CN112232273A - Early warning method and system based on machine learning identification image - Google Patents
Early warning method and system based on machine learning identification image Download PDFInfo
- Publication number
- CN112232273A CN112232273A CN202011205888.4A CN202011205888A CN112232273A CN 112232273 A CN112232273 A CN 112232273A CN 202011205888 A CN202011205888 A CN 202011205888A CN 112232273 A CN112232273 A CN 112232273A
- Authority
- CN
- China
- Prior art keywords
- video data
- violation
- labels
- type
- machine learning
- 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
- 238000010801 machine learning Methods 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000006399 behavior Effects 0.000 claims abstract description 89
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 239000012634 fragment Substances 0.000 claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 8
- 230000000007 visual effect Effects 0.000 claims abstract description 7
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 230000000391 smoking effect Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 12
- 238000012544 monitoring process Methods 0.000 description 6
- 238000012216 screening Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000005070 sampling Methods 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
- G06V20/40—Scenes; Scene-specific elements in video content
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
Abstract
The invention discloses an early warning method and system based on machine learning identification images, wherein the method comprises the following steps: the method comprises the steps that a neural network video detection module obtains original video data, and the original video data are detected through a machine learning visual identification algorithm to obtain detected video data fragments; processing the detected video data fragments according to the grade influencing the working efficiency to increase behavior labels; and playing the video data segments after detection including the illegal action labels according to the generation time sequence of the original video data, and playing the video data segments after detection in a first screen area when two illegal action labels of a first type, a second type and a third type appear in the same video data segment after detection. According to the early warning method and system based on the machine learning identification image, the video of the violation, particularly the serious violation, is automatically screened and visually displayed in real time through the machine learning identification image, and the efficiency of video data processing is improved.
Description
Technical Field
The invention relates to the field of video identification processing, in particular to an early warning method and system based on machine learning identification images.
Background
At present, the conventional wharf is continuously developed towards an automatic wharf, and four operation links are covered in the whole system of the operation of the automatic wharf, including a bridge crane (transportation operation between a ship and a shore), an internal container truck (operation between the shore and a yard), a tower crane (operation between the yard and an external container truck) and an external container truck (outward transportation of containers). Because a large number of operation processes need to be completed in a short time in a large wharf, the accident rate needs to be reduced to the lowest extent in four links in the wharf operation process, otherwise, accidents or delay together will cause a large amount of economic loss.
In the prior art, the behavior in the monitoring cabin cannot obtain the real-time safety early warning of the monitoring video, the traditional manual detection mode is usually used for detecting the real-time monitoring video acquired by hundreds of devices, but the manual detection of the monitoring video brings great cost, the manual detection of the monitoring video is very easy to make mistakes, and the quality of the safety early warning cannot be guaranteed. In addition, the mode of manually detecting the surveillance video brings great difficulty to wharf safety supervision work, and because the amount of the surveillance video data is huge, the surveillance video data can only be subjected to afterward tracing after sampling detection, or the surveillance video can be reviewed after an accident occurs, so that real-time early warning of the surveillance video can not be achieved, comprehensive inspection of all the surveillance videos can not be achieved, and potential safety hazards are brought to the operation of an automatic wharf.
Therefore, it is necessary to provide an early warning system and method based on machine learning identification, which can perform real-time early warning on the monitoring video.
Disclosure of Invention
The invention aims to solve the technical problem of providing an early warning method and system based on machine learning identification images, automatically screening and visually displaying illegal behaviors, particularly videos of serious illegal behaviors in real time through the machine learning identification images, and improving the efficiency of video data processing.
The invention adopts the technical scheme that an early warning method for recognizing images based on machine learning is provided to solve the technical problems, and the method comprises the following steps:
the method comprises the steps that a neural network video detection module obtains original video data from a plurality of detection servers, and the original video data are detected through a machine learning visual identification algorithm to obtain detected video data fragments;
processing the detected video data fragments according to the grade influencing the working efficiency to increase behavior labels, wherein the behavior labels comprise violation labels and safety behavior labels, and the violation labels comprise a first type of violation labels, a second type of violation labels and a third type of violation labels;
the early warning sequencing algorithm module plays the video data fragments including the violation labels after detection on a display module according to the generation time sequence of the original video data, wherein the display module comprises a first screen area and a second screen area;
and when two violation behavior labels of the first type violation behavior label, the second type violation behavior label and the third type violation behavior label appear in the same detected video data segment, playing the detected video data segment in the first screen area.
Preferably, when two violation labels of the first type violation label, the second type violation label, and the third type violation label appear in a plurality of detected video data segments, the detected video data segment with the longest duration is played in the first screen area.
Preferably, when the first type of violation labels, the second type of violation labels, and the third type of violation labels appear in the detected video data segment at the same time, the detected video data segment is played in the first screen area and voice warning and acousto-optic warning are performed.
Preferably, when the original video data are generated at the same time and the detected video data segments including the violation tags are sorted according to the device number of the camera generating the original video data, the detected video data segments including the violation tags are played in the first screen area and the second screen area.
Preferably, the violations include smoking behaviors, electronic device operation behaviors, and belt-off behaviors, and the safety behaviors include work behaviors.
Preferably, the detected video data segments are stored in a video stream buffer pool, first-in first-out is performed according to a generation time sequence of original video data, and the detected video data segments which are first-in the video stream buffer pool are played in the first screen area first.
Preferably, the detected video data segments including the violation labels played in the first screen area and the second screen area are not played again after being subjected to resolution processing.
Preferably, when the detected video data segments including the violation labels played in the first screen area and the second screen area have violation label errors, the detected video data segments including the violation labels are stored, and the neural network video detection module is updated.
Preferably, the second screen region includes a plurality of sub-screen regions for playing the detected video data segment including the violation label in the generation time order of the original video data.
The invention also adopts the technical scheme of solving the technical problems and provides a machine learning identification-based early warning system using the machine learning identification-based early warning method.
Compared with the prior art, the invention has the following beneficial effects: according to the early warning method and system based on the machine learning identification image, the original video data are detected through a machine learning visual identification algorithm to obtain detected video data fragments, the detected video data fragments are processed according to the grade influencing the working efficiency to increase behavior labels, the behavior labels comprise violation labels and safety behavior labels, and the violation labels comprise a first type of violation labels, a second type of violation labels and a third type of violation labels; the early warning sequencing algorithm module plays the video data fragments including the violation labels after detection on a display module according to the generation time sequence of the original video data, wherein the display module comprises a first screen area and a second screen area; when two violation labels of the first type violation label, the second type violation label and the third type violation label appear in the same detected video data segment, playing the detected video data segment in the first screen area, and displaying the violation in the first screen area and the second screen area of the display module intuitively in real time through automatic screening, especially playing a video of a serious violation in the first screen area, so that the serious violation can be seen clearly and emphatically, and the efficiency of video data processing is improved;
further, when two violation labels of the first type violation label, the second type violation label and the third type violation label appear in a plurality of detected video data segments, playing the detected video data segment with the longest duration in the first screen area, wherein the detected video data segment with the longest duration represents that the violation is performed for a long time, and the video data segment is required to be played in the first screen area for a serious violation so as to achieve the purpose of reminding;
further, when the first type of violation labels, the second type of violation labels, and the third type of violation labels simultaneously appear on the detected video data segment, the detected video data segment is played in the first screen area and voice warning and acousto-optic warning are performed, and when three types of violation behaviors simultaneously appear, the video data segment is very serious, and the voice warning and the acousto-optic warning are required to be performed while the video data segment is played in the first screen area, so that the purposes of reminding and warning are achieved.
Drawings
FIG. 1 is a flow chart of a method for identifying an image early warning based on machine learning according to an embodiment of the present invention;
FIG. 2 is a block diagram of an early warning method for recognizing images based on machine learning according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating a display module of an early warning method for recognizing images based on machine learning according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a display module of an early warning method based on machine learning identification images according to an embodiment of the present invention;
FIG. 5 is another diagram of a display module of a pre-warning method for recognizing images based on machine learning according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating a statistical module of an early warning method for recognizing images based on machine learning according to an embodiment of the present invention;
fig. 7 is a block diagram of an early warning system based on machine learning recognition images according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the figures and examples.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be apparent to one of ordinary skill in the art that the present invention may be practiced without these specific details. Accordingly, the particular details set forth are merely exemplary, and the particular details may be varied from the spirit and scope of the present invention and still be considered within the spirit and scope of the present invention.
Referring now to fig. 1, a flow chart of an early warning method for recognizing images based on machine learning in an embodiment of the present invention is shown. An early warning method based on machine learning recognition images comprises the following steps:
step 101: the method comprises the steps that a neural network video detection module obtains original video data from a plurality of detection servers, and the original video data are detected through a machine learning visual identification algorithm to obtain detected video data fragments;
step 102: processing the detected video data fragments according to the grade influencing the working efficiency to increase behavior labels, wherein the behavior labels comprise violation labels and safety behavior labels, and the violation labels comprise a first type of violation labels, a second type of violation labels and a third type of violation labels;
step 103: the early warning sequencing algorithm module plays the video data fragments including the violation labels after detection on a display module according to the generation time sequence of the original video data, wherein the display module comprises a first screen area and a second screen area;
step 104: and when two violation behavior labels of the first type violation behavior label, the second type violation behavior label and the third type violation behavior label appear in the same detected video data segment, playing the detected video data segment in the first screen area.
In a specific implementation, the original video data is detected through a machine learning visual recognition algorithm, including dividing an image in the original video data into 9 × 9 cells, and transferring the image divided into 9 × 9 cells to a Convolutional Neural Network (CNN), so as to apply an image classification and positioning process to each mesh, and obtain a border and a corresponding class probability of an illegal activity object, where the output format is y ═ pc, bx, by, bh, bw, c1, c2, c3, pc is a probability that the object exists in the mesh, and bx, by, bh, bw are bounding boxes and c1, c2, c3 are monitored categories.
And when two violation behavior labels of the first type violation behavior label, the second type violation behavior label and the third type violation behavior label appear in a plurality of detected video data fragments, playing the detected video data fragment with the longest duration in the first screen area. The first type of violation behavior tag may be a smoking behavior tag, the second type of violation behavior tag may be an operating electronic device behavior, and the third type of violation behavior tag may be an unworn waistband behavior tag.
And when the first type of illegal action label, the second type of illegal action label and the third type of illegal action label appear in the detected video data segment at the same time, playing the detected video data segment in the first screen area and carrying out voice warning and acousto-optic warning.
When the generation time of the original video data is the same and the detected video data segments including the violation labels are included, sorting according to the equipment number of the camera generating the original video data so as to play the detected video data segments including the violation labels in the first screen area and the second screen area.
The illegal behaviors comprise smoking behaviors, electronic equipment operation behaviors and waistband-free behaviors, and the safety behaviors comprise working behaviors.
Referring now to fig. 2, fig. 2 is a block diagram of an early warning method for recognizing images based on machine learning according to an embodiment of the present invention. The plurality of detection servers 12 are connected with the plurality of cameras 11, the plurality of cameras 11 are used for shooting original video data, the plurality of detection servers 12 are connected with the display 13, and the display 13 comprises a display module 14. The display module 14 is connected to a data storage module 15, which is used for storing the detected video data segments.
Referring now to fig. 3, fig. 3 is a schematic diagram of a display module of an early warning method based on machine learning recognition images according to an embodiment of the present invention. The display module 14 includes a first screen area 21 and a second screen area 22, the first screen area 21 being a main screen for playing heavily offending video data clips. The second screen area 22 includes a plurality of sub-screen areas, the sub-screen areas are used for playing the detected video data segments including the violation labels according to the generation time sequence of the original video data, and the second screen area 22 includes a sub-screen 0, a sub-screen 1, a sub-screen 2, a sub-screen 3, a sub-screen 4, and a sub-screen 5. The second screen area 22 is used for playing the detected video data segment including the violation label in the order of the generation time of the original video data. Referring now to fig. 4, fig. 4 is a schematic diagram of a display module of an early warning method based on machine learning recognition images according to an embodiment of the present invention. The severe violation area may include when two or three of the first, second, and third types of violation tags are present in the plurality of detected video data segments, such as smoking behavior and electronic device operation behavior. The light violation area may include when one or two of the first type violation tags, the second type violation tags, and the third type violation tags occur in multiple detected video data segments, such as an operating electronic device behavior and a non-wearing belt behavior. The safety behavior region includes safety behavior, i.e., working behavior. Referring now to fig. 5, fig. 5 is another schematic diagram of a display module of a warning method for recognizing images based on machine learning according to an embodiment of the present invention. The main screen, the auxiliary screen 0, the auxiliary screen 1, the auxiliary screen 2, the auxiliary screen 3, the auxiliary screen 4 and the auxiliary screen 5 are arranged in sequence and used for playing the serious violation 1, the serious violation 2, the serious violation 3, the light violation 1, the light violation 2, the light violation 3, the safety behavior 1, the safety behavior 2 and the safety behavior 3 in sequence.
In an implementation, the display module 14 further includes a touch key area 23, and the touch key area 23 includes a notification key, a forwarding key, a resolved key, and an error notification key. And the notification key is used for transferring the detected video data segment including the violation label to a loudspeaker in the cockpit to play warning voice. The forwarding key is used for selecting a corresponding responsibility group for notification. The resolved key is used to empty the warning message at that time and not enter the queue within 1 minute, this key operation is to allow time for processing activity and at the same time prevent overly intensive alerts. The error-reporting key is used for storing the video evidence locally when an error-reporting prompt pops up, and the detected video data segments including the violation labels and played in the first screen area 21 and the second screen area 22 are not played any more after being subjected to resolution processing. When the detected video data segments including the violation labels played in the first screen area 21 and the second screen area 22 have violation label errors, storing the detected video data segments including the violation labels, and updating the neural network video detection module.
In a specific implementation, the display module 14 further includes a statistics module 24, and the statistics module 24 is configured to count the number of video data segments of the working behaviors, the number of video data segments of violation behaviors 1, namely, the first type of violation behaviors, the number of video data segments of violation behaviors 2, namely, the second type of violation behaviors, and the number of video data segments of violation behaviors 3, namely, the third type of violation behaviors. Referring to fig. 6, fig. 6 is a schematic diagram of a statistical module of an early warning method for recognizing an image based on machine learning according to an embodiment of the present invention, where the number of video data segments of a working behavior is 12, the number of video data segments of violation behavior 1, i.e., a first type of violation behavior, is 13, the number of video data segments of violation behavior 2, i.e., a second type of violation behavior, is 1, and the number of video data segments of violation behavior 3, i.e., a third type of violation behavior, is 4.
Referring now to fig. 7, fig. 7 is a block diagram of a warning system for recognizing images based on machine learning according to an embodiment of the present invention. The neural network video detection module 69 acquires original video data from a plurality of cameras including a container truck 1 camera 61, a bridge crane 1 camera 62, a bridge crane 15 camera 63, a tower crane 2 camera 64, a container truck 121 camera 64 and a bridge crane 34 camera 66, and the early warning sorting algorithm module 68 plays a detected video data segment including an illegal action label on the display module according to the generation time sequence of the original video data, wherein the display module includes a first screen area and a second screen area. The detected video data segments are stored in the video stream buffer pool 67, and first-in first-out is performed according to the generation time sequence of the original video data, and the detected video data segments which are first-in the video stream buffer pool 67 are played in the first screen area first. The first screen area is a main screen, and the second screen area comprises a sub screen 0, a sub screen 1, a sub screen 2, a sub screen 3, a sub screen 4 and a sub screen 5. The video stream cache pool 67 includes a severe violation video, a light violation video, and a security violation video.
The invention also adopts the technical scheme of solving the technical problems and provides a machine learning identification-based early warning system using the machine learning identification-based early warning method.
In summary, according to the early warning method and system based on machine learning identification image provided by this embodiment, the original video data is detected through a machine learning visual identification algorithm to obtain a detected video data segment, and the detected video data segment is processed to add a behavior tag according to a level affecting working efficiency, where the behavior tag includes an illegal behavior tag and a security behavior tag, and the illegal behavior tag includes a first type of illegal behavior tag, a second type of illegal behavior tag, and a third type of illegal behavior tag; the early warning sequencing algorithm module plays the video data fragments including the violation labels after detection on a display module according to the generation time sequence of the original video data, wherein the display module comprises a first screen area and a second screen area; when two violation labels of the first type violation label, the second type violation label and the third type violation label appear in the same detected video data segment, playing the detected video data segment in the first screen area, and displaying the violation in the first screen area and the second screen area of the display module intuitively in real time through automatic screening, especially playing a video of a serious violation in the first screen area, so that the serious violation can be seen clearly and emphatically, and the efficiency of video data processing is improved;
further, when two violation labels of the first type violation label, the second type violation label and the third type violation label appear in a plurality of detected video data segments, playing the detected video data segment with the longest duration in the first screen area, wherein the detected video data segment with the longest duration represents that the violation is performed for a long time, and the video data segment is required to be played in the first screen area for a serious violation so as to achieve the purpose of reminding;
further, when the first type of violation labels, the second type of violation labels, and the third type of violation labels simultaneously appear on the detected video data segment, the detected video data segment is played in the first screen area and voice warning and acousto-optic warning are performed, and when three types of violation behaviors simultaneously appear, the video data segment is very serious, and the voice warning and the acousto-optic warning are required to be performed while the video data segment is played in the first screen area, so that the purposes of reminding and warning are achieved.
Although the present invention has been described with respect to the preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. An early warning method for recognizing images based on machine learning is characterized by comprising the following steps:
the method comprises the steps that a neural network video detection module obtains original video data from a plurality of detection servers, and the original video data are detected through a machine learning visual identification algorithm to obtain detected video data fragments;
processing the detected video data fragments according to the grade influencing the working efficiency to increase behavior labels, wherein the behavior labels comprise violation labels and safety behavior labels, and the violation labels comprise a first type of violation labels, a second type of violation labels and a third type of violation labels;
the early warning sequencing algorithm module plays the video data fragments including the violation labels after detection on a display module according to the generation time sequence of the original video data, wherein the display module comprises a first screen area and a second screen area;
and when two violation behavior labels of the first type violation behavior label, the second type violation behavior label and the third type violation behavior label appear in the same detected video data segment, playing the detected video data segment in the first screen area.
2. The early warning method based on machine learning identification images according to claim 1, wherein when two of the first type violation labels, the second type violation labels and the third type violation labels appear in a plurality of detected video data segments, the detected video data segment with the longest duration is played in the first screen area.
3. The early warning method based on machine learning recognition images according to claim 1, wherein when the first type of violation labels, the second type of violation labels and the third type of violation labels appear in the detected video data segment at the same time, the detected video data segment is played in the first screen area and voice warning and acousto-optic warning are performed.
4. The machine-learning-based recognition image early warning method of claim 1, wherein when the original video data are generated at the same time and include the detected video data segment of the violation label, the original video data are sorted according to a device number of a camera generating the original video data to play the detected video data segment including the violation label in the first screen region and the second screen region.
5. The machine learning identification image-based early warning method according to claim 1, wherein the violation behaviors include smoking behaviors, electronic device operation behaviors, and belt-off-wearing behaviors, and the safety behaviors include work behaviors.
6. The early warning method based on machine learning identification image according to claim 1, wherein the detected video data segment is stored in a video stream buffer pool, first-in first-out is performed according to a generation time sequence of original video data, and the detected video data segment which is first-in first in the video stream buffer pool is played first in the first screen area.
7. The machine-learning-based recognition image early warning method according to claim 1, wherein the detected video data segments including the violation labels played in the first screen region and the second screen region are not played again after being subjected to resolution processing.
8. The early warning method based on machine learning identification images according to claim 1, wherein when the detected video data segment including the violation label played in the first screen region and the second screen region has a violation label error, the detected video data segment including the violation label is stored and the neural network video detection module is updated.
9. The machine-learning-based recognition image early-warning method according to claim 1, wherein the second screen region includes a plurality of sub-screen regions for playing the detected video data segment including the violation label in the generation time order of the original video data.
10. A machine learning identification based early warning system using the machine learning identification based early warning method according to any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011205888.4A CN112232273A (en) | 2020-11-02 | 2020-11-02 | Early warning method and system based on machine learning identification image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011205888.4A CN112232273A (en) | 2020-11-02 | 2020-11-02 | Early warning method and system based on machine learning identification image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112232273A true CN112232273A (en) | 2021-01-15 |
Family
ID=74122005
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011205888.4A Pending CN112232273A (en) | 2020-11-02 | 2020-11-02 | Early warning method and system based on machine learning identification image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112232273A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065444A (en) * | 2021-03-26 | 2021-07-02 | 北京大米科技有限公司 | Behavior detection method and device, readable storage medium and electronic equipment |
| CN113221845A (en) * | 2021-06-07 | 2021-08-06 | 北京猎豹移动科技有限公司 | Advertisement auditing method, device, equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020019703A1 (en) * | 2000-08-02 | 2002-02-14 | Levine Alfred B. | Automotive safety enhansing system |
| CN102053795A (en) * | 2011-01-11 | 2011-05-11 | 上海华平信息技术股份有限公司 | Method and system for freely laying out video display window |
| CN103109025A (en) * | 2010-09-17 | 2013-05-15 | 日立建机株式会社 | Work machine display system |
| CN103337176A (en) * | 2013-07-09 | 2013-10-02 | 辽宁天久信息科技产业有限公司 | Traffic violation snapshotting system and traffic violation snapshotting method |
| CN208376628U (en) * | 2018-06-26 | 2019-01-15 | 比歪尔科技（深圳）有限公司 | A kind of safe driving assistant system |
| CN109872528A (en) * | 2019-04-02 | 2019-06-11 | 地洲智慧能源(浙江)有限公司 | A kind of two-way wisdom traffic lights of solar energy |
| CN110765848A (en) * | 2019-09-08 | 2020-02-07 | 天津大学 | Chemical plant personnel safety guarantee system and early warning method based on artificial intelligence image processing algorithm |
| CN210558713U (en) * | 2019-07-26 | 2020-05-19 | 华鼎世纪（北京）国际科技有限公司 | Elevator thing networking wisdom analysis terminal |
| CN111368696A (en) * | 2020-02-28 | 2020-07-03 | 淮阴工学院 | Dangerous chemical transport vehicle illegal driving behavior detection method and system based on visual cooperation |
| CN111444843A (en) * | 2020-03-26 | 2020-07-24 | 中科海微(北京)科技有限公司 | Multimode driver and vehicle illegal behavior monitoring method and system |
-
2020
- 2020-11-02 CN CN202011205888.4A patent/CN112232273A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020019703A1 (en) * | 2000-08-02 | 2002-02-14 | Levine Alfred B. | Automotive safety enhansing system |
| CN103109025A (en) * | 2010-09-17 | 2013-05-15 | 日立建机株式会社 | Work machine display system |
| CN102053795A (en) * | 2011-01-11 | 2011-05-11 | 上海华平信息技术股份有限公司 | Method and system for freely laying out video display window |
| CN103337176A (en) * | 2013-07-09 | 2013-10-02 | 辽宁天久信息科技产业有限公司 | Traffic violation snapshotting system and traffic violation snapshotting method |
| CN208376628U (en) * | 2018-06-26 | 2019-01-15 | 比歪尔科技（深圳）有限公司 | A kind of safe driving assistant system |
| CN109872528A (en) * | 2019-04-02 | 2019-06-11 | 地洲智慧能源(浙江)有限公司 | A kind of two-way wisdom traffic lights of solar energy |
| CN210558713U (en) * | 2019-07-26 | 2020-05-19 | 华鼎世纪（北京）国际科技有限公司 | Elevator thing networking wisdom analysis terminal |
| CN110765848A (en) * | 2019-09-08 | 2020-02-07 | 天津大学 | Chemical plant personnel safety guarantee system and early warning method based on artificial intelligence image processing algorithm |
| CN111368696A (en) * | 2020-02-28 | 2020-07-03 | 淮阴工学院 | Dangerous chemical transport vehicle illegal driving behavior detection method and system based on visual cooperation |
| CN111444843A (en) * | 2020-03-26 | 2020-07-24 | 中科海微(北京)科技有限公司 | Multimode driver and vehicle illegal behavior monitoring method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113065444A (en) * | 2021-03-26 | 2021-07-02 | 北京大米科技有限公司 | Behavior detection method and device, readable storage medium and electronic equipment |
| CN113221845A (en) * | 2021-06-07 | 2021-08-06 | 北京猎豹移动科技有限公司 | Advertisement auditing method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112232273A (en) | Early warning method and system based on machine learning identification image | |
| CN111241343A (en) | Road information monitoring and analyzing detection method and intelligent traffic control system | |
| CN111564015B (en) | Method and device for monitoring perimeter intrusion of rail transit | |
| CN102348128A (en) | Surveillance camera system having camera malfunction detection function | |
| CN106990109B (en) | A kind of underground belt foreign matter detection system and method based on multi-cam judgement | |
| CN111629181A (en) | Fire-fighting life passage monitoring system and method | |
| CN112560816A (en) | Equipment indicator lamp identification method and system based on YOLOv4 | |
| CN114998234A (en) | Self-supervision spring defect detection method based on data enhancement strategy | |
| CN108357517A (en) | Motor train unit train operation troubles information detecting method and system | |
| CN211184122U (en) | Intelligent video analysis system for linkage of railway operation safety prevention and control and large passenger flow early warning | |
| CN113239854A (en) | Ship identity recognition method and system based on deep learning | |
| CN113420673B (en) | Garbage classification method, device, equipment and storage medium | |
| CN102745212A (en) | Method and system for monitoring shutting safety for vehicle inspection and maintenance | |
| CN112686130A (en) | Wisdom fishing boat supervision decision-making system | |
| JPS59192944A (en) | Classifying method of pattern defect | |
| CN205336414U (en) | Automatic switch emergent terminal equipment | |
| CN113992984B (en) | Elevator advertisement monitoring and broadcasting method | |
| CN115359416A (en) | Intelligent early warning system for railway freight yard sky eye | |
| CN115147770A (en) | Belt foreign matter vision recognition system based on image processing | |
| CN110769042B (en) | System for railway loading and unloading operation information acquisition | |
| CN113850773A (en) | Detection method, device, equipment and computer readable storage medium | |
| CN114399671A (en) | Target identification method and device | |
| CN110991286A (en) | Automatic electric shovel bucket defect identification device and method based on convolutional neural network | |
| CN114205667B (en) | Broadcast television broadcast abnormal picture identification and fault analysis system | |
| CN116486324B (en) | Subway seat trampling behavior detection method, device, equipment and storage medium |
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 |