add text annotations, and also includes customizing text and configuring new information such as line name, line type, speed, mileage and annotations, wherein the graphics adjustment items include brightness, contrast, gamma value and saturation;
[0044] Step (E6), the resolution accuracy of the captured image is not less than 1 mm, and the minimum adjustable exposure time is less than 50 μs, and the maximum inspection speed of the system is not less than 10 km / h;
[0045] In step (E7), the system has a storage capacity of no less than 200 kilometers of photographs, and the relevant data can be easily and quickly exported or copied to a ground computer.
[0046] The beneficial effects of the present invention are as follows: an inspection method based on a self-service inspection system for a fire pipe network of the present invention provides a technical basis for intelligent detection of a fire pipe network with the progress of advanced science and technology such as robots, artificial intelligence, big data, cloud computing, the Internet of Things, driverless, and 5G communications; the present invention adopts a dual-track ultrasonic flaw detector trolley platform as a carrier, and takes a single-chip microcomputer STM32F103VET6 as a core, and then utilizes the installation position, power interface, lighting, encoder mileage signal and other interfaces provided by the platform, and utilizes high-definition CCD industrial camera acquisition technology to complete the overall scheme design of the system; effectively realizes real-time monitoring of the pipeline environment, and adopts Bayesian and BP neural network related fusion technology to realize identification and classification of monitoring results and visualization of detection results of abnormal situations, thereby improving the efficiency and quality of operation and maintenance work; through this self-service inspection system for a fire pipe network, the problem of insufficient manual records can be overcome, the level of detection information can be improved, and the difficulty of detection work can be reduced, which plays a promoting role in timely detection of rail transit tunnel conditions, urban fire inspection work and daily information maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS
[0047] Figure 1 It is a structural schematic diagram of an inspection method based on a fire protection pipe network self-service inspection system of the present invention;
[0048] Figure 2 It is a schematic diagram of the composition of the fire protection pipe network self-service inspection system of the present invention;
[0049] Figure 3 is a schematic diagram of the position of the area array camera of the present invention;
[0050] Figure 4 It is a schematic diagram of the working principle of the intelligent algorithm system of the present invention;
[0051] Figure 5 It is a schematic diagram of the intelligent algorithm flow of the present invention;
[0052] Figure 6 It is a schematic diagram of the BP neural network structure of the present invention;
[0053] Figure 7 It is a BP neural network training flow chart of the present invention;
[0054] Figure 8 It is a schematic diagram of the Bayesian fusion result of the present invention;
[0055] Fig. 9 It is a schematic diagram of the BP neural network classification result of the present invention;
[0056] Figure 1 Middle: 1. Disc; 2. Bottom mounting structure layer; 3. Power supply battery; 4. Middle electrical layer; 5. Display. DETAILED DESCRIPTION
[0057] The present invention will be further described below in conjunction with the accompanying drawings.
[0058] like Figure 1-9 As shown, the inspection method based on the fire protection pipe network self-service inspection system of the present invention comprises the following steps:
[0059] Step (A), designing and purchasing hardware equipment of the fire pipe network self-service inspection system, and assembling the fire pipe network self-service inspection system, wherein the fire pipe network self-service inspection system includes a control unit, a detection unit, a positioning unit, an operating unit and a vehicle platform, the control unit includes but is not limited to a host, a circuit board, a light source controller and an encoder, the detection unit includes but is not limited to an array camera and a stroboscopic light source, the positioning unit uses an RFID positioning tag, the operating unit uses a touch screen, and the vehicle-mounted platform uses a double-track flaw detection trolley;
[0060] Among them, Figure 1 or Figure 3 As shown in the figure, the system uses a double-track rail ultrasonic flaw detector trolley platform as a carrier, and uses the installation position, power interface, lighting, encoder mileage signal and other interfaces provided by the platform, and uses high-definition CCD industrial camera acquisition technology to quickly and clearly capture and save 270° circumferential lining image photos of the pipeline; the system uses six array cameras and seven stroboscopic array light sources, which are evenly distributed and firmly installed on a ring-shaped fixture; the ring-shaped device is fixedly installed at the rear of the vehicle body and protrudes from the vehicle body. The installation height is 2 meters from the rail surface, and the center of the equipment is located at the center of the tunnel, ensuring that the image of the inner wall of the subway tunnel (including the side wall and top surface of the tunnel) can be collected at 270°, and the system can complete clear photography of different viewing distances of the entire pipeline network.
[0061] Step (B), based on the assembled fire pipe network self-service inspection system, design the system's software modules, wherein the system's software modules include an image debugging acquisition module, an image playback module, a puzzle module and an intelligent recognition module, the image debugging acquisition module is used to receive and display the collected data in real time, the image playback module is used to operate historical data, the puzzle module is used to piece together the images collected by six cameras into a complete tunnel surface, and the intelligent recognition module is used to identify the defect type and generate a report;
[0062] Among them, Figure 2 As shown, the hardware program design of the system is written in C language, using the programming software keiluVision5 and the burning software stc-isp-v6.85. The overall modular programming is adopted, which is conducive to the inspection and optimization of subsequent programs. The monitoring software adopts the independently developed pipeline inspection system.
[0063] Step (C), constructing the intelligent algorithm flow of the system according to the software module of the system, wherein the intelligent algorithm flow of the system includes the following steps:
[0064] Among them, Figure 4 or Figure 5 As shown, this method can effectively improve the detection accuracy and lay the foundation for subsequent manual screening;
[0065] Step (C1), collecting and inputting tunnel images, using YOLO to detect foreign objects, and binarizing the bottom ten lines of each image, and comparing the binarized image information with the set threshold. Information fusion is used in collecting and inputting tunnel images, and the information fusion uses Bayesian fusion and BP neural network. The specific steps are as follows:
[0066] Step (C11), obtain Bayesian fusion. In the process of using multiple sensors to measure a certain parameter, there are two random variables, one is the measured parameter μ, and the other is the measurement result x of each sensor. i ,i=1,2,…,m, and both variables are considered to obey normal distribution, as shown in formula (1),
[0067]
[0068] In this way, the conditional probability density of each measurement value is shown in formula (2):
[0069]
[0070] make Obviously, α is a constant and has nothing to do with the measured parameter μ, as shown in formula (3),
[0071]
[0072] From P = (μ|x 1 ,x 2 ,..,x k ) The derivation results show that it still satisfies the normal distribution form. Here, we assume that but Then μ N The value of is shown in formula (4),
[0073]
[0074] The optimal Bayesian fusion of the measured parameters is shown in formula (5):
[0075]
[0076] Step (C11), constructing a BP neural network, wherein the BP neural network is a multi-layer neural network based on an error back propagation algorithm, having a structure of three layers or more, and comprising an input layer, a hidden layer, and an output layer, and each layer is composed of a number of neurons, and there is no connection relationship between the neurons, and the layers are fully connected, and the weights between each connection are different.
[0077] Among them, Figure 6 and Figure 7 As shown in the figure, the training methods of BP neural network are divided into two types: unsupervised learning and supervised learning. The BP algorithm steps involve important parameters such as the number of network layers, the number of hidden layer neurons, the selection of initial weights, and learning errors and rates. These parameters should all be taken into consideration when designing the BP network. Figure 8 and Fig. 9 As shown in the figure, they are the effects of using Bayesian fusion and BP neural network respectively, so that accurate decision-making can be achieved by fusing multi-sensor information;
[0078] Step (C2), if it is higher than the threshold, the current image is determined to be an abnormal image, and then the splicing operation is performed, and the fastener position is located by YOLO, and then the algorithm is used to locate the abnormal image, and the remaining part is subdivided into defects;
[0079] Step (C3), the types of defect detection include crack detection and water leakage detection, and the detection results are output to the defect database.
[0080] Step (D) is to design the system usage process based on the system's intelligent algorithm process, wherein the system usage process includes mechanical assembly, electrical assembly debugging, system commissioning, real-time monitoring interface, image playback interface, intelligent recognition and statistical reports and detection completion. The specific steps are as follows:
[0081] Step (D1), mechanical assembly, after assembling the inspection robot, install it on the vehicle body platform;
[0082] Step (D2), electrical assembly and debugging, connecting various communication interfaces and power interfaces of the equipment;
[0083] Step (D3), system commissioning, open the camera commissioning interface, adjust the camera parameters and check the shooting status, and save the collected data after confirming that the status is good;
[0084] Step (D4), real-time monitoring interface, input basic operation information, start the detection function and start collecting data, check the mileage speed and other information during the operation, and complete the operation within the skylight time;
[0085] Step (D5), image playback interface, view historical images and perform image stitching, then view defective marked images, and compare the two sets of historical data, record and save the differences;
[0086] Step (D6), intelligent identification and statistical report, select defect type and detect defect images, complete the detection and filter out the images that pass the inspection, then process the defect images and generate statistical reports with one click;
[0087] Step (D7), detection is completed, close the software and equipment, and receive the detection.
[0088] Step (E) completes the intelligent inspection of the fire protection pipe network according to the system usage process. The specific steps are as follows:
[0089] Step (E1), using high-definition camera shooting technology to quickly inspect the pipeline network, and the camera shooting position and the light source position cover the tunnel lining in a circumferential direction of not less than 270°, and can complete clear photography of the pipeline network at different viewing distances;
[0090] Step (E2), the system records the location information of the damage and reflects it in the image information, and the location information includes but is not limited to the track direction side and circumferential area position, and multiple longitudinal and circumferential images are synthesized to form a continuous picture;
[0091] Step (E3), the camera must be able to clearly capture water leakage, missing corners, falling blocks, misalignment, cracks of the lining, and the falling off of pipelines in the tunnel. Water leakage includes wet marks, water seepage, dripping, and mud leakage, and can identify faults through system software;
[0092] Step (E4), the system has image display, storage and playback functions, and can compare any two images based on a common mileage benchmark, and the mileage of the image can be adjusted;
[0093] Step (E5), the system playback software has the functions of selecting the mileage range for viewing the continuous images, adjusting the graphics and zooming the images, and can add text annotations, and also includes customizing text and configuring new information such as line name, line type, speed, mileage and annotations, wherein the graphics adjustment items include brightness, contrast, gamma value and saturation;
[0094] Step (E6), the resolution accuracy of the captured image is not less than 1 mm, and the minimum adjustable exposure time is less than 50 μs, and the maximum inspection speed of the system is not less than 10 km / h;
[0095] In step (E7), the system has a storage capacity of no less than 200 kilometers of photographs, and the relevant data can be easily and quickly exported or copied to a ground computer.
[0096] In summary, the inspection method based on the fire pipe network self-service inspection system of the present invention first adopts a dual-track ultrasonic flaw detector trolley platform as a carrier, and takes the single-chip microcomputer STM32F103VET6 as the core, and then uses the installation position, power interface, lighting, encoder mileage signal and other interfaces provided by the platform, and uses the high-definition CCD industrial camera acquisition technology to complete the overall system design; effectively realizes real-time monitoring of the pipeline environment, and adopts Bayesian and BP neural network related fusion technology to realize the identification and classification of monitoring results and visualization of abnormal situation detection results, thereby improving the efficiency and quality of operation and maintenance work; through this fire pipe network self-service inspection system, the problem of insufficient manual records can be overcome, the level of detection information can be improved, the difficulty of detection work can be reduced, and it plays a promoting role in timely detection of rail transit tunnel conditions, urban fire inspection work and daily information maintenance work, and has the advantages of scientific and reasonable methods, strong applicability and good effects.
[0097] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. An inspection method based on a fire protection pipe network self-service inspection system. Features: The following steps are included: Step (A), designing and purchasing hardware equipment of the fire pipe network self-service inspection system, and assembling the fire pipe network self-service inspection system; Step (B), based on the assembled fire pipe network self-service inspection system, design the software module of the system; Step (C), constructing the intelligent algorithm flow of the system according to the software module of the system, wherein the intelligent algorithm flow of the system includes the following steps: Step (C1), collecting and inputting tunnel images, using YOLO to detect foreign objects, and binarizing the bottom ten lines of each image, and comparing the binarized image information with the set threshold. Information fusion is used in collecting and inputting tunnel images, and the information fusion uses Bayesian fusion and BP neural network. The specific steps are as follows: Step (C11), obtain Bayesian fusion. In the process of using multiple sensors to measure a certain parameter, there are two random variables, one is the measured parameter μ, and the other is the measurement result x of each sensor. i ,i=1,2,…,k, and both variables are considered to obey normal distribution, as shown in formula (1), The conditional probability density of each measurement value is shown in formula (2): make Obviously, α is a constant and has nothing to do with the measured parameter μ, as shown in formula (3), From P = (μ|x 1 ,x 2 ,..,x k ) The derivation results show that it still satisfies the normal distribution form. Here, we assume that but Then μ N The value of is shown in formula (4), The optimal Bayesian fusion of the measured parameters is shown in formula (5): Step (C12), constructing a BP neural network, wherein the BP neural network is a multi-layer neural network based on an error back propagation algorithm, has a structure of three or more layers, and is composed of an input layer, a hidden layer, and an output layer, and each layer is composed of a number of neurons, and there is no connection relationship between the neurons, and the layers are fully connected, and the weights between each connection are different; Step (C2), if it is higher than the threshold, the current image is determined to be an abnormal image, and then the splicing operation is performed, and the fastener position is located by YOLO, and then the algorithm is used to locate the abnormal image, and the remaining part is subdivided into defects; Step (C3), the types of defect detection include crack detection and water leakage detection, and the detection results are output to the defect database; Step (D), designing the system usage process based on the system's intelligent algorithm process; Step (E) completes the intelligent inspection of the fire protection pipe network according to the system usage process. The specific steps are as follows: Step (E1), using high-definition camera shooting technology to quickly inspect the pipeline network, and the camera shooting position and the light source position cover the tunnel lining in a circumferential direction of not less than 270°, and can complete clear photography of the pipeline network at different viewing distances; Step (E2), the system records the location information of the damage and reflects it in the image information, and the location information includes but is not limited to the track direction side and circumferential area position, and multiple longitudinal and circumferential images are synthesized to form a continuous picture; Step (E3), the camera must be able to clearly capture water leakage, missing corners, falling blocks, misalignment, cracks of the lining, and the falling off of pipelines in the tunnel. Water leakage includes wet marks, water seepage, dripping, and mud leakage, and can identify faults through system software; Step (E4), the system has image display, storage and playback functions, and can compare any two images based on a common mileage benchmark, and the mileage of the image can be adjusted; Step (E5), the system playback software has the functions of selecting the mileage range for viewing the continuous images, adjusting the graphics and zooming the images, and can add text annotations, and also includes customizing text and configuring new information including line name, line type, speed, mileage and annotations, wherein the graphics adjustment items include brightness, contrast, gamma value and saturation; Step (E6), the resolution accuracy of the captured image is not less than 1 mm, and the minimum adjustable exposure time is less than 50 μs, and the maximum inspection speed of the system is not less than 10 km / h; In step (E7), the system has a storage capacity of no less than 200 kilometers of photographs, and the relevant data can be easily and quickly exported or copied to a ground computer.
2. According to claim 1, a patrol method based on a fire protection pipe network self-service patrol system, Features: Step (A), designing and purchasing hardware equipment of the fire pipe network self-service inspection system, assembling the fire pipe network self-service inspection system, wherein the fire pipe network self-service inspection system includes a control unit, a detection unit, a positioning unit, an operating unit and a vehicle platform, the control unit includes but is not limited to a host, a circuit board, a light source controller and an encoder, the detection unit includes but is not limited to an array camera and a stroboscopic light source, the positioning unit adopts an RFID positioning tag, the operating unit adopts a touch screen, and the vehicle platform adopts a double-track flaw detection trolley.
3. According to claim 2, a patrol inspection method based on a fire protection pipe network self-service patrol inspection system, Features: Step (B), based on the assembled fire pipe network self-service inspection system, design the system's software modules, wherein the system's software modules include an image debugging acquisition module, an image playback module, a puzzle module and an intelligent recognition module, the image debugging acquisition module is used to receive and display the collected data in real time, the image playback module is used to operate on historical data, the puzzle module is used to piece together the images collected by six cameras into a complete tunnel surface, and the intelligent recognition module is used to identify the defect type and generate a report.
4. According to claim 3, a patrol inspection method based on a fire protection pipe network self-service patrol inspection system, Features: Step (D) is to design the system usage process based on the system's intelligent algorithm process, wherein the system usage process includes mechanical assembly, electrical assembly debugging, system commissioning, real-time monitoring interface, image playback interface, intelligent recognition and statistical reports and detection completion. The specific steps are as follows: Step (D1), mechanical assembly, after assembling the inspection robot, install it on the vehicle body platform; Step (D2), electrical assembly and debugging, connecting various communication interfaces and power interfaces of the equipment; Step (D3), system commissioning, open the camera commissioning interface, adjust the camera parameters and check the shooting status, and save the collected data after confirming that the status is good; Step (D4), real-time monitoring interface, input basic operation information, start the detection function and start collecting data, check the mileage and speed information during the operation, and complete the operation within the skylight time; Step (D5), image playback interface, view historical images and perform image stitching, then view defective marked images, and compare the two sets of historical data, record and save the differences; Step (D6), intelligent identification and statistical report, select defect type and detect defect images, complete the detection and filter out the images that pass the inspection, then process the defect images and generate statistical reports with one click; Step (D7), detection is completed, close the software and equipment, and receive the detection.
