CN111178198B - Automatic monitoring method for potential safety hazards of laboratory dangerous goods based on machine vision - Google Patents
Automatic monitoring method for potential safety hazards of laboratory dangerous goods based on machine vision Download PDFInfo
- Publication number
- CN111178198B CN111178198B CN201911321071.0A CN201911321071A CN111178198B CN 111178198 B CN111178198 B CN 111178198B CN 201911321071 A CN201911321071 A CN 201911321071A CN 111178198 B CN111178198 B CN 111178198B
- Authority
- CN
- China
- Prior art keywords
- potential safety
- laboratory
- key frame
- image
- article
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000012544 monitoring process Methods 0.000 title claims abstract description 18
- 238000012937 correction Methods 0.000 claims abstract description 13
- 238000007689 inspection Methods 0.000 claims description 25
- 238000001514 detection method Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 238000011835 investigation Methods 0.000 abstract description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 4
- 239000007789 gas Substances 0.000 description 4
- 239000001301 oxygen Substances 0.000 description 4
- 229910052760 oxygen Inorganic materials 0.000 description 4
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention provides a machine vision-based automatic monitoring, identifying and early warning method for potential safety hazards in laboratory objects, which is used for a potential safety hazard investigation and correction method. According to the method, the appearance characteristics of the articles are extracted from the acquired images, and whether the articles are dangerous articles or not is judged by utilizing a model trained by an article library. If the article is judged to be dangerous article, then the inference engine and the state library are matched one by one to obtain whether the dangerous article has potential safety hazard. If the potential safety hazard exists, the safety library is searched to obtain a safety clause number, whether the judgment is correct is judged by human identification, if the judgment is correct, the correction is ordered, and the monitoring is continued during the correction. If the object is judged to be wrong, the object placing image data is uploaded to an object library for self-learning. The invention has the advantages of real time, wide area, high efficiency and accuracy.
Description
Technical Field
The invention belongs to the technical field of laboratory safety related to safety engineering disciplines, and particularly relates to a method for automatically monitoring and identifying potential safety hazards of dangerous goods in a laboratory by utilizing machine vision, image processing, pattern recognition and expert system technology.
Background
The potential safety hazard of dangerous goods storage in a laboratory essentially belongs to unsafe states of the goods. Traditionally, potential safety hazards have relied on professionals to identify and monitor in the field or through remote video. The recognition capability of professionals depends on their experience accumulation and ability level, and is difficult to achieve in full time, real-time, wide-area, accurate, and efficient.
Disclosure of Invention
The invention aims to provide a machine vision-based automatic monitoring, identifying and early warning method for potential safety hazards in a laboratory, which is used for a potential safety hazard investigation and correction method.
The method comprises the following steps:
step 1: and establishing a security inspection standard database and two target characteristic knowledge graphs. The two target feature knowledge graph libraries are a standard library and a state library respectively. The standard library contains images of dangerous goods in a laboratory; the state library contains images of the laboratory contents in unsafe states.
Step 2: and collecting video stream images of indoor articles, and extracting key frame images.
Step 3: and performing image preprocessing on the key frame image.
Step 4: establishing an algorithm model for image recognition, and extracting features of the key frame images to obtain feature parameters of each article in the key frame images; the characteristic parameters include shape, pose, and position.
Step 5: searching and comparing the object images in each key frame image extracted by the characteristic parameters with images in a standard library; and judging whether each article in the key frame image is a dangerous article according to the shape of the article image.
Step 6: searching and comparing the object images in each key frame image extracted by the characteristic parameters with the images in the state library; and judging whether the position or the posture of each article placed in the key frame image is in an unsafe state according to the position and the posture of the article in the article image.
Step 7: after the potential safety hazard of the dangerous goods is identified, searching unsafe item clause numbers corresponding to the potential safety hazard in a safety inspection standard database, and outputting content description of the potential safety hazard recorded in the safety inspection standard database. Potential safety hazards include the presence of dangerous articles in the laboratory that cannot occur and the presence of articles in the laboratory in an unsafe condition.
Preferably, the security check standard database contains a description of a plurality of numbered security non-conforming items. The image of each typical dangerous goods potential safety hazard in the state library corresponds to the safety non-conforming item clause number and the related content description in the safety inspection standard database.
Preferably, the image sensor for acquiring the video stream image of the indoor article in step 2 adopts a charge coupled device.
Preferably, in the video stream image, a key frame image is acquired every n frames;
preferably, the image preprocessing in step 3 includes image enhancement and image segmentation.
Preferably, in step 4, the characteristic parameter further includes color information; and fifthly, taking the shape and the color of the objects as judgment basis to judge whether each object in the key frame image is a dangerous object.
Preferably, three algorithm models are built in the step 4; the three algorithm models are a shape recognition algorithm, a color recognition algorithm and a position recognition algorithm respectively. The shape recognition algorithm is used for recognizing the shape and the gesture of each object in each key frame image; and the color recognition algorithm is used for extracting color characteristic parameters of each object in each key frame image. And the position recognition algorithm is used for recognizing the relative positions of all objects and the environment in all the key frame images.
Preferably, the shape recognition algorithm adopts a boundary feature method and a Fourier shape description method; the color recognition algorithm employs a color histogram, color set, color moment, or color aggregate vector algorithm. The location recognition algorithm employs a target detection algorithm.
Preferably, after the step 7 is performed, steps 8 to 10 are performed, and the procedure is as follows:
step 8: generating a content description and a security check list of the live photo, wherein the content description and the security check list comprise serial numbers, unsafe item clause numbers and potential safety hazards; the live photo is a key frame image of which the potential safety hazard is checked; the safety inspector judges whether the identified potential safety hazard is correct according to the safety inspection table; if the safety inspector judges that the potential safety hazards recorded in the safety inspection table are correct, a limit correction notice is issued and sent to the safety related personnel in the laboratory, and the potential safety hazards are subjected to correction in a specified period.
Step 9: and continuously monitoring the potential safety hazard of dangerous articles in the laboratory until the modification requirement is met in the modification period, so that the safety check forms a closed loop.
Step 10: and storing the potential safety hazard correction condition of the dangerous goods in the laboratory into a database of a laboratory safety management system.
Preferably, in step 8, if the security inspector determines that the potential safety hazard recorded in the security inspection table is incorrect; the standard library and the state library are updated by deep learning based on the live photos.
The invention has the beneficial effects that:
according to the invention, the key frame image is firstly intercepted in the video shot by the image sensor, and then the key frame image is identified, judged and compared, so that compared with the prior art that safety personnel are relied on to detect the safety of a laboratory, the potential safety hazard can be identified more quickly, the probability of risk occurrence of the laboratory is greatly reduced, and the cost of laboratory safety inspection is reduced.
2. The invention uses a whole set of processes similar to an expert system to realize automatic detection of the safety condition of the laboratory, and overturns the prior need of manually detecting the safety of the laboratory, thereby saving manpower and improving the accuracy.
3. According to the invention, the shape, position, posture and color of each article in a laboratory are identified by an image identification algorithm according to a security check list; all articles in the laboratory can be detected, and the defect that potential safety hazards are left easily due to negligence in manual detection is overcome.
4. The invention can be applied to different laboratories as long as the database is replaced, without the need to specially restart the design in other laboratories.
Drawings
FIG. 1 is a general flow chart of an automatic monitoring method for judging the potential safety hazard of laboratory dangerous goods according to the invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, the automatic monitoring method for the potential safety hazard of the laboratory dangerous goods based on the machine vision comprises the following specific steps:
step 1: based on the laboratory safety management system, a safety inspection standard database and two target characteristic knowledge graphs are established according to dangerous goods types and technical safety requirements of various professional laboratories. The security check criteria database contains a description of a plurality of numbered security non-compliance items. The two target feature knowledge graph libraries are a standard library and a state library respectively. The standard library contains images of typical dangerous goods in a laboratory; the state library contains images of the articles in the laboratory in unsafe states; the image of each typical dangerous goods potential safety hazard in the state library corresponds to the safety non-conforming item clause number and the related content description in the safety inspection standard database.
Step 2: acquiring video stream images of indoor articles by using an image sensor, and acquiring a key frame image every n frames at intervals, wherein n=20; the image sensor employs a charge coupled device (i.e., charge coupled Device, CCD for short). The charge coupled device can acquire an RGB color image, each pixel consisting of three components of red (R) green (G) blue (B), representing a point in the RGB color space.
Step 3: and carrying out image enhancement and image segmentation processing on each key frame image to divide the images of each article in the key frame images.
Step 4: establishing three algorithm models; the three algorithm models are a shape recognition algorithm, a color recognition algorithm and a position recognition algorithm respectively. The shape recognition algorithm is used for recognizing the shape and the gesture of each object in each key frame image; and the color recognition algorithm is used for extracting color characteristic parameters of each object in each key frame image. And the position recognition algorithm is used for recognizing the relative positions of all objects and the environment in all the key frame images. The shape recognition algorithm adopts a boundary characteristic method and a Fourier shape description method; the color recognition algorithm employs a color histogram, color set, color moment, or color aggregate vector algorithm. The location recognition algorithm employs a target detection algorithm.
Step 5: and identifying the shape, the gesture, the position and the color of the object in each key frame image by using a shape identification algorithm, a color identification algorithm and a position identification algorithm. The shape, position, structure and pose of an object are a high-level visual feature in an image, and are intuitive and interpretable, and are often used for describing the properties and information of a target object.
Step 6: searching and comparing the object images in each key frame image extracted by the characteristic parameters with images in a standard library through an inference engine; judging whether each article in the key frame image is a dangerous article or not according to the shape and the color of the article image; if dangerous goods exist in the key frame image, an alarm is sent out, and what dangerous goods are further determined; for example: the image sensor is used for collecting a gas cylinder at a certain position in a laboratory, and the gas cylinder is identified as an oxygen cylinder (combustion-supporting gas) through light blue.
Step 7: searching and comparing the object images in each key frame image extracted by the characteristic parameters with the images in the state library through an inference engine; judging whether the position or the posture of each article placed in the key frame image is in an unsafe state or not according to the position information and the posture information of the articles in the article image; if the articles in the key frame image are in an unsafe state, an alarm is sent out, and the unsafe state of each dangerous article is further determined; for example: whether the identified oxygen cylinder is at a specified position, whether a fixing device is arranged, whether the fixing position is proper, and whether a light green hydrogen cylinder or a brown methane cylinder (combustible gas) is arranged around the oxygen cylinder is mixed with the oxygen cylinder.
Step 8: after the potential safety hazard of the dangerous goods is identified, searching unsafe item clause numbers corresponding to the potential safety hazard in a safety inspection standard database, and outputting content description of the potential safety hazard recorded in the safety inspection standard database. Potential safety hazards include the presence of dangerous articles in the laboratory that cannot occur and the presence of articles in the laboratory in an unsafe condition.
Step 9: generating a content description and a security check list of the live photo, wherein the content description and the security check list comprise serial numbers, unsafe item clause numbers and potential safety hazards; the live photo is a key frame image of which the potential safety hazard is checked; the safety inspector judges whether the identified potential safety hazard is correct according to the safety inspection table; if the safety inspector judges that the potential safety hazards recorded in the safety inspection table are incorrect; the standard library and the state library are updated by deep learning based on the live photos. If the safety inspector judges that the potential safety hazards recorded in the safety inspection table are correct, a limit correction notice is issued and sent to the safety related personnel in the laboratory, and the potential safety hazards are subjected to correction in a specified period.
Step 10: and in the rectifying period, the potential safety hazard of dangerous articles in the laboratory is continuously monitored until the rectifying requirement is met, so that the safety inspection forms a closed loop.
Step 11: the potential safety hazard correction conditions of the dangerous goods in the laboratory are stored in a database of a laboratory safety management system, so that a manager can grasp the safety conditions of the dangerous goods in the laboratory in real time.
Claims (8)
1. The automatic monitoring method for the potential safety hazards of the dangerous goods in the laboratory based on machine vision is characterized by comprising the following steps of: step 1: establishing a security inspection standard database and two target feature knowledge graphs; the two target characteristic knowledge graph libraries are a standard library and a state library respectively; the standard library contains images of dangerous goods in a laboratory; the state library contains images of the articles in the laboratory in unsafe states;
step 2: collecting video stream images of indoor articles, and extracting key frame images;
step 3: performing image preprocessing on the key frame image;
step 4: establishing an algorithm model for image recognition, and extracting features of the key frame images to obtain feature parameters of each article in the key frame images; the characteristic parameters include shape, attitude and position;
step 5: searching and comparing the object images in each key frame image extracted by the characteristic parameters with images in a standard library; judging whether each article in the key frame image is a dangerous article according to the shape of the article image;
step 6: searching and comparing the object images in each key frame image extracted by the characteristic parameters with the images in the state library; judging whether the position or the posture of each article placed in the key frame image is in an unsafe state according to the position and the posture of the article in the article image;
step 7: after the potential safety hazard of the dangerous goods is identified, searching unsafe item clause numbers corresponding to the potential safety hazard in a safety inspection standard database, and outputting content description of the potential safety hazard recorded in the safety inspection standard database; the potential safety hazards comprise dangerous articles which cannot appear in a laboratory and unsafe articles in the laboratory; the safety inspection standard database contains a plurality of numbered descriptions of safety non-conforming items; the images of the potential safety hazards of each typical dangerous article in the state library correspond to the safety non-conforming item clause numbers and related content descriptions in the safety inspection standard database;
step 8: generating a content description and a security check list of the live photo, wherein the content description and the security check list comprise serial numbers, unsafe item clause numbers and potential safety hazards; the live photo is a key frame image of which the potential safety hazard is checked; the safety inspector judges whether the identified potential safety hazard is correct according to the safety inspection table; if the safety inspector judges that the potential safety hazards recorded in the safety inspection table are correct, issuing a limit correction notice, and sending the notice to a safety related personnel in a laboratory, and accounting to complete the correction of the potential safety hazards within a specified period;
step 9: continuously monitoring dangerous goods potential safety hazards in a laboratory within the rectification period until the rectification requirement is met, so that safety inspection forms a closed loop;
step 10: and storing the potential safety hazard correction condition of the dangerous goods in the laboratory into a database of a laboratory safety management system.
2. The automatic monitoring method for potential safety hazards in laboratory based on machine vision according to claim 1, wherein the method comprises the following steps: the image sensor for collecting the video stream image of the indoor article in the step 2 adopts a charge coupled device.
3. The automatic monitoring method for potential safety hazards in laboratory based on machine vision according to claim 1, wherein the method comprises the following steps: in the video stream image, a key frame image is acquired every n frames at intervals.
4. The automatic monitoring method for potential safety hazards in laboratory based on machine vision according to claim 1, wherein the method comprises the following steps: the image preprocessing in the step 3 comprises image enhancement and image segmentation.
5. The automatic monitoring method for potential safety hazards in laboratory based on machine vision according to claim 1, wherein the method comprises the following steps: in step 4, the characteristic parameters further include color information; and fifthly, taking the shape and the color of the objects as judgment basis to judge whether each object in the key frame image is a dangerous object.
6. The automatic monitoring method for potential safety hazards in laboratory based on machine vision according to claim 1, wherein the method comprises the following steps: step 4, three algorithm models are built in total; the three algorithm models are a shape recognition algorithm, a color recognition algorithm and a position recognition algorithm respectively; the shape recognition algorithm is used for recognizing the shape and the gesture of each object in each key frame image; the color recognition algorithm is used for extracting color characteristic parameters of each object in each key frame image; and the position recognition algorithm is used for recognizing the relative positions of all objects and the environment in all the key frame images.
7. The automatic monitoring method for potential safety hazards in laboratory based on machine vision as claimed in claim 6, wherein the method comprises the following steps: the shape recognition algorithm adopts a boundary characteristic method and a Fourier shape description method; the color recognition algorithm adopts a color histogram, a color set, a color moment or a color aggregate vector algorithm; the location recognition algorithm employs a target detection algorithm.
8. The automatic monitoring method for potential safety hazards in laboratory based on machine vision according to claim 1, wherein the method comprises the following steps: in step 8, if the safety inspector judges that the potential safety hazard recorded in the safety inspection table is incorrect; the standard library and the state library are updated by deep learning based on the live photos.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911321071.0A CN111178198B (en) | 2019-12-19 | 2019-12-19 | Automatic monitoring method for potential safety hazards of laboratory dangerous goods based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911321071.0A CN111178198B (en) | 2019-12-19 | 2019-12-19 | Automatic monitoring method for potential safety hazards of laboratory dangerous goods based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111178198A CN111178198A (en) | 2020-05-19 |
| CN111178198B true CN111178198B (en) | 2024-02-09 |
Family
ID=70647333
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911321071.0A Active CN111178198B (en) | 2019-12-19 | 2019-12-19 | Automatic monitoring method for potential safety hazards of laboratory dangerous goods based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111178198B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112488376A (en) * | 2020-11-25 | 2021-03-12 | 航天信息股份有限公司 | Method and system for managing and controlling potential safety hazards of operation site |
| CN112881600A (en) * | 2021-01-11 | 2021-06-01 | 中南民族大学 | Laboratory hazard detection method, laboratory hazard detection equipment, storage medium and laboratory hazard detection device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101413811A (en) * | 2008-11-27 | 2009-04-22 | 河北工业大学 | Autonomous identifying method of hazardous article target |
| CN109074889A (en) * | 2016-02-22 | 2018-12-21 | 拉皮斯坎系统股份有限公司 | System and method for detecting dangerous material and contraband in cargo |
| CN110472922A (en) * | 2018-05-09 | 2019-11-19 | 洛阳华清天木生物科技有限公司 | A kind of management system based on wisdom laboratory safety inspection |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009040217A1 (en) * | 2007-09-24 | 2009-04-02 | Siemens Aktiengesellschaft | Method and device for controlling traffic flows having vehicles transporting hazardous goods, the vehicles moving through a security-critical traffic area of a road network, in particular through a road tunnel |
-
2019
- 2019-12-19 CN CN201911321071.0A patent/CN111178198B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101413811A (en) * | 2008-11-27 | 2009-04-22 | 河北工业大学 | Autonomous identifying method of hazardous article target |
| CN109074889A (en) * | 2016-02-22 | 2018-12-21 | 拉皮斯坎系统股份有限公司 | System and method for detecting dangerous material and contraband in cargo |
| CN110472922A (en) * | 2018-05-09 | 2019-11-19 | 洛阳华清天木生物科技有限公司 | A kind of management system based on wisdom laboratory safety inspection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111178198A (en) | 2020-05-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111259892B (en) | Inspection method, inspection device, inspection equipment and inspection medium for state of indicator lamp | |
| CN110738135B (en) | Method and system for judging and guiding worker operation step standard visual recognition | |
| CN111949809B (en) | Intelligent processing method for infrared inspection data of power transmission line | |
| US20190311470A1 (en) | Apparel production monitoring system using image recognition | |
| CN110414320B (en) | Method and system for safety production supervision | |
| CN109858367B (en) | Visual automatic detection method and system for worker through supporting unsafe behaviors | |
| CN108038424B (en) | Visual automatic detection method suitable for high-altitude operation | |
| CN111178198B (en) | Automatic monitoring method for potential safety hazards of laboratory dangerous goods based on machine vision | |
| CN112088387A (en) | System and method for detecting defects in imaged articles | |
| CN112712093B (en) | Security check image identification method and device, electronic equipment and storage medium | |
| CN112927461B (en) | Early warning decision method and device for charging pile of new energy automobile | |
| CN103970771A (en) | Search method and system for human body | |
| CN115393789A (en) | Crowd gathering early warning method, system, equipment and storage medium for dangerous area of building site | |
| CN111310844B (en) | Vehicle identification model construction method and device and identification method and device | |
| CN109001210B (en) | System and method for detecting aging and cracking of sealing rubber strip of civil air defense door | |
| CN109858829A (en) | A kind of accident risks method for early warning based on block chain | |
| CN115311601A (en) | Fire detection analysis method based on video analysis technology | |
| CN114049543A (en) | Automatic identification method for scrap steel unloading change area based on deep learning | |
| CN117253176B (en) | Safe production Al intelligent detection method based on video analysis and computer vision | |
| CN112465747A (en) | Intelligent identification method for transmission tower hardware corrosion | |
| CN114494103A (en) | Defect detection method and device | |
| CN112149579A (en) | Improved yolo _ v 3-based bird nest hidden danger detection method | |
| CN112052912A (en) | Intelligent flame combustion state identification method for fire-fighting robot | |
| CN113792658B (en) | Automatic control method and system for closed loop of fire operation in whole process intelligent monitoring | |
| CN111723690B (en) | Method and system for monitoring state of circuit equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |