CN113052140A - Video-based substation personnel and vehicle violation detection method and system - Google Patents
Video-based substation personnel and vehicle violation detection method and system Download PDFInfo
- Publication number
- CN113052140A CN113052140A CN202110450263.2A CN202110450263A CN113052140A CN 113052140 A CN113052140 A CN 113052140A CN 202110450263 A CN202110450263 A CN 202110450263A CN 113052140 A CN113052140 A CN 113052140A
- Authority
- CN
- China
- Prior art keywords
- personnel
- detection
- vehicle
- person
- violation
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 87
- 235000019504 cigarettes Nutrition 0.000 claims abstract description 28
- 238000012544 monitoring process Methods 0.000 claims description 7
- 230000006399 behavior Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 238000010008 shearing Methods 0.000 claims description 2
- 238000000034 method Methods 0.000 abstract description 15
- 230000000391 smoking effect Effects 0.000 abstract description 9
- 238000013135 deep learning Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000012855 volatile organic compound Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Alarm Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for detecting personnel and vehicle violation of transformer substation based on video, belonging to the technical field of image detection and comprising the following steps: s1: personnel and vehicle target detection; s2: personnel and vehicle location violation detection; s3: cutting a person image; s4: safety helmet, cigarette detection and frock classification. The invention adopts a cascading method to solve the violation problem of the transformer substation, uses a real-time detector at the first level, specifically analyzes each specific problem at the later level, and adopts a method combining classification and detection, so that the precision can be greatly improved on the premise of sacrificing a small amount of speed; the problem of violation of tooling of personnel is solved by classification, so that the manual data marking amount can be reduced, more tooling data can be covered, the false detection rate is reduced, and meanwhile, the classification can ensure that each personnel can be divided into normal tooling and non-worn tooling certainly, and the missing detection rate is reduced; the personnel images are cut out independently for smoking and safety helmet detection, and the detection precision is improved.
Description
Technical Field
The invention relates to the technical field of image detection, in particular to a method and a system for detecting personnel and vehicles violation of a transformer substation based on video.
Background
Equipment operation in the transformer substation is complex, dangerous areas are many, and the illegal operation of people and vehicles, such as no tool, no safety helmet, smoking, line crossing break-in, vehicle break-and-stop, etc., is very likely to cause electric power system accidents, and casualties and property loss are caused. The current video monitoring system is mainly used for obtaining evidence afterwards and cannot give an alarm in advance.
At present, the violation detection of the transformer substation is mainly solved through general target detection, each frame of an original image is detected by using a trained deep learning network, and generally labeled types include persons without tools, heads without safety helmets, cigarettes and cigarette ends. The general target detection network model based on deep learning is simple in training, can detect violation of vehicles of partial personnel, has good robustness, but needs to be based on a large amount of manually marked data, and can cause false detection and missed detection if training data is lacked. For example, the types of tools are various, a large amount of strongly supervised data needs to be marked, a large amount of labor is consumed, and false detection may be caused if a tool which is not found in training data occurs in actual use. For example, the cigarette butts have small targets in the monitoring video, so that the detection omission is easily caused. Complicated and various devices in the transformer substation, such as wires, railings and the like, are also easily mistakenly detected into cigarette ends. Therefore, a transformer substation personnel and vehicle violation detection method and system based on videos are provided.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: how to utilize videos to timely detect and identify violation behaviors of transformer substation personnel and prevent dangerous behaviors, and the method for detecting the violation behaviors of the transformer substation personnel and vehicles based on the videos is provided.
The invention solves the technical problems through the following technical scheme, and the invention comprises the following steps:
s1: person and vehicle object detection
Carrying out cascade detection on each frame of an original video, and carrying out personnel and vehicle detection by using a first-stage detector;
s2: personnel and vehicle location violation detection
Determining whether line crossing break-in and illegal parking exist in personnel and vehicles by combining preset position information of the monitoring camera;
s3: person image cutting
Cutting out an image of a person according to the rectangular frame generated by the first-stage detector;
s4: safety helmet and cigarette detection and tool classification
Classifying the personnel images obtained by cutting in the step S3 by using a tool classifier, and classifying the personnel images into normal tools and non-worn tools; and simultaneously, detecting the cigarettes and the safety helmet by using a second-level detector, judging that the personnel violate the rules if the cigarettes are detected or the safety helmet is not detected, and judging that the personnel do not violate the rules if the cigarettes are not detected or the safety helmet is detected.
Further, in step S1, the first-stage detector selects a real-time object detection network, and the data set of the real-time object detection network selects an open data set including a human vehicle.
Further, in the step S2, the positions of the person and the vehicle in the video image are determined by the rectangular frame obtained by the first-level detector, an forbidden area and an illegal parking area are defined in the video image in advance by combining with the camera preset bit information, and whether the person and the vehicle are in the corresponding areas is determined by comparing, so as to determine whether the person crosses the line and whether the vehicle is illegal.
Further, in the step S3, the image of the person is cut out from the original image according to the rectangular frame of the person obtained by the first-level detector, and the rectangular frame of the person is enlarged by ten percent when the person is cut out.
Further, in the step S4, the tool classifier selects a fine-grained classification network.
Further, in the step S4, it is not necessary to detect each frame of the original video, and only the frames including people detected by the first-stage detector need to be processed.
The invention also provides a video-based transformer substation personnel and vehicle violation detection system, which adopts the violation detection method to detect personnel violation behaviors and comprises the following steps:
the first detection module is used for carrying out cascade detection on each frame of the original video and utilizing the first-stage detector to carry out personnel and vehicle detection;
the second detection module is used for determining whether line crossing intrusion and illegal parking exist in the personnel and the vehicles by combining preset position information of the monitoring camera;
the image cutting module is used for cutting out the image of the person according to the rectangular frame generated by the first-level detector;
the third detection module is used for classifying the personnel images obtained by cutting and shearing by using the tool classifier, and dividing the personnel images into normal tools and non-worn tools; meanwhile, cigarette and safety helmet detection is carried out on the image of the person by using a second-level detector, if the cigarette is detected or the safety helmet is not detected, the violation of the rules of the person is judged, and if the cigarette is not detected or the safety helmet is detected, the violation of the rules of the person is judged;
the central processing module is used for sending instructions to other modules to complete related actions;
the first detection module, the second detection module, the image cutting module and the third detection module are all electrically connected with the central processing module.
Compared with the prior art, the invention has the following advantages: according to the video-based transformer substation personnel and vehicle violation detection method, transformer substation violation problems are solved by adopting a cascading method, a real-time detector is used at the first level, each specific problem is specifically analyzed at the later level, and the precision can be greatly improved on the premise of sacrificing a small amount of speed by adopting a method of combining classification and detection; the problem of violation of tooling of personnel is solved by classification, so that the manual data marking amount can be reduced, more tooling data can be covered, the false detection rate is reduced, and meanwhile, the classification can ensure that each personnel can be divided into normal tooling and non-worn tooling certainly, and the missing detection rate is reduced; the personnel image is cut out separately for smoking and safety helmet detection, the proportion of small objects in the image can be amplified, and redundant irrelevant background information is removed, so that the detection precision is improved, and the method is worthy of being popularized and used.
Drawings
FIG. 1 is a schematic flow chart in an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating the determination of contraband and violation in the embodiment of the present invention;
FIG. 3 is a schematic diagram of cropping of a person portion of an image in an embodiment of the invention.
Detailed Description
The following examples are given for the detailed implementation and specific operation of the present invention, but the scope of the present invention is not limited to the following examples.
As shown in fig. 1, the present embodiment provides a technical solution: a transformer substation personnel and vehicle violation detection method based on videos comprises the following steps:
the method comprises the following steps: carrying out cascade detection on each frame of an original video, and carrying out personnel and vehicle detection by using a first-stage detector;
in the first step, the first-level people-vehicle detector needs to process each frame of the original video and needs to ensure the efficiency of the network, so that a real-time target detection network is selected. The personnel vehicle has a large number of public data sets that can be used without additional manual labeling.
The target detection network can be a network such as YOLO, SSD, FASTER-RCNN, Center-Net, etc. Detection of personnel there are a number of public data sets, such as COCO, VOCs, etc., that contain this category of personnel. The personnel data set is mainly annotated with the coordinates of the top left and bottom right corners of the bounding box of the person. Ready data can be trained.
Step two: determining whether line crossing break-in and illegal parking exist in personnel and vehicles by combining preset position information of the monitoring camera;
the specific process of the second step is as follows:
the positions of the personnel and the vehicle in the video image can be determined through the rectangular frame obtained by the man-vehicle detector (the position information is the coordinate of the upper left corner and the lower right corner of the personnel surrounding frame in the image, and the corresponding coordinate is the position of the personnel), an forbidden region and an illegal parking region are defined in the video image in advance by combining the preset position information of the camera, and whether the personnel and the vehicle are in the regions is judged by comparison, so that whether the personnel cross the lines or not is judged, and whether the vehicle is illegal or not is judged. This part has little extra computation time overhead.
In step two, as shown in fig. 2, the shaded area in the drawing is a well-defined forbidden area, and when the intersection range of the detection frame (enclosure frame) area of the person or vehicle (represented by the person in the drawing) and the forbidden or forbidden area is divided by the area of the person or vehicle >0.5 (the value can be adjusted as required), it is considered that the boundary crossing occurs, and the determination formula is as follows:
step three: cutting out images of the personnel according to the rectangular frame generated by the human-vehicle detector;
in step three, the image of the person is cut out from the original image according to the rectangular frame obtained by the first-level people-vehicle detector. When in cutting, the rectangular frame is expanded by ten percent, so that the safety helmet on the head and the cigarettes on the hands are prevented from being lost;
as shown in fig. 3, the personnel are cut out from the original image, so that the interference of the background in the substation can be effectively removed, the cut image can be classified and detected, the details of the tool can be enlarged, the occupation ratio of small objects such as cigarette ends in the image can be improved, and the detection and classification precision can be further improved.
Step four: a second level of detection and classification is performed. The classifier classifies images of personnel, divides the images into normal and non-worn tools, simultaneously detects cigarettes and safety helmets by using the detector, and judges that the personnel violate rules if the cigarettes are detected or the safety helmets are not detected.
In the fourth step, the tool classifier selects a fine-grained classification network, and the fine-grained classification network is mainly used for distinguishing fine images (such as the types of clearly distinguished birds and the types of vehicles), and is particularly suitable for tool classification scenes. Most of the existing fine-grained classification networks are trained by weakly supervised data, only photos of tools and non-tools need to be collected, and rectangular frames do not need to be marked manually, so that the difficulty in manufacturing data sets is very low.
B-CNN, DFL-CNN and the like are used for the fine-grained classification network, the network extracts the characteristics of people and classifies the people by softmax, and the characteristic extraction is learned through a tool data set. The tool data set marking process divides people into two categories, wherein the tool is normally 0, and the tool which is not worn is 1.
The smoking-headgear detector need not detect every frame of the original video, only the frames of people detected by the first-level detector need to be processed. In addition, data information of the picture cut in the third step is relatively reduced, which is beneficial to improving the detection efficiency, so that the smoking and safety helmet detector adopts a slightly complex point network to ensure the detection accuracy;
it should be noted that the safety helmet and the smoking detection network are the same as the human detection network, and if the surrounding frame of the safety helmet does not appear in the surrounding frame of the human, it is determined that the safety helmet is not worn. If the surrounding frame of the cigarette end appears in the surrounding frame of the person, the person judges that the cigarette is smoked.
In the aspect of data set marking, firstly, people in the original data set are cut by a people and vehicle detector, and safety caps and cigarettes are marked in the cut pictures. And training the deep learning neural network by using the labeled data set, wherein the trained deep learning neural network can be used for detecting safety helmets and cigarettes. If the smoking safety helmet detector detects cigarettes, the situation that smoking violation occurs to personnel is judged; and if the smoking safety helmet detector does not detect the safety helmet, judging that the personnel has the illegal action of not wearing the safety helmet.
To sum up, the transformer substation personnel and vehicle violation detection method based on video in the embodiment adopts a cascading method to solve the transformer substation violation problem, uses a real-time detector at the first level, specifically analyzes each specific problem at the later level, and adopts a method combining classification and detection, so that the precision can be greatly improved on the premise of sacrificing a small amount of speed; the problem of violation of tooling of personnel is solved by classification, so that the manual data marking amount can be reduced, more tooling data can be covered, the false detection rate is reduced, and meanwhile, the classification can ensure that each personnel can be divided into normal tooling and non-worn tooling certainly, and the missing detection rate is reduced; the personnel image is cut out separately for smoking and safety helmet detection, the proportion of small objects in the image can be amplified, and redundant irrelevant background information is removed, so that the detection precision is improved, and the method is worthy of being popularized and used.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (7)
1. A transformer substation personnel and vehicle violation detection method based on videos is characterized by comprising the following steps:
s1: person and vehicle object detection
Carrying out cascade detection on each frame of an original video, and carrying out personnel and vehicle detection by using a first-stage detector;
s2: personnel and vehicle location violation detection
Determining whether line crossing break-in and illegal parking exist in personnel and vehicles by combining preset position information of the monitoring camera;
s3: person image cutting
Cutting out an image of a person according to the rectangular frame generated by the first-stage detector;
s4: safety helmet and cigarette detection and tool classification
Classifying the personnel images obtained by cutting in the step S3 by using a tool classifier, and classifying the personnel images into normal tools and non-worn tools; and simultaneously, detecting the cigarettes and the safety helmet by using a second-level detector, judging that the personnel violate the rules if the cigarettes are detected or the safety helmet is not detected, and judging that the personnel do not violate the rules if the cigarettes are not detected or the safety helmet is detected.
2. The video-based substation personnel and vehicle violation detection method according to claim 1, characterized in that: in step S1, the first-stage detector selects a real-time target detection network, and the data set of the real-time target detection network is trained with an open data set including a human vehicle.
3. The video-based substation personnel and vehicle violation detection method according to claim 2, characterized in that: in step S2, the positions of the person and the vehicle in the video image are determined by the rectangular frame obtained by the first-level detector, an forbidden area and an illegal parking area are defined in the video image in advance by combining with the camera preset bit information, and whether the person and the vehicle are in the corresponding areas is determined by comparing, so as to determine whether the person crosses the line and whether the vehicle is illegal.
4. The video-based substation personnel and vehicle violation detection method according to claim 3, characterized in that: in step S3, the image of the person is cut out from the original image according to the rectangular frame of the person obtained by the first-level detector, and the rectangular frame of the person is enlarged by ten percent when the image is cut out.
5. The video-based substation personnel and vehicle violation detection method according to claim 3, characterized in that: in step S4, the tool classifier selects a fine-grained classification network.
6. The video-based substation personnel and vehicle violation detection method according to claim 3, characterized in that: in step S4, it is not necessary to detect each frame of the original video, and only the frames containing people detected by the first-level detector need to be processed.
7. A video-based transformer substation personnel and vehicle violation detection system is characterized in that the violation detection method of any one of claims 1-6 is adopted to detect personnel violation behaviors, and comprises the following steps:
the first detection module is used for carrying out cascade detection on each frame of the original video and utilizing the first-stage detector to carry out personnel and vehicle detection;
the second detection module is used for determining whether line crossing intrusion and illegal parking exist in the personnel and the vehicles by combining preset position information of the monitoring camera;
the image cutting module is used for cutting out the image of the person according to the rectangular frame generated by the first-level detector;
the third detection module is used for classifying the personnel images obtained by cutting and shearing by using the tool classifier, and dividing the personnel images into normal tools and non-worn tools; meanwhile, cigarette and safety helmet detection is carried out on the image of the person by using a second-level detector, if the cigarette is detected or the safety helmet is not detected, the violation of the rules of the person is judged, and if the cigarette is not detected or the safety helmet is detected, the violation of the rules of the person is judged;
the central processing module is used for sending instructions to other modules to complete related actions;
the first detection module, the second detection module, the image cutting module and the third detection module are all electrically connected with the central processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110450263.2A CN113052140A (en) | 2021-04-25 | 2021-04-25 | Video-based substation personnel and vehicle violation detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110450263.2A CN113052140A (en) | 2021-04-25 | 2021-04-25 | Video-based substation personnel and vehicle violation detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113052140A true CN113052140A (en) | 2021-06-29 |
Family
ID=76520562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110450263.2A Pending CN113052140A (en) | 2021-04-25 | 2021-04-25 | Video-based substation personnel and vehicle violation detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113052140A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115393905A (en) * | 2022-11-01 | 2022-11-25 | 合肥中科类脑智能技术有限公司 | Helmet wearing detection method based on attitude correction |
| CN115497030A (en) * | 2022-10-27 | 2022-12-20 | 中国水利水电科学研究院 | Fishing behavior identification method based on deep learning |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109086792A (en) * | 2018-06-26 | 2018-12-25 | 上海理工大学 | Based on the fine granularity image classification method for detecting and identifying the network architecture |
| CN109800665A (en) * | 2018-12-28 | 2019-05-24 | 广州粤建三和软件股份有限公司 | A kind of Human bodys' response method, system and storage medium |
| CN110991315A (en) * | 2019-11-28 | 2020-04-10 | 江苏电力信息技术有限公司 | Method for detecting wearing state of safety helmet in real time based on deep learning |
| CN111222420A (en) * | 2019-12-24 | 2020-06-02 | 重庆市通信产业服务有限公司 | FTP protocol-based low-bandwidth-requirement helmet identification method |
| CN111428641A (en) * | 2020-03-24 | 2020-07-17 | 深圳供电局有限公司 | Secure dressing detection method and device, computer equipment and readable storage medium |
| CN111680610A (en) * | 2020-06-03 | 2020-09-18 | 合肥中科类脑智能技术有限公司 | Construction scene abnormity monitoring method and device |
| CN112052804A (en) * | 2020-09-10 | 2020-12-08 | 公安部第三研究所 | Video intelligent analysis and alarm system and method for realizing safety management |
| CN112183472A (en) * | 2020-10-28 | 2021-01-05 | 西安交通大学 | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet |
| CN112464797A (en) * | 2020-11-25 | 2021-03-09 | 创新奇智(成都)科技有限公司 | Smoking behavior detection method and device, storage medium and electronic equipment |
-
2021
- 2021-04-25 CN CN202110450263.2A patent/CN113052140A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109086792A (en) * | 2018-06-26 | 2018-12-25 | 上海理工大学 | Based on the fine granularity image classification method for detecting and identifying the network architecture |
| CN109800665A (en) * | 2018-12-28 | 2019-05-24 | 广州粤建三和软件股份有限公司 | A kind of Human bodys' response method, system and storage medium |
| CN110991315A (en) * | 2019-11-28 | 2020-04-10 | 江苏电力信息技术有限公司 | Method for detecting wearing state of safety helmet in real time based on deep learning |
| CN111222420A (en) * | 2019-12-24 | 2020-06-02 | 重庆市通信产业服务有限公司 | FTP protocol-based low-bandwidth-requirement helmet identification method |
| CN111428641A (en) * | 2020-03-24 | 2020-07-17 | 深圳供电局有限公司 | Secure dressing detection method and device, computer equipment and readable storage medium |
| CN111680610A (en) * | 2020-06-03 | 2020-09-18 | 合肥中科类脑智能技术有限公司 | Construction scene abnormity monitoring method and device |
| CN112052804A (en) * | 2020-09-10 | 2020-12-08 | 公安部第三研究所 | Video intelligent analysis and alarm system and method for realizing safety management |
| CN112183472A (en) * | 2020-10-28 | 2021-01-05 | 西安交通大学 | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet |
| CN112464797A (en) * | 2020-11-25 | 2021-03-09 | 创新奇智(成都)科技有限公司 | Smoking behavior detection method and device, storage medium and electronic equipment |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115497030A (en) * | 2022-10-27 | 2022-12-20 | 中国水利水电科学研究院 | Fishing behavior identification method based on deep learning |
| CN115393905A (en) * | 2022-11-01 | 2022-11-25 | 合肥中科类脑智能技术有限公司 | Helmet wearing detection method based on attitude correction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108053427B (en) | Improved multi-target tracking method, system and device based on KCF and Kalman | |
| CN108062349B (en) | Video monitoring method and system based on video structured data and deep learning | |
| CN111967393B (en) | Safety helmet wearing detection method based on improved YOLOv4 | |
| Gong et al. | A real-time fire detection method from video with multifeature fusion | |
| CN108009473B (en) | Video structuralization processing method, system and storage device based on target behavior attribute | |
| CN108052859B (en) | Abnormal behavior detection method, system and device based on clustering optical flow characteristics | |
| CN102799873B (en) | Human body abnormal behavior recognition method | |
| CN108629282A (en) | A kind of smoking detection method, storage medium and computer | |
| CN113052140A (en) | Video-based substation personnel and vehicle violation detection method and system | |
| CN112906481A (en) | Method for realizing forest fire detection based on unmanned aerial vehicle | |
| CN103996045B (en) | A kind of smog recognition methods of the various features fusion based on video | |
| CN111222478A (en) | Construction site safety protection detection method and system | |
| CN112115775A (en) | Smoking behavior detection method based on computer vision in monitoring scene | |
| Gunawaardena et al. | Computer vision based fire alarming system | |
| CN112183472A (en) | Method for detecting whether test field personnel wear work clothes or not based on improved RetinaNet | |
| CN114782897A (en) | Dangerous behavior detection method and system based on machine vision and deep learning | |
| CN112711996A (en) | System for detecting occupancy of fire fighting access | |
| CN111695432A (en) | Artificial intelligent face abnormity detection system and method under video monitoring scene | |
| CN111832450B (en) | Knife holding detection method based on image recognition | |
| CN113887372A (en) | Target aggregation detection method and computer-readable storage medium | |
| CN113657250A (en) | Flame detection method and system based on monitoring video | |
| CN113392706A (en) | Device and method for detecting smoking and using mobile phone behaviors | |
| KR101552564B1 (en) | Fusion security system based on gas sensor and IP network camera | |
| CN117475353A (en) | Video-based abnormal smoke identification method and system | |
| CN117197713A (en) | Extraction method based on digital video monitoring system |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210629 |