Indoor gas safety inspection auxiliary identification method and system and electronic equipment

Abstract

The invention relates to the field of intelligent assistance of indoor gas safety inspection, discloses an auxiliary identification method for indoor gas safety inspection, and aims to solve the problems of low efficiency and low quality of gas safety inspection. The method comprises the following steps: establishing a standardized inspection process, and inspecting whether gas pipelines, gas meters, gas appliances and ventilation conditions meet the conditions or not; checking the acquired image, judging whether the image meets the requirement or not, and reserving the image meeting the requirement to obtain a third-step image; step 3, judging whether the image in the step 3 has a shooting element or not, reserving the image with the shooting element as a shooting element image, screening out photos with screens of the gas meter and the gas detector, and identifying the reading of the gas meter and the detection condition of the gas detector by utilizing target detection and OCR technologies; and carrying out abnormal condition identification on the shot element image, and sending the gas serial number of which the abnormal condition is identified to an inspector. The system can assist personnel in safety check, and has remarkable advantages.

Description

Technical Field

[0001] The present invention relates to the field of intelligent assistance for indoor gas safety inspection, and in particular to an auxiliary identification method, system and electronic equipment for indoor gas safety inspection. Background Art

[0002] With the acceleration of my country's urbanization process, gas, as a clean and efficient energy source, has been widely used in residents' lives and corporate production. However, gas safety accidents occur from time to time, posing a serious threat to the safety of people's lives and property. Indoor gas safety inspection is an important means to prevent gas accidents, but at present, indoor gas safety inspection mainly relies on manual inspection, and there are the following problems: the level of inspectors is uneven, and some inspectors lack professional knowledge and experience, making it difficult to find potential safety hazards. During the inspection process, some personnel are lazy and cannot ensure the quality of safety inspections. In order to solve the above problems, it is necessary to study an indoor gas safety inspection auxiliary identification method and system to improve the efficiency and quality of gas safety inspections. With the rapid development of artificial intelligence technology, intelligent recognition technology, including image recognition, target detection, and text recognition, has been vigorously developed and applied, and has been widely used in security, medical, automatic driving and other fields. Applying intelligent recognition technology to the field of indoor safety inspections to assist personnel in safety inspections has significant advantages. It can realize the identification and recording of indoor gas safety status, timely discover hidden dangers, improve the efficiency and quality of gas safety inspections, prevent problems before they happen, and protect people's lives and property.

[0003] Therefore, an auxiliary identification method for indoor gas safety inspection is proposed. Summary of the invention

[0004] The indoor gas safety inspection auxiliary identification method, system and electronic equipment of the present invention apply intelligent identification technology to the field of indoor safety inspection to assist personnel in safety inspection, which has significant advantages. It can realize the identification and recording of indoor gas safety status, discover hidden dangers in time, improve the efficiency and quality of gas safety inspection, prevent problems before they occur, and protect people's lives and property.

[0005] A first aspect of the present invention provides an indoor gas safety inspection auxiliary identification method, the method comprising:

[0006] Establish a standardized inspection process to check whether the gas pipelines, gas meters, gas appliances, and ventilation conditions meet the requirements, and take photos to obtain images during the inspection process;

[0007] Based on deep learning image classification technology, the image is checked to determine whether the image meets the requirements. The images that do not meet the requirements are removed, and the images that meet the requirements are retained to obtain the third step image;

[0008] Based on deep learning technology, it is determined whether the image in the third step has the shooting element, and the image without the shooting element is removed, and the image with the shooting element is retained as the shooting element image.

[0009] Based on the captured element images, photos with gas meters and gas detector screens are selected, and the gas meter readings and gas detector detection conditions are identified using target detection and OCR technology;

[0010] The abnormal situation is identified in the captured element image, and the gas number of the identified abnormal situation is sent to the inspector to remind the inspector to conduct a focused inspection.

[0011] In a preferred embodiment, judging whether the image meets the requirements includes three steps of detection, namely, a first step detection, a second step detection and a third step detection;

[0012] The first step detection includes: detecting whether the image is blurry or has insufficient light. If so, the image is judged to be unqualified; if not, the image is retained as the first step image, and then the second step detection is performed on the first step image.

[0013] In a preferred embodiment, performing the second step detection on the first step image includes:

[0014] Determine whether the first step image is a screen shot photo. If it is a screen shot photo, remove it. If it is not a screen shot photo, keep the first step image as the second step image.

[0015] In a preferred embodiment, performing the third step detection on the second step image includes:

[0016] The second-step images are compared with the existing database, and the second-step images with a similarity of 100% are judged as unqualified images and removed, and the initial images with a similarity less than 100% are used as the third-step images.

[0017] In a preferred embodiment, the photographing elements include a gas stove, a gas hose, a gas meter and a gas detector screen; when the third step image includes one of the photographing elements, it is determined that the third step image has the photographing element.

[0018] In a preferred embodiment, identifying abnormal situations on the captured element images includes:

[0019] For the captured element images with the gas meter and the gas detector screen, it is determined whether the gas meter reading and the gas concentration of the gas detector are within the set range, and the situation that is not within the set range is classified as an abnormal situation;

[0020] For the captured element images without gas meter and gas detector screens, determine whether there is pipeline corrosion, whether the gas stove has flameout protection, and whether a standard hose is used in the captured element images. Pipe corrosion, gas stoves without flameout protection, and failure to use standard hoses are all classified as abnormal situations.

[0021] In a preferred embodiment, the indoor gas safety inspection auxiliary identification method further includes:

[0022] Record and save the data and results of each inspection, including inspection time, location, inspectors, problems found and disposal measures.

[0023] A second aspect of the present invention provides an indoor gas safety inspection auxiliary identification system, based on the above-mentioned indoor gas safety inspection auxiliary identification method, the system comprises:

[0024] Process establishment module, used to establish standardized inspection processes;

[0025] The first recognition module is used to check the image based on deep learning image classification technology to determine whether the image meets the requirements, remove the image that does not meet the requirements, and retain the image that meets the requirements to obtain the third step image;

[0026] The second recognition module is used to determine whether the third step image has the shooting element based on the deep learning technology, remove the image without the shooting element, and retain the image with the shooting element as the shooting element image,

[0027] The third recognition module is used to filter out photos with gas meters and gas detector screens based on the captured element images, and use target detection and OCR technology to identify the gas meter readings and the detection conditions of the gas detector;

[0028] The abnormality recognition module is used to identify abnormal situations in the captured element images, and send the gas number of the identified abnormal situation to the inspector to remind the inspector to conduct focused inspection.

[0029] The third aspect of the present invention provides an electronic device, comprising: at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned indoor gas safety inspection auxiliary identification method.

[0030] A fourth aspect of the present invention provides a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to be executed by the computer to implement the above-mentioned indoor gas safety inspection auxiliary identification method.

[0031] At least one of the above technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects:

[0032] The present invention establishes a standardized process so that inspectors can conduct a comprehensive safety inspection of indoor gas safety in accordance with standard requirements, avoiding the risk of missing important inspection items. The present invention strengthens the control of the indoor safety inspection process by intelligently distinguishing image quality and identifying key elements, avoiding the problem of some inspectors being lazy and reporting with blurred photos, duplicate photos, and other photos. The present invention recognizes safety hazards through reading recognition based on artificial intelligence technology, which helps to quickly discover and respond to safety hazards and greatly improves the efficiency and effectiveness of gas safety inspections. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0034] Figure 1 A schematic diagram of an indoor gas safety inspection auxiliary identification method provided in an embodiment of this specification;

[0035] Figure 2 A flowchart of an indoor gas safety inspection auxiliary identification method provided in an embodiment of this specification;

[0036] Figure 3 A schematic diagram of an indoor gas safety inspection auxiliary identification system provided in accordance with an embodiment of this specification. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0038] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0039] Figure 1 An indoor gas safety inspection auxiliary identification method is provided in an embodiment of this specification, see Figure 1 , the method comprising:

[0040] Establish a standardized inspection process to check whether the gas pipelines, gas meters, gas appliances, and ventilation conditions meet the requirements, and take photos to obtain images during the inspection process;

[0041] Based on deep learning image classification technology, the image is checked to determine whether the image meets the requirements. The images that do not meet the requirements are removed, and the images that meet the requirements are retained to obtain the third step image;

[0042] Based on deep learning technology, it is determined whether the image in the third step has the shooting element, and the image without the shooting element is removed, and the image with the shooting element is retained as the shooting element image.

[0043] Based on the captured element images, photos with gas meters and gas detector screens are selected, and the gas meter readings and gas detector detection conditions are identified using target detection and OCR technology;

[0044] The abnormal situation is identified in the captured element image, and the gas number of the identified abnormal situation is sent to the inspector to remind the inspector to conduct a focused inspection.

[0045] In order to specifically illustrate the indoor gas safety inspection auxiliary identification method of the present application, the following is combined with Figure 2 The method is specifically described as follows:

[0046] Establish a standardized inspection process to check whether the gas pipelines, gas meters, gas appliances, and ventilation conditions meet the requirements, and take photos to obtain images during the inspection process;

[0047] Based on deep learning image classification technology, the image is checked to determine whether the image meets the requirements. The images that do not meet the requirements are removed, and the images that meet the requirements are retained to obtain the third step image;

[0048] In this embodiment, judging whether the image meets the requirements includes three steps of detection, namely, the first step detection, the second step detection and the third step detection;

[0049] The first step detection includes: detecting whether the image is blurry or has insufficient light. If so, the image is judged to be unqualified; if not, the image is retained as the first step image, and then the second step detection is performed on the first step image.

[0050] The second step detection of the first step image includes: determining whether the first step image is a screen shot photo, if it is a screen shot photo, removing it, if it is not a screen shot photo, retaining the first step image as the second step image.

[0051] The third step detection of the second step image includes: comparing the second step image with the existing database, judging the second step image with a similarity of 100% as an unqualified image and removing it, and the initial image with a similarity less than 100% as the third step image.

[0052] Based on deep learning technology, determine whether the image in the third step has shooting elements, remove the image without shooting elements, and retain the image with shooting elements as the shooting element image;

[0053] In this embodiment, the photographing elements include a gas stove, a gas hose, a gas meter and a gas detector screen; when the third step image includes one of the photographing elements, it is determined that the third step image has the photographing element.

[0054] Based on the captured element images, photos with gas meters and gas detector screens are selected, and the gas meter readings and gas detector detection conditions are identified using target detection and OCR technology;

[0055] The abnormal situation is identified in the captured element image, and the gas number of the identified abnormal situation is sent to the inspector to remind the inspector to conduct a focused inspection.

[0056] Identifying abnormal situations in captured element images includes:

[0057] For the captured element images with the gas meter and the gas detector screen, it is determined whether the gas meter reading and the gas concentration of the gas detector are within the set range, and the situation that is not within the set range is classified as an abnormal situation;

[0058] For the captured element images without gas meter and gas detector screens, determine whether there is pipeline corrosion, whether the gas stove has flameout protection, and whether a standard hose is used in the captured element images. Pipe corrosion, gas stoves without flameout protection, and failure to use standard hoses are all classified as abnormal situations.

[0059] The indoor gas safety inspection auxiliary identification method of the present invention also includes: recording and saving the data and results of each inspection, including the inspection time, location, inspectors, problems found and disposal measures.

[0060] The second aspect of the present invention further provides an indoor gas safety inspection auxiliary identification system, which is based on the indoor gas safety inspection auxiliary identification method, and the system includes:

[0061] Process establishment module, used to establish standardized inspection processes;

[0062] The first recognition module is used to check the image based on deep learning image classification technology to determine whether the image meets the requirements, remove the image that does not meet the requirements, and retain the image that meets the requirements to obtain the third step image;

[0063] The second recognition module is used to determine whether the third step image has the shooting element based on the deep learning technology, remove the image without the shooting element, and retain the image with the shooting element as the shooting element image,

[0064] The third recognition module is used to filter out photos with gas meters and gas detector screens based on the captured element images, and use target detection and OCR technology to identify the gas meter readings and the detection conditions of the gas detector;

[0065] The abnormality recognition module is used to identify abnormal situations in the captured element images, and send the gas number of the identified abnormal situation to the inspector to remind the inspector to conduct focused inspection.

[0066] The indoor gas safety inspection auxiliary identification system of the present invention also provides an example, such as Figure 3 As shown, it includes standardized process configuration module, mobile inspector inspection module, image quality analysis module, intelligent analysis module, and data management module:

[0067] Standardized process configuration module: used to customize the inspection process and set inspection standards for inspectors, including inspection order, inspection content, whether to take photos of the scene, photo recognition association algorithm, recognition content, and operations that do not meet recognition requirements.

[0068] Mobile inspector inspection module: used by inspectors to conduct indoor inspections, including login, inspection, inspection record query, system settings and other functional modules. The inspection module loads the inspection process, and the inspector completes the inspection item by item according to the system prompts, takes photos of the scene as required, retakes photos or proceeds to the next step according to the photo inspection results, and records and handles hidden dangers according to the system prompts.

[0069] Image quality analysis module: includes image repeatability inspection and image quality judgment. It is used to judge the image quality and automatically analyze the quality of the photos taken by the inspectors on site in terms of clarity, angle, light, etc. by using deep learning-based image classification technology, and returns the results of whether the photos are qualified to the mobile inspector inspection module.

[0070] Intelligent analysis module: used to perform corresponding intelligent recognition of photos. According to the analysis request and photos sent by the mobile inspector inspection module, it performs corresponding intelligent analysis, including key element recognition, reading recognition, and safety hazard recognition functions, and returns the photo recognition results to the mobile inspector inspection module.

[0071] Data management module: used for the management, statistics and analysis of inspection records, providing data support for managers to conduct subsequent safety management and risk assessment, and assisting decision-making.

[0072] The third aspect of the present invention provides an electronic device, comprising: at least one processor; and a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the above-mentioned indoor gas safety inspection auxiliary identification method.

[0073] A fourth aspect of the present invention provides a computer-readable storage medium storing computer instructions, wherein the computer instructions are used to be executed by the computer to implement the above-mentioned indoor gas safety inspection auxiliary identification method.

[0074] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. An indoor gas safety inspection auxiliary identification method, characterized in that: The method comprises: Establish a standardized inspection process to check whether the gas pipelines, gas meters, gas appliances, and ventilation conditions meet the requirements, and take photos to obtain images during the inspection process; Based on deep learning image classification technology, the image is checked to determine whether the image meets the requirements. The images that do not meet the requirements are removed, and the images that meet the requirements are retained to obtain the third step image; Based on deep learning technology, determine whether the image in the third step has shooting elements, remove the image without shooting elements, and retain the image with shooting elements as the shooting element image; Based on the captured element images, photos with gas meters and gas detector screens are selected, and the gas meter readings and gas detector detection conditions are identified using target detection and OCR technology; The abnormal situation is identified in the captured element image, and the gas number of the identified abnormal situation is sent to the inspector to remind the inspector to conduct a focused inspection.

2. The indoor gas safety inspection auxiliary identification method according to claim 1 is characterized in that: Judging whether an image meets the requirements includes three steps of detection, namely the first step detection, the second step detection and the third step detection; The first step detection includes: detecting whether the image is blurry or has insufficient light. If so, the image is judged to be unqualified; if not, the image is retained as the first step image, and then the second step detection is performed on the first step image.

3. The indoor gas safety inspection auxiliary identification method according to claim 2 is characterized in that: The second step of detecting the first step image includes: Determine whether the first step image is a screen shot photo. If it is a screen shot photo, remove it. If it is not a screen shot photo, keep the first step image as the second step image.

4. The indoor gas safety inspection auxiliary identification method according to claim 3 is characterized in that: The third step of detecting the second step image includes: The second-step images are compared with the existing database, and the second-step images with a similarity of 100% are judged as unqualified images and removed, and the initial images with a similarity less than 100% are used as the third-step images.

5. The indoor gas safety inspection auxiliary identification method according to claim 4 is characterized in that: The photographing elements include a gas stove, a gas hose, a gas meter and a gas detector screen; when the third step image includes one of the photographing elements, it is determined that the third step image has the photographing element.

6. The indoor gas safety inspection auxiliary identification method according to claim 5 is characterized in that: Identifying abnormal situations in captured element images includes: For the captured element images with the gas meter and the gas detector screen, it is determined whether the gas meter reading and the gas concentration of the gas detector are within the set range, and the situation that is not within the set range is classified as an abnormal situation; For the captured element images without gas meter and gas detector screens, determine whether there is pipeline corrosion, whether the gas stove has flameout protection, and whether a standard hose is used in the captured element images. Pipe corrosion, gas stoves without flameout protection, and failure to use standard hoses are all classified as abnormal situations.

7. The indoor gas safety inspection auxiliary identification method according to claim 6 is characterized in that: The indoor gas safety inspection auxiliary identification method also includes: Record and save the data and results of each inspection, including inspection time, location, inspectors, problems found and disposal measures.

8. An indoor gas safety inspection auxiliary identification system, based on the indoor gas safety inspection auxiliary identification method according to any one of claims 1 to 7, characterized in that: The system comprises: Process establishment module, used to establish standardized inspection processes; The first recognition module is used to check the image based on deep learning image classification technology to determine whether the image meets the requirements, remove the image that does not meet the requirements, and retain the image that meets the requirements to obtain the third step image; The second recognition module is used to determine whether the third step image has the shooting element based on the deep learning technology, remove the image without the shooting element, and retain the image with the shooting element as the shooting element image, The third recognition module is used to filter out photos with gas meters and gas detector screens based on the captured element images, and use target detection and OCR technology to identify the gas meter readings and the detection conditions of the gas detector; The abnormality recognition module is used to identify abnormal situations in the captured element images, and send the gas number of the identified abnormal situation to the inspector to remind the inspector to conduct focused inspection.

9. An electronic device, characterized in that: include: at least one processor; And a memory communicatively connected to at least one of the processors; wherein the memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to implement the indoor gas safety inspection auxiliary identification method described in claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to be executed by the computer to implement the indoor gas safety inspection auxiliary identification method described in claims 1-7.

Images