CN112307896A - Method for detecting lewd behavior abnormity of elevator under community monitoring scene - Google Patents
Method for detecting lewd behavior abnormity of elevator under community monitoring scene Download PDFInfo
- Publication number
- CN112307896A CN112307896A CN202011029847.4A CN202011029847A CN112307896A CN 112307896 A CN112307896 A CN 112307896A CN 202011029847 A CN202011029847 A CN 202011029847A CN 112307896 A CN112307896 A CN 112307896A
- Authority
- CN
- China
- Prior art keywords
- elevator
- community
- detecting
- lewd
- passenger
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012544 monitoring process Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 58
- 230000006399 behavior Effects 0.000 claims abstract description 31
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000004364 calculation method Methods 0.000 claims abstract description 6
- 238000002372 labelling Methods 0.000 claims abstract description 6
- 230000000694 effects Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 238000013135 deep learning Methods 0.000 abstract description 7
- 238000013527 convolutional neural network Methods 0.000 description 9
- 238000013461 design Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000001217 buttock Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
Images
Classifications
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5066—Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs
-
- 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
- G06N3/08—Learning methods
-
- 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
-
- 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
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- 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)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of deep learning, and particularly discloses a method for detecting lewd behavior abnormity of an elevator under a community monitoring scene, which comprises the following steps: passenger video streams in the elevator environment in the community are obtained, passenger images are obtained through decoding, and a data set is made through multi-attribute labeling; inputting the image data set into a designed target multi-attribute detection network model TMD-CNN for training and learning until optimal parameters are obtained; detecting the community elevator environment by using a trained TMD-CNN target multi-attribute detection model, and outputting the passenger number, the gender and the position information of a key point; calculating the distance of the opposite sex key points by using a distance calculation formula, comparing the distance with a set threshold value, and judging whether lewd behavior occurs or not; and adopting a GPU scheduling strategy to perform GPU scheduling. The method and the system use the GPU scheduling strategy to analyze the GPU real-time condition, improve the running speed of the network model, improve the detection efficiency and improve the applicability and the practicability to the elevator environment in the community.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to a method for detecting lewd behavior abnormity of an elevator under a community monitoring scene.
Background
With the continuous development of scientific technology and network technology, the life of people gradually changes to the direction of digitalization, networking and intellectualization, and the living standard of people is improved. In recent years, the rapid rise of artificial intelligence technology and machine learning promotes the development of society towards intelligence again. The intelligent community and the intelligent city are products of artificial intelligence development. More and more researchers begin to pay attention to the field of deep learning, wherein the deep learning is the highest level of the development stage of machine learning at present, and the convolutional neural network is a typical representative of the deep learning and has remarkable effects in the aspects of image detection, image recognition and the like. For feature extraction, the convolutional neural network has the advantage of automatic learning, replaces a complicated traditional manual extraction method, reduces manual intervention and improves the accuracy of feature extraction. The convolution nerve not only has remarkable effect in the aspect of feature extraction, but also plays an irreplaceable role in other aspects such as object detection and identification.
The detection of human bodies by deep learning is a research hotspot at present. The effective detection of human targets can be applied in various fields, such as community life, commercial monitoring, military assistance and the like. In recent years, the target detection is in a continuous development state, and the detection accuracy and efficiency are higher and higher. The fast R-CNN is one of the most effective methods in the field of target detection and identification, and has the advantages that the lifting part of the candidate frame is put on a GPU for operation, the extraction part of the area candidate frame is embedded into the network from the network, and the feature map after convolution can be used for obtaining the area candidate frame. Similar target recognition networks also include Mask-RCNN, YOLO, SSD and the like, although the accuracy of many technologies in experimental effect reaches more than 96%, the applicability of the similar target recognition networks depends heavily on the detection environment, and target detection is influenced by objective reasons such as weather and illumination, uncertainty of target change and other external factors, so that the accuracy of many detection models is reduced. The algorithms cannot meet the requirements of various complex environments, and target detection and identification still have no complete system, so that how to design an algorithm for improving the accuracy of target identification for a specific environment is still the focus of current research.
Lewd acts are a big 'stubborn disease' in the community, and due to the fact that effective monitoring means are lacked all the time, the wind and air environment of the community is affected, and huge potential safety hazards are brought to residents in the community. In recent years, corresponding solutions are provided, the false alarm rate of intelligent monitoring and detection is high due to the influence of complex human body actions and the like, and the monitoring efficiency is greatly reduced. Based on the method, how to perform anti-interference training on the monitoring video analysis function in the elevator through a deep learning algorithm is achieved, so that accurate recognition of lewd behavior of a target under an elevator scene is achieved.
Disclosure of Invention
In order to solve the lewd behavior which is a social 'obstinate' problem, the invention provides a lewd behavior abnormity detection method aiming at an elevator scene in a community. In order to make the Detection model more suitable for the elevator environment, a Target Multi-attribute Detection Network model TMD-CNN (Target Multi-attribute Detection-Convolition neural Network) based on a convolutional neural Network is designed, and the Network model is trained and learned by using an image making data set of the elevator environment to obtain a high-applicability Target Detection model. The detection targets are the gender and the key points of the body of the passenger, including 5 parts of the head, the chest, the hands and the buttocks. And judging whether the lewd abnormal behavior occurs or not by calculating the key point distance between the opposite sex. Because the space of the elevator environment is small, the detection omission phenomenon occurs at the key points, and the distance between any key points is set to be smaller than a set threshold value, so that the lewd behavior can be considered. The design of the TMD-CNN target multi-attribute detection network model improves the applicability and the practicability of target detection, and the GPU scheduling strategy is adopted to improve the operation efficiency of the model. Experiments show that the method can effectively identify the lewd behavior abnormity detection method under the community monitoring scene. The potential safety hazard of community residents is solved, and the wind and gas environment of the community is improved.
In order to solve the technical problems, the technical scheme of the invention is as follows:
a method for detecting lewd behavior abnormity of an elevator under a community monitoring scene comprises the steps of collecting an elevator passenger data set, training a target multi-attribute detection network model, detecting elevator passenger information, calculating a human body key point distance to judge whether lewd behavior occurs or not, and scheduling GPU resources, and further comprises the following steps:
step 1: passenger video streams in the elevator environment in the community are obtained, passenger images are obtained through decoding, and a data set is made through multi-attribute labeling;
step 2: inputting the image data set into a designed target multi-attribute detection network model TMD-CNN for training and learning until optimal parameters are obtained;
and step 3: and detecting the community elevator environment by using the trained TMD-CNN target multi-attribute detection model, and outputting the information of the number, the gender and the position of a key point of passengers.
And 4, step 4: calculating the distance of the opposite sex key points by using a distance calculation formula, comparing the distance with a set threshold value, and judging whether lewd behavior occurs or not;
and 5: and adopting a GPU scheduling strategy to perform GPU scheduling.
Preferably, the step 1 includes: a plurality of high-definition cameras or video acquisition devices are installed in each elevator in the community, an area needing to be monitored is selected, and all video streams in an elevator environment area are obtained. The video stream is decoded to obtain an image set with elevator passengers, in order to obtain a target multi-attribute detection network model with strong applicability, the image set needs to be labeled to make a data set to train the model, and the labeling attributes are as follows: the passenger individual labels analyze the sex, and the position labels of the key points of the human body comprise five parts, namely a head part, a chest part, two hands and a hip part.
Preferably, the step 2 includes: the invention designs a Target Multi-attribute Detection Network model TMD-CNN (Target Multi-attribute Detection-Convolition neural Network) based on a convolutional neural Network, and the Network structure is shown in FIG. 2 and is as follows: the 1 st and 2 nd layers all contain conv, pool and norm, the 3 rd and 4 th layers use the same conv, then are connected with conv5 and pool5, the 6 th is fc, the 7 th and 8 th layers use 6 fc in parallel, and the loss layer is calculated by softmax + cross entropy loss function. Training and learning the data set manufactured in the step 1, wherein the training process comprises the following steps: firstly, training by using default parameters, and continuously adjusting the initial weight, the training rate and the iteration times according to a training intermediate result until the network achieves the optimal detection effect with preset efficiency.
Preferably, the step 3 includes: and (3) detecting the elevator monitoring area in the community in real time by using the trained TMD-CNN target multi-attribute detection network model, and outputting the number of passengers in the elevator, the gender of each passenger and the spatial position information of key points.
Preferably, in the step 4: because the elevator environment space is less, the phenomenon of omission easily appears, in order to avoid that some parts can not be detected due to shielding, the distance between any two key points arranged between opposite sex is smaller than a set threshold value, namely the lewd can be considered to occur, and the threshold value of the experiment is set to be 10 cm. Calculating the distance between the two points according to the spatial position information of the key points detected in the step 3, wherein the distance calculation formula comprises the following steps:
preferably, in step 5, the usage of the GPUs in the GPU processing cluster is monitored in real time, and a proper scheduling policy is adopted to schedule the GPUs in real time.
By adopting the technical scheme, the method for detecting the lewd behavior abnormity of the elevator under the community monitoring scene has the following beneficial effects that:
(1) in order to enable the detection model to be more suitable for the elevator environment, a target multi-attribute detection network model TMD-CNN based on a convolutional neural network is designed, and the network model is trained and learned by using an image making data set of the elevator environment to obtain a high-applicability target detection model.
(2) The positions of 5 key points are detected, because the space of the elevator environment is small, the detection omission phenomenon occurs on the key points, and the distance between any key points between opposite sex is set to be smaller than a set threshold value, so that the lewd behavior can be considered. The method is used for judging the lewd abnormal behavior, and the detection accuracy is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for detecting lewd behavior anomaly in an elevator under a community monitoring scenario in accordance with the present invention;
FIG. 2 is a schematic structural diagram of a Target Multi-attribute Detection-contribution neural Network model TMD-CNN (Target Multi-attribute Detection-contribution neural Network) designed based on a convolutional neural Network structure according to the present invention;
FIG. 3 is a diagram of a GPU resource scheduling policy in a GPU processor cluster according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, a method for detecting lewd prader exception in an elevator in a community monitoring environment includes: the method comprises the steps of collecting an elevator passenger data set, training a target multi-attribute detection network model, detecting elevator passenger information, calculating an opposite sex key point distance, judging whether lewd behaviors occur or not and scheduling GPU resources.
The following describes in detail a method for detecting lewd behavior anomaly in an elevator under a community monitoring environment:
as shown in fig. 1, passenger video streams in the elevator environment in a community are acquired, passenger images are obtained through decoding, and a multi-attribute labeling data set is made; inputting the image data set into a designed target multi-attribute detection network model TMD-CNN for training and learning until optimal parameters are obtained; and detecting the community elevator environment by using the trained TMD-CNN target multi-attribute detection model, and outputting the information of the number, the gender and the position of a key point of passengers. Calculating the distance of the opposite sex key points by using a distance calculation formula, comparing the distance with a set threshold value, and judging whether lewd behavior occurs or not; and adopting a GPU scheduling strategy to perform GPU scheduling. The invention designs a Target Multi-attribute Detection Network model TMD-CNN (Target Multi-attribute Detection-Convolvulation neural Network) based on a convolutional neural Network, detects elevator passengers by utilizing the trained TMD-CNN Target Multi-attribute Detection model, and counts the number and the gender information of the elevator passengers and the marks of key points of the passengers. The detection network model is very easily influenced by the detection environment, and in order to adapt to the human body multi-attribute detection in the specific elevator environment, the network model is trained and learned by collecting passenger image making data sets under the elevator environment in a community at multiple angles through multiple cameras, so that the accuracy and the applicability of TMD-CNN model detection are improved. The target lewd behavior is judged by calculating the distance between key points of the human body of an opposite sex passenger in the elevator, the key points of the human body are set to be five parts of a head, a chest, two hands and a hip, in order to avoid shielding, some parts cannot be detected, and the lewd behavior can be considered to occur when the distance between any two key points between the opposite sex is set to be smaller than a set threshold value. In addition, the invention uses the GPU scheduling strategy to analyze the real-time condition of the GPU, thereby improving the running speed of the network model and improving the detection efficiency. The design of the TMD-CNN network model improves the applicability and the practicability of the method to the elevator environment in the community. The lewd behavior is analyzed through the distance between the key points of the human body, so that the detection precision is effectively improved.
The GPU resource scheduling layer monitors the current GPU resource use condition in real time according to a scheduling strategy as shown in figure 3, before a GPU processor cluster distributes tasks, whether the current GPU consumption is too large is checked, if the consumption is too large, a GPU use condition list and a GPU computing capacity list are checked, and a GPU receiving task is reselected.
The invention provides a method for detecting lewd behavior abnormity of an elevator under a community monitoring scene, which is characterized in that in order to enable a detection model to be more suitable for an elevator environment, a target multi-attribute detection network model TMD-CNN based on a convolutional neural network is designed, and an image making data set of the elevator environment is used for training and learning the network model to obtain a high-applicability target detection model. The target lewd behavior is judged by calculating the distance between opposite sex key points of passengers in the elevator, the key points of a human body are set to be four parts of a head, a chest, two hands and a hip, in order to avoid shielding to cause that some parts cannot be detected, the lewd behavior can be considered to occur by setting the distance between any two key points to be smaller than a set threshold value, and the detection accuracy is improved. In addition, the invention uses the GPU scheduling strategy to analyze the real-time condition of the GPU, thereby improving the running speed of the network model and improving the detection efficiency.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (6)
1. A method for detecting lewd behavior abnormity of an elevator under a community monitoring scene is characterized in that: the method comprises the following steps:
step 1: passenger video streams in the elevator environment in the community are obtained, passenger images are obtained through decoding, and a data set is made through multi-attribute labeling;
step 2: inputting the image data set into a designed target multi-attribute detection network model TMD-CNN for training and learning until optimal parameters are obtained;
and step 3: detecting the community elevator environment by using a trained TMD-CNN target multi-attribute detection model, and outputting the passenger number, the gender and the position information of a key point;
and 4, step 4: calculating the distance of the opposite sex key points by using a distance calculation formula, comparing the distance with a set threshold value, and judging whether lewd behavior occurs or not;
and 5: and adopting a GPU scheduling strategy to perform GPU scheduling.
2. The method for detecting lewd behavior anomaly in an elevator under a community monitoring scenario according to claim 1, wherein: in the step 1, installing a high-definition camera or a video acquisition device in each elevator of the community, selecting an area needing to be monitored, and acquiring all video streams in an elevator environment area; the video stream is decoded to obtain an image set with elevator passengers, in order to obtain a target multi-attribute detection network model with strong applicability, the image set needs to be labeled to make a data set to train the model, and the labeling attributes are as follows: the passenger individual labels analyze the sex, and the position labels of the key points of the human body comprise five parts, namely a head part, a chest part, two hands and a hip part.
3. The method for detecting lewd behavior anomaly in an elevator under a community monitoring scenario according to claim 1, wherein: in the step 2, the method further includes training and learning the target multi-attribute detection network model TMD-CNN by using the data set produced in the step 1, and the training process: firstly, training by using default parameters, and continuously adjusting the initial weight, the training rate and the iteration times according to a training intermediate result until the network achieves the optimal detection effect with preset efficiency.
4. The method for detecting lewd behavior anomaly in an elevator under a community monitoring scenario according to claim 1, wherein: and in the step 3, the trained TMD-CNN target multi-attribute detection network model is used for detecting elevator monitoring areas in the community in real time, and outputting the number of passengers in the elevator, the gender of each passenger and the spatial position information of key points.
5. The method for detecting lewd behavior anomaly in an elevator under a community monitoring scenario according to claim 1, wherein: in the step 4, the lewd may be considered to occur when the distance between any two key points between the opposite sex is smaller than a set threshold value; calculating the distance between the two points according to the spatial position information of the key points detected in the step 3, wherein the distance calculation formula comprises the following steps:
6. the method for detecting lewd behavior anomaly in an elevator under a community monitoring scenario according to claim 1, wherein: in the step 5, the usage of the GPUs in the GPU processing cluster is monitored in real time, and a proper scheduling strategy is adopted to schedule the GPUs in real time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011029847.4A CN112307896A (en) | 2020-09-27 | 2020-09-27 | Method for detecting lewd behavior abnormity of elevator under community monitoring scene |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011029847.4A CN112307896A (en) | 2020-09-27 | 2020-09-27 | Method for detecting lewd behavior abnormity of elevator under community monitoring scene |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112307896A true CN112307896A (en) | 2021-02-02 |
Family
ID=74488428
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011029847.4A Withdrawn CN112307896A (en) | 2020-09-27 | 2020-09-27 | Method for detecting lewd behavior abnormity of elevator under community monitoring scene |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112307896A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113031464A (en) * | 2021-03-22 | 2021-06-25 | 北京市商汤科技开发有限公司 | Device control method, device, electronic device and storage medium |
-
2020
- 2020-09-27 CN CN202011029847.4A patent/CN112307896A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113031464A (en) * | 2021-03-22 | 2021-06-25 | 北京市商汤科技开发有限公司 | Device control method, device, electronic device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107330920B (en) | Monitoring video multi-target tracking method based on deep learning | |
| CN110070033B (en) | Method for detecting wearing state of safety helmet in dangerous working area in power field | |
| CN109615019B (en) | Abnormal behavior detection method based on space-time automatic encoder | |
| CN105426820B (en) | More people's anomaly detection methods based on safety monitoring video data | |
| Yu et al. | Railway obstacle detection algorithm using neural network | |
| CN109359697A (en) | Graph image recognition methods and inspection system used in a kind of power equipment inspection | |
| CN109255298A (en) | Safety helmet detection method and system in dynamic background | |
| CN103839065A (en) | Extraction method for dynamic crowd gathering characteristics | |
| CN110728252B (en) | Face detection method applied to regional personnel motion trail monitoring | |
| CN112287827A (en) | Complex environment pedestrian mask wearing detection method and system based on intelligent lamp pole | |
| Ni et al. | An improved Faster R-CNN for defect recognition of key components of transmission line | |
| CN115761537B (en) | Power transmission line foreign matter intrusion identification method oriented to dynamic feature supplementing mechanism | |
| CN112801009B (en) | Facial emotion recognition method, device, medium and equipment based on double-flow network | |
| Cao et al. | YOLO-SF: YOLO for fire segmentation detection | |
| CN113158983A (en) | Airport scene activity behavior recognition method based on infrared video sequence image | |
| CN112800975A (en) | Behavior identification method in security check channel based on image processing | |
| CN104090658A (en) | Human body image detection method based on HOG features | |
| CN110210399A (en) | Face recognition method based on uncertainty quantization probability convolution neural network | |
| CN113205039A (en) | Power equipment fault image identification and disaster investigation system and method based on multiple DCNNs | |
| CN109002753A (en) | One kind being based on the cascade large scene monitoring image method for detecting human face of convolutional neural networks | |
| CN113936034B (en) | Apparent motion combined weak and small moving object detection method combined with inter-frame optical flow | |
| Gupta et al. | Progression modelling for online and early gesture detection | |
| CN112307896A (en) | Method for detecting lewd behavior abnormity of elevator under community monitoring scene | |
| CN112270207A (en) | Smoke and fire detection method in community monitoring scene | |
| CN107729811B (en) | Night flame detection method based on scene modeling |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210202 |
|
| WW01 | Invention patent application withdrawn after publication |