CN111339901B - Image-based intrusion detection method and device, electronic equipment and storage medium - Google Patents
Image-based intrusion detection method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111339901B CN111339901B CN202010108343.5A CN202010108343A CN111339901B CN 111339901 B CN111339901 B CN 111339901B CN 202010108343 A CN202010108343 A CN 202010108343A CN 111339901 B CN111339901 B CN 111339901B
- Authority
- CN
- China
- Prior art keywords
- person
- snap
- human body
- area
- shoe
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 127
- 230000002093 peripheral effect Effects 0.000 claims abstract description 60
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000012544 monitoring process Methods 0.000 claims abstract description 7
- 230000009545 invasion Effects 0.000 claims description 24
- 230000006399 behavior Effects 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 12
- 230000033001 locomotion Effects 0.000 claims description 11
- 230000001133 acceleration Effects 0.000 claims description 10
- 238000002372 labelling Methods 0.000 claims description 6
- 238000012790 confirmation Methods 0.000 claims description 4
- 230000001419 dependent effect Effects 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 description 16
- 238000012795 verification Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010200 validation analysis 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
- G06V20/50—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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Social Psychology (AREA)
- Psychiatry (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides an intrusion detection method and device based on images, electronic equipment and a storage medium, wherein the method comprises the following steps: step 1: detecting and monitoring the snap shot picture by adopting a first algorithm; step 2: marking a perimeter region in the snap shot picture; step 3: executing step 4 when the detection frame is partially overlapped with the peripheral area; step 4: identifying the position of the shoes of the person in the detection frame of the person by adopting a second algorithm; step 5: determining a relationship of a position of a person's shoe to a perimeter region; intrusion occurs when the location of the person's shoe is within the perimeter region. According to the image-based intrusion detection method, event detection is carried out by judging the position relation between the shoes or the feet and the peripheral area, rather than only by identifying the position relation between the detection frame of the human body and the peripheral area, the false alarm problem of the first logic can be effectively avoided, and the false alarm problem of the second logic can be also avoided; the accuracy of intrusion alert in the perimeter area is improved.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to an image-based intrusion detection method, an image-based intrusion detection device, an electronic device, and a storage medium.
Background
At present, the wire-mixing detection algorithm and the perimeter intrusion detection method are similar, and in terms of the perimeter intrusion detection algorithm, the current detection logic method generally has two methods:
logic method one: one is based on whether the human body detection frame coincides with the perimeter, and once the coincidence is judged, the human body detection frame is considered as an intrusion alarm, as shown in fig. 1, and as long as the coincidence of the human body detection frame and the perimeter area (roi) is judged in four directions of A, B, C, D, the human body detection frame is considered as perimeter intrusion;
logic method II: the other is to judge whether the two are completely overlapped, if so, the perimeter intrusion is determined;
since the object in video monitoring is presented as a 2d picture, referring to fig. 1, the personnel detection frame is rectangular, and the perimeter area is generally an irregular polygon, a standard rectangle is adopted in fig. 1 for better explanation of the problem;
for logic one, it is found from the monitoring that the personnel detection frame coincides with the perimeter frame line, and possibly enters the guard area, as in the direction C in fig. 1, the personnel may merely pass near the boundary (a certain distance is still provided from the perimeter area, but the personnel detection frame already coincides with the perimeter area partially), and according to the existing logic, the personnel detection frame is considered as intrusion, but is actually false, so that logic 1 causes more false alarms;
For the second logic, if the detection frame does not enter the perimeter area completely, but the foot enters (the foot is generally located at the bottom of the detection frame), the alarm event should be actually detected, but the intrusion cannot be calculated according to the second logic 2, so that more missed reports are caused;
such missing and false alarms can affect the actual use experience of the user and can be considered as inadequate algorithm accuracy;
disclosure of Invention
The invention aims to provide an image-based intrusion detection method, which is used for identifying shoes (or feet) of a person, carrying out event detection by judging the position relation between the shoes or the feet and a peripheral area, and judging not only by identifying the position relation between a detection frame of a human body and the peripheral area, so that the false alarm problem of a logic one can be effectively avoided, and the false alarm problem of a logic two can also be avoided; the accuracy of intrusion alert in the perimeter area is improved.
The intrusion detection method based on the image provided by the embodiment of the invention comprises the following steps:
step 1: detecting and monitoring the snap shot picture by adopting a first algorithm, and determining whether personnel exist; when personnel exist in the snap shot picture, marking a detection frame of the personnel in the snap shot picture;
Step 2: marking a perimeter region in the snap shot picture;
step 3: confirming whether the detection frame and the peripheral area are partially overlapped; executing step 4 when the detection frame is partially overlapped with the peripheral area; when the detection frame and the peripheral area part are not overlapped, no invasion occurs;
step 4: identifying the position of the shoes of the person in the detection frame of the person by adopting a second algorithm;
step 5: determining a relationship of a position of a person's shoe to a perimeter region; no intrusion occurs when the position of the person's shoe is outside the perimeter area; intrusion occurs when the location of the person's shoe is within the perimeter region.
Preferably, the detection frame for marking the personnel in the snap shot picture specifically comprises:
identifying the human body outline in the snap shot picture;
determining contour points of the most edge of the human body contour in the four directions of up, down, left and right;
and constructing a detection frame by using the contour points.
Preferably, the step of confirming whether the detection frame and the perimeter area are partially overlapped specifically includes:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the detection frame as a first coordinate set and the coordinates in the peripheral area as a second coordinate set;
comparing the first coordinate set with the second coordinate set to determine whether the same coordinate element exists; when present, the detection frame partially overlaps the perimeter region.
Preferably, determining the relationship of the position of the person's shoe to the perimeter region comprises:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the position of the shoes of the person as a third coordinate set and taking the coordinates in the peripheral area as a second coordinate set;
comparing the third coordinate set with the second coordinate set to determine whether the same coordinate element exists; when not present, then the position of the person's shoe is outside the perimeter area; when present, determining whether all coordinate elements in the third coordinate set are in the second coordinate set, and when all coordinate elements in the third coordinate set are in the second coordinate set, then the location of the person's shoe is within the perimeter region.
Preferably, determining the relationship of the position of the person's shoe to the perimeter region further comprises:
determining a duty ratio of the coordinate element in a third coordinate set, which is an intersection of the second coordinate set and the third coordinate set; when the duty ratio is larger than or equal to a preset value, the position of the shoes of the personnel is in the peripheral area; when the duty cycle is less than the preset value, then the position of the person's shoe is outside the perimeter area.
Preferably, the image-based intrusion detection method further includes:
acquiring at least two snap shots;
Acquiring the area of the shoes of the person on at least two snap-shot pictures;
acquiring the reference point of the shoes of the person in the area of the shoes of the person on at least two snap shots;
and drawing the track of the reference point, and outputting intrusion early warning information when the extension line of the track passes through the perimeter area.
Preferably, the track drawing of the reference point includes:
constructing a motion model; extracting the head position and the gesture vector of a human body, the position and the gesture vector of limbs and the position of an obstacle from the snap shot picture;
analyzing the head position and the gesture vector of the extracted human body, extracting the position and the gesture vector of limbs and predicting the position of the human body in the next snap shot picture by using a walking behavior database of the human body;
determining the position of the next reference point according to the predicted position of the human body, and fitting a track based on the position of the reference point;
the walking behavior database of the person comprises:
when the head position and the posture of the human body are opposite to the hands of the human body, predicting the position of the human body in the next snap-shot picture based on the last two snap-shot pictures;
when the head position and the posture of the human body are opposite to the obstacle, predicting the position of the human body in the next shot picture based on the last two shot pictures, and correcting the position of the human body in the next shot picture based on the position of the obstacle, so that the distance between the position of the human body in the next shot picture and the position of the obstacle is greater than the preset distance.
When the position and posture vector of the four limbs show the acceleration walking behavior, predicting the position of the human body in the next captured picture based on the last two captured pictures, and correcting the position of the human body in the next captured picture based on the preset acceleration.
Preferably, the image-based intrusion detection method further includes:
acquiring two snap shots;
acquiring the areas of the shoes of the personnel on the two snap shots;
determining whether personnel invade the peripheral area according to the areas of shoes on the two pictures; when the personnel are determined to invade the perimeter area, the invasion early-warning information is output;
determining whether personnel invade the peripheral area according to the shoe area on the two pictures; when the personnel can invade the perimeter area, the invasion early warning information is output, and the method is concretely implemented as follows:
establishing a coordinate system by taking the lower left corner of the snap-shot picture as the origin of coordinates, so that the whole picture is positioned in a first quadrant of the coordinate system;
wherein ,in the area of the shoe for the person +.>Dependent variable of the individual point fitting function, +. >In the area of the shoe for the person +.>Independent variable of the individual point fitting function, +.>In the area of the shoe for the person depicted in the first figure +.>The abscissa of the individual points,/>In the area of the shoe for the person depicted in the first figure +.>Ordinate of individual points, +.>In the area of the shoe for the person depicted in the second figure +.>The abscissa of the individual points,/>In the area of the person's shoe as depicted in the second figureOrdinate of individual points, +.>A total number of points that are the areas of the person's shoe;
wherein ,is the +.>Point and->Relation value of fitting straight line equation, +.>Is the +.>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the first picture +.>The abscissa of the individual points,/>Is the first picture +.>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
wherein ,is the +.>The person's shoe area in the second picture is marked with a dot +.>Distance of individual points, +.>Is the +.>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
The invention also provides an intrusion detection device based on the image, which comprises
The personnel marking module is used for detecting the monitored snap shot picture by adopting a first algorithm and determining whether personnel exist or not; when personnel exist in the snap shot picture, marking a detection frame of the personnel in the snap shot picture;
the perimeter region labeling module is used for labeling a perimeter region in the snap shot picture;
the pre-intrusion confirming module is used for confirming whether the detecting frame and the peripheral area are partially overlapped or not; turning to the identification module when the detection frame is partially overlapped with the perimeter region; when the detection frame and the peripheral area part are not overlapped, no invasion occurs;
The identification module is used for identifying the position of the shoes of the person in the detection frame of the person by adopting a second algorithm;
an intrusion verification module for determining a relationship of a person's shoe position to a perimeter area; no intrusion occurs when the position of the person's shoe is outside the perimeter area; intrusion occurs when the location of the person's shoe is within the perimeter region.
The present invention also provides an electronic device including: the device comprises a display screen, a processor and a memory;
the processor is electrically connected with the memory and the display screen respectively;
the memory stores instructions that, when executed by the processor, cause the processor to perform any of the image-based intrusion detection methods of the present invention.
The invention also provides a computer readable storage medium having program code stored therein, the program code being executable by a processor to invoke the method of performing any of the image-based intrusion detection methods of the invention.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic diagram of an intrusion detection method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of another intrusion detection method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an image-based intrusion detection method according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an image processing apparatus according to an embodiment of the invention.
In the figure:
11. a personnel marking module; 12. a perimeter region labeling module; 13. a pre-intrusion validation module; 14. an identification module; 15. and an intrusion confirmation module.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
The embodiment of the invention provides an intrusion detection method based on images, which is shown in fig. 2 and 3 and comprises the following steps:
Step 1: detecting and monitoring the snap shot picture by adopting a first algorithm, and determining whether personnel exist; when personnel exist in the snap shot picture, marking a detection frame of the personnel in the snap shot picture;
step 2: marking a perimeter region in the snap shot picture;
step 3: confirming whether the detection frame and the peripheral area are partially overlapped; executing step 4 when the detection frame is partially overlapped with the peripheral area; when the detection frame and the peripheral area part are not overlapped, no invasion occurs;
step 4: identifying the position of the shoes of the person in the detection frame of the person by adopting a second algorithm;
step 5: determining a relationship of a position of a person's shoe to a perimeter region; no intrusion occurs when the position of the person's shoe is outside the perimeter area; intrusion occurs when the location of the person's shoe is within the perimeter region.
The working principle and the beneficial effects of the technical scheme are as follows:
the first algorithm detects and monitors the snap shot picture, mainly identifies whether personnel exist in the picture; the first algorithm adopts an open-source edge recognition algorithm to recognize the human body. Personnel realize the demarcation perimeter area in the picture; when the detection frame of the person in the picture is overlapped with the perimeter region, a detection algorithm model (namely a second algorithm) of the shoe (foot) attribute of the person is required to be obtained through training by adopting a large amount of picture data in advance, the relation between the coordinate position of the shoe and the coordinate position of the perimeter region is judged, if the person is outside the perimeter region, the person is considered not invaded, and no event alarm is generated; if the foot enters a forbidden zone, we will call the forbidden zone invasion, and the foot will not enter the forbidden zone.
According to the image-based intrusion detection method, the shoes (or feet) of the person are identified, event detection is carried out by judging the position relation between the shoes or the feet and the peripheral area, and the position relation between the detection frame of the human body and the peripheral area is not only identified, so that the false alarm problem of the first logic can be effectively avoided, and the false alarm problem of the second logic can also be avoided; the accuracy of intrusion alert in the perimeter area is improved.
In one embodiment, the detection frame for marking the personnel in the snap shot picture specifically comprises:
identifying the human body outline in the snap shot picture;
determining contour points of the most edge of the human body contour in the four directions of up, down, left and right;
and constructing a detection frame by using the contour points.
The working principle and the beneficial effects of the technical scheme are as follows:
the detection frame is marked based on the human body outline, so that the marking of the detection frame is more accurate, and the accuracy of judging whether invasion occurs is improved.
In one embodiment, determining whether the detection frame and the perimeter region overlap partially specifically includes:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the detection frame as a first coordinate set and the coordinates in the peripheral area as a second coordinate set;
Comparing the first coordinate set with the second coordinate set to determine whether the same coordinate element exists; when present, the detection frame partially overlaps the perimeter region.
The working principle and the beneficial effects of the technical scheme are as follows:
the detection frame and the perimeter region are thinned into two different coordinate sets, and whether the detection frame and the perimeter region are partially overlapped is confirmed from the quantization relation of the coordinate sets. The accuracy of judging whether the intrusion occurs is improved.
In one embodiment, determining a relationship of a person's shoe position to a perimeter region comprises:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the position of the shoes of the person as a third coordinate set and taking the coordinates in the peripheral area as a second coordinate set;
comparing the third coordinate set with the second coordinate set to determine whether the same coordinate element exists; when not present, then the position of the person's shoe is outside the perimeter area; when present, determining whether all coordinate elements in the third coordinate set are in the second coordinate set, and when all coordinate elements in the third coordinate set are in the second coordinate set, then the location of the person's shoe is within the perimeter region.
The working principle and the beneficial effects of the technical scheme are as follows:
The shoes and the peripheral area of the person are thinned into two different coordinate sets, and whether the shoes of the person and the peripheral area are partially overlapped is confirmed from the quantitative relation of the coordinate sets. The accuracy of judging whether the intrusion occurs is improved.
In one embodiment, determining the relationship of the location of the person's shoe to the perimeter region further comprises:
determining a duty ratio of the coordinate element in a third coordinate set, which is an intersection of the second coordinate set and the third coordinate set; when the duty ratio is larger than or equal to a preset value, the position of the shoes of the personnel is in the peripheral area; when the duty cycle is less than the preset value, then the position of the person's shoe is outside the perimeter area.
The working principle and the beneficial effects of the technical scheme are as follows:
the person's shoes have not only an inner and an outer relationship with the perimeter region, but also a case that the person's shoes are on the boundary of the perimeter region, and the case that the person's shoes are on the boundary of the perimeter region is divided by introducing a preset value, so that the person's shoes are on the boundary of the perimeter region. The accuracy of judging whether the intrusion occurs is improved.
In one embodiment, the image-based intrusion detection method further comprises:
acquiring at least two snap shots;
Acquiring the area of the shoes of the person on at least two snap-shot pictures;
acquiring the reference point of the shoes of the person in the area of the shoes of the person on at least two snap shots;
and drawing the track of the reference point, and outputting intrusion early warning information when the extension line of the track passes through the perimeter area.
The working principle and the beneficial effects of the technical scheme are as follows:
drawing the motion trail of the personnel through continuously snapshot pictures, predicting whether the personnel can pass through the perimeter area, realizing early warning in advance, and preventing invasion.
In one embodiment, the locus of reference points is plotted, comprising:
constructing a motion model; extracting the head position and the gesture vector of a human body, the position and the gesture vector of limbs and the position of an obstacle from the snap shot picture;
analyzing the head position and the gesture vector of the extracted human body, extracting the position and the gesture vector of limbs and predicting the position of the human body in the next snap shot picture by using a walking behavior database of the human body;
determining the position of the next reference point according to the predicted position of the human body, and fitting a track based on the position of the reference point;
the walking behavior database of the person comprises:
when the head position and the posture of the human body are opposite to the hands of the human body, predicting the position of the human body in the next snap-shot picture based on the last two snap-shot pictures;
When the head position and the posture of the human body are opposite to the obstacle, predicting the position of the human body in the next shot picture based on the last two shot pictures, and correcting the position of the human body in the next shot picture based on the position of the obstacle, so that the distance between the position of the human body in the next shot picture and the position of the obstacle is greater than the preset distance.
When the position and posture vector of the four limbs show the acceleration walking behavior, predicting the position of the human body in the next captured picture based on the last two captured pictures, and correcting the position of the human body in the next captured picture based on the preset acceleration.
The working principle and the beneficial effects of the technical scheme are as follows:
in order to improve the accuracy of motion trail prediction of a person, the behavior of the person is predicted from the head and limbs of the person, and then the motion trail of the person is predicted.
In one embodiment, the image-based intrusion detection method further comprises:
acquiring two snap shots;
acquiring the areas of the shoes of the personnel on the two snap shots;
determining whether personnel invade the peripheral area according to the areas of shoes on the two pictures; when the personnel are determined to invade the perimeter area, the invasion early-warning information is output;
Determining whether personnel invade the peripheral area according to the shoe area on the two pictures; when the personnel can invade the perimeter area, the invasion early warning information is output, and the method is concretely implemented as follows:
establishing a coordinate system by taking the lower left corner of the snap-shot picture as the origin of coordinates, so that the whole picture is positioned in a first quadrant of the coordinate system;
wherein ,in the area of the shoe for the person +.>Dependent variable of the individual point fitting function, +.>In the area of the shoe for the person +.>Independent variable of the individual point fitting function, +.>In the area of the shoe for the person depicted in the first figure +.>The abscissa of the individual points,/>In the area of the shoe for the person depicted in the first figure +.>Ordinate of individual points, +.>In the area of the shoe for the person depicted in the second figure +.>The abscissa of the individual points,/>In the area of the person's shoe as depicted in the second figureOrdinate of individual points, +.>A total number of points that are the areas of the person's shoe;
wherein ,is the +. >Point and->Relation value of fitting straight line equation, +.>Is the +.>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the first picture +.>The abscissa of the individual points,/>Is the first picture +.>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
wherein ,is the +.>From a point to the second picture of the area of the person's shoeDistance of individual points, +.>Is the +.>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
The beneficial effects are that: in the above technical solution, two snap shots are adopted to determine whether the perimeter area is possibly invaded, if so, further calculation is performed, otherwise calculation is not needed; and further calculating the distance between the peripheral area and the area of the shoes of the personnel, and if the distance reaches a preset threshold value, invading the early warning information, otherwise, not outputting the invading early warning information. By the aid of the technology, whether the personnel possibly invades the peripheral area can be predicted in advance, and invasion early warning information can be output to remind when the shoes of the personnel are about to enter the peripheral area under the condition of possible invasion.
As shown in fig. 4, the present invention further provides an image-based intrusion detection apparatus including:
the personnel marking module 11 is used for detecting the monitored snap shot picture by adopting a first algorithm to determine whether personnel exist; when personnel exist in the snap shot picture, marking a detection frame of the personnel in the snap shot picture;
a perimeter region labeling module 12, configured to label a perimeter region in the snap shot picture;
a pre-intrusion confirmation module 13 for confirming whether the detection frame and the perimeter area are partially overlapped; turning to the identification module 14 when the detection frame partially overlaps the perimeter region; when the detection frame and the peripheral area part are not overlapped, no invasion occurs;
An identification module 14 for identifying the position of the person's shoe within the person's detection frame using a second algorithm;
an intrusion verification module 15 for determining the relationship of the position of the person's shoe to the perimeter area; no intrusion occurs when the position of the person's shoe is outside the perimeter area; intrusion occurs when the location of the person's shoe is within the perimeter region.
The working principle and the beneficial effects of the technical scheme are as follows:
the first algorithm detects and monitors the snap shot picture, mainly identifies whether personnel exist in the picture; the first algorithm adopts an open-source edge recognition algorithm to recognize the human body. Personnel realize the demarcation perimeter area in the picture; when the detection frame of the person in the picture is overlapped with the perimeter region, a detection algorithm model (namely a second algorithm) of the shoe (foot) attribute of the person is required to be obtained through training by adopting a large amount of picture data in advance, the relation between the coordinate position of the shoe and the coordinate position of the perimeter region is judged, if the person is outside the perimeter region, the person is considered not invaded, and no event alarm is generated; if the foot enters a forbidden zone, we will call the forbidden zone invasion, and the foot will not enter the forbidden zone.
The image-based intrusion detection device provided by the invention can be used for identifying the shoes (or feet) of the personnel, and carrying out event detection by judging the position relationship between the shoes or the feet and the peripheral area, so that the false alarm problem of the first logic can be effectively avoided, and the false alarm problem of the second logic can be also avoided; the accuracy of intrusion alert in the perimeter area is improved.
In one embodiment, the person tagging module 11 performs operations including:
identifying the human body outline in the snap shot picture;
determining contour points of the most edge of the human body contour in the four directions of up, down, left and right;
and constructing a detection frame by using the contour points.
The working principle and the beneficial effects of the technical scheme are as follows:
the detection frame is marked based on the human body outline, so that the marking of the detection frame is more accurate, and the accuracy of judging whether invasion occurs is improved.
In one embodiment, the pre-intrusion verification module 13 specifically includes the following operations:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the detection frame as a first coordinate set and the coordinates in the peripheral area as a second coordinate set;
comparing the first coordinate set with the second coordinate set to determine whether the same coordinate element exists; when present, the detection frame partially overlaps the perimeter region.
The working principle and the beneficial effects of the technical scheme are as follows:
the detection frame and the perimeter region are thinned into two different coordinate sets, and whether the detection frame and the perimeter region are partially overlapped is confirmed from the quantization relation of the coordinate sets. The accuracy of judging whether the intrusion occurs is improved.
In one embodiment, the intrusion verification module 15 specifically includes the following operations:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the position of the shoes of the person as a third coordinate set and taking the coordinates in the peripheral area as a second coordinate set;
comparing the third coordinate set with the second coordinate set to determine whether the same coordinate element exists; when not present, then the position of the person's shoe is outside the perimeter area; when present, determining whether all coordinate elements in the third coordinate set are in the second coordinate set, and when all coordinate elements in the third coordinate set are in the second coordinate set, then the location of the person's shoe is within the perimeter region.
The working principle and the beneficial effects of the technical scheme are as follows:
the shoes and the peripheral area of the person are thinned into two different coordinate sets, and whether the shoes of the person and the peripheral area are partially overlapped is confirmed from the quantitative relation of the coordinate sets. The accuracy of judging whether the intrusion occurs is improved.
In one embodiment, the intrusion verification module 15 further includes the following operations:
determining a duty ratio of the coordinate element in a third coordinate set, which is an intersection of the second coordinate set and the third coordinate set; when the duty ratio is larger than or equal to a preset value, the position of the shoes of the personnel is in the peripheral area; when the duty cycle is less than the preset value, then the position of the person's shoe is outside the perimeter area.
The working principle and the beneficial effects of the technical scheme are as follows:
the person's shoes have not only an inner and an outer relationship with the perimeter region, but also a case that the person's shoes are on the boundary of the perimeter region, and the case that the person's shoes are on the boundary of the perimeter region is divided by introducing a preset value, so that the person's shoes are on the boundary of the perimeter region. The accuracy of judging whether the intrusion occurs is improved.
In one embodiment, the image-based intrusion detection apparatus further comprises: a prediction module;
the prediction module performs operations including:
acquiring at least two snap shots;
acquiring the area of the shoes of the person on at least two snap-shot pictures;
acquiring the reference point of the shoes of the person in the area of the shoes of the person on at least two snap shots;
And drawing the track of the reference point, and outputting intrusion early warning information when the extension line of the track passes through the perimeter area.
The working principle and the beneficial effects of the technical scheme are as follows:
drawing the motion trail of the personnel through continuously snapshot pictures, predicting whether the personnel can pass through the perimeter area, realizing early warning in advance, and preventing invasion.
In one embodiment, the prediction module further performs operations comprising:
constructing a motion model; extracting the head position and the gesture vector of a human body, the position and the gesture vector of limbs and the position of an obstacle from the snap shot picture;
analyzing the head position and the gesture vector of the extracted human body, extracting the position and the gesture vector of limbs and predicting the position of the human body in the next snap shot picture by using a walking behavior database of the human body;
determining the position of the next reference point according to the predicted position of the human body, and fitting a track based on the position of the reference point;
the walking behavior database of the person comprises:
when the head position and the posture of the human body are opposite to the hands of the human body, predicting the position of the human body in the next snap-shot picture based on the last two snap-shot pictures;
when the head position and the posture of the human body are opposite to the obstacle, predicting the position of the human body in the next shot picture based on the last two shot pictures, and correcting the position of the human body in the next shot picture based on the position of the obstacle, so that the distance between the position of the human body in the next shot picture and the position of the obstacle is greater than the preset distance.
When the position and posture vector of the four limbs show the acceleration walking behavior, predicting the position of the human body in the next captured picture based on the last two captured pictures, and correcting the position of the human body in the next captured picture based on the preset acceleration.
The working principle and the beneficial effects of the technical scheme are as follows:
in order to improve the accuracy of motion trail prediction of a person, the behavior of the person is predicted from the head and limbs of the person, and then the motion trail of the person is predicted.
In one embodiment, the prediction module further performs operations comprising:
acquiring two snap shots;
acquiring the areas of the shoes of the personnel on the two snap shots;
determining whether personnel invade the peripheral area according to the areas of shoes on the two pictures; when the personnel are determined to invade the perimeter area, the invasion early-warning information is output;
determining whether personnel invade the peripheral area according to the shoe area on the two pictures; when the personnel can invade the perimeter area, the invasion early warning information is output, and the method is concretely implemented as follows:
establishing a coordinate system by taking the lower left corner of the snap-shot picture as the origin of coordinates, so that the whole picture is positioned in a first quadrant of the coordinate system;
wherein ,in the area of the shoe for the person +.>Dependent variable of the individual point fitting function, +.>In the area of the shoe for the person +.>Independent variable of the individual point fitting function, +.>In the area of the shoe for the person depicted in the first figure +.>The abscissa of the individual points,/>In the area of the shoe for the person depicted in the first figure +.>Ordinate of individual points, +.>In the area of the shoe for the person depicted in the second figure +.>The abscissa of the individual points,/>In the area of the person's shoe as depicted in the second figureOrdinate of individual points, +.>A total number of points that are the areas of the person's shoe;
wherein ,is the +.>Point and->Relation value of fitting straight line equation, +.>Is the +.>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the first picture +.>The abscissa of the individual points,/>Is the first picture +.>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/ >Is the +.>The ordinate of the individual points;
wherein ,is the +.>From a point to the second picture of the area of the person's shoeDistance of individual points, +.>For the circumference ofFirst%>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
The beneficial effects are that: in the above technical solution, two snap shots are adopted to determine whether the perimeter area is possibly invaded, if so, further calculation is performed, otherwise calculation is not needed; and further calculating the distance between the peripheral area and the area of the shoes of the personnel, and if the distance reaches a preset threshold value, invading the early warning information, otherwise, not outputting the invading early warning information. By the aid of the technology, whether the personnel possibly invades the peripheral area can be predicted in advance, and invasion early warning information can be output to remind when the shoes of the personnel are about to enter the peripheral area under the condition of possible invasion.
The present invention also provides an electronic device including: the device comprises a display screen, a processor and a memory;
the processor is electrically connected with the memory and the display screen respectively;
the memory stores instructions that, when executed by the processor, cause the processor to perform any of the image-based intrusion detection methods of the present invention.
The invention also provides a computer readable storage medium having program code stored therein, the program code being executable by a processor to invoke the method of performing any of the image-based intrusion detection methods of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (8)
1. An image-based intrusion detection method, comprising:
step 1: detecting and monitoring the snap shot picture by adopting a first algorithm, and determining whether personnel exist; when personnel exist in the snap shot picture, marking a detection frame of the personnel in the snap shot picture;
step 2: marking a perimeter region in the snap shot picture;
Step 3: confirming whether the detection frame and the perimeter area are partially overlapped; executing step 4 when the detection frame is partially overlapped with the perimeter area; when the detection frame and the peripheral area part are not overlapped, no invasion occurs;
step 4: identifying the position of the shoes of the person in the detection frame of the person by adopting a second algorithm;
step 5: determining a relationship of a position of the person's shoe to the perimeter region; no intrusion occurs when the location of the person's shoe is outside the perimeter region; intrusion occurs when the location of the person's shoe is within the perimeter region;
the image-based intrusion detection method further comprises the following steps:
acquiring at least two snap shots;
acquiring the areas of the shoes of the personnel on the at least two snap shots;
determining a reference point of the shoes of the person in the area of the shoes of the person acquired on the at least two snap shots;
drawing the track of the reference point, and outputting intrusion early warning information when an extension line of the track passes through the perimeter area;
the track drawing of the reference point comprises the following steps:
Constructing a motion model; extracting the head position and the posture vector of the human body, the position and the posture vector of the limbs and the obstacle position from the snap shot picture;
analyzing the head position and the gesture vector of the extracted human body, extracting the position and the gesture vector of limbs and predicting the position of the human body in the next snap-shot picture by using a walking behavior database of the human body;
determining the position of a next reference point according to the predicted position of the human body, and fitting the track based on the position of the reference point;
the walking behavior database of the person comprises:
when the head position and the posture of the human body are opposite to the hands of the human body, predicting the position of the human body in the next snap shot picture based on the last two snap shot pictures;
when the head position and the posture of the human body are opposite to the obstacle, predicting the position of the human body in the next snap-shot picture based on the last two snap-shot pictures, and correcting the position of the human body in the next snap-shot picture based on the position of the obstacle, so that the distance between the position of the human body in the next snap-shot picture and the position of the obstacle is larger than a preset distance;
When the position and posture vector of the four limbs show the acceleration walking behavior, predicting the position of the human body in the next snap-shot picture based on the last two snap-shot pictures, and correcting the position of the human body in the next snap-shot picture based on the preset acceleration.
2. The image-based intrusion detection method according to claim 1, wherein the marking of the detection frame of the person in the snap shot picture specifically comprises:
identifying the human body outline in the snap shot picture;
determining contour points of the most edge of the human body contour in the four directions of up, down, left and right;
and constructing the detection frame by the contour points.
3. The image-based intrusion detection method of claim 1, wherein confirming whether the detection frame and the perimeter region overlap in part comprises:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the detection frame as a first coordinate set and the coordinates in the perimeter area as a second coordinate set;
comparing the first coordinate set and the second coordinate set to determine whether the same coordinate element exists; when present, the detection frame partially overlaps the perimeter region.
4. The image-based intrusion detection method of claim 1, wherein determining the relationship of the location of the person's shoe to the perimeter area comprises:
establishing a coordinate system by taking the center point of the snap-shot picture as an origin;
taking the coordinates in the position of the shoes of the person as a third coordinate set and taking the coordinates in the peripheral area as a second coordinate set;
comparing the third coordinate set with the second coordinate set to determine whether the same coordinate element exists; when not present, then the location of the person's shoe is outside of the perimeter region; determining if all coordinate elements in the third set of coordinates are in the second set of coordinates when present, and if all coordinate elements in the third set of coordinates are in the second set of coordinates, then the location of the person's shoe is within the perimeter region;
and/or the number of the groups of groups,
determining a duty cycle of the coordinate element in the third coordinate set, which is an intersection of the second coordinate set and the third coordinate set; when the duty ratio is larger than or equal to a preset value, the position of the shoes of the personnel is in the peripheral area; when the duty cycle is less than the preset value, then the position of the person's shoe is outside the perimeter region.
5. The image-based intrusion detection method of claim 1, further comprising:
acquiring two snap shots;
acquiring the areas of the shoes of the personnel on the two snap shots;
determining whether personnel invade the peripheral area according to the areas of shoes on the two pictures; when the personnel are determined to invade the perimeter area, the invasion early-warning information is output;
determining whether personnel invade the peripheral area according to the shoe area on the two pictures; when the personnel can invade the perimeter area, the invasion early warning information is output, and the method is concretely implemented as follows:
establishing a coordinate system by taking the lower left corner of the snap-shot picture as the origin of coordinates, so that the whole picture is positioned in a first quadrant of the coordinate system;
step 11, calculating a fitting function equation of corresponding coordinates of the areas of the shoes of the person in the two snap pictures;
wherein ,in the area of the shoe for the person +.>Dependent variable of the individual point fitting function, +.>In the area of the shoe for the person +.>Independent variable of the individual point fitting function, +.>In the area of the shoe for the person depicted in the first figure +.>The abscissa of the individual points,/ >In the area of the shoe for the person depicted in the first figure +.>Ordinate of individual points, +.>In the area of the shoe for the person depicted in the second figure +.>The abscissa of the individual points,/>In the area of the shoe for the person depicted in the second figure +.>Ordinate of individual points, +.>A total number of points that are the areas of the person's shoe;
step 12, calculating a relation value between the perimeter region and a fitting linear equation;
wherein ,is the +.>Point and->Relation value of fitting straight line equation, +.>Is the +.>The abscissa of the individual points,/>The>Ordinate of individual points, +.>Is the first picture in the first figureThe abscissa of the individual points,/>Is the first picture +.>Ordinate of individual points, +.>Is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
step 13, when the relation value is equal to zero, judging that the personnel possibly invades the peripheral area, and carrying out step 14, otherwise, judging that the personnel does not invade the peripheral area;
step 14, calculating the distance between the point of the area of the shoe of the person and the midpoint of the peripheral area in the second picture;
wherein ,is the +.>The person's shoe area in the second picture is marked with a dot +.>Distance of individual points, +.>Is the +.>The abscissa of the individual points,/>The>The ordinate of the individual points is taken to mean,is the +.>The abscissa of the individual points,/>Is the +.>The ordinate of the individual points;
step 15, when there isWhen the intrusion detection value is smaller than or equal to a preset threshold value, intrusion early warning information is output, and when the intrusion detection value is +.>And when the intrusion detection value is larger than the preset threshold value, no intrusion early warning information is output.
6. An image-based intrusion detection device, comprising
The personnel marking module (11) is used for detecting the pictures shot by monitoring by adopting a first algorithm and determining whether personnel exist or not; when personnel exist in the snap shot picture, marking a detection frame of the personnel in the snap shot picture;
the perimeter region labeling module (12) is used for labeling a perimeter region in the snap shot picture;
a pre-intrusion confirmation module (13) for confirming whether the detection frame and the perimeter region are partially overlapped; turning to an identification module (14) when the detection frame partially overlaps the perimeter region; when the detection frame and the peripheral area part are not overlapped, no invasion occurs;
The identification module (14) is used for identifying the position of the shoes of the person in the detection frame of the person by adopting a second algorithm;
an intrusion confirmation module (15) for determining a relationship of a position of the person's shoe to the perimeter area; no intrusion occurs when the location of the person's shoe is outside the perimeter region; intrusion occurs when the location of the person's shoe is within the perimeter region;
furthermore, the image-based intrusion detection apparatus further includes: a prediction module;
the prediction module performs operations including:
acquiring at least two snap shots;
acquiring the area of the shoes of the person on at least two snap-shot pictures;
acquiring the reference point of the shoes of the person in the area of the shoes of the person on at least two snap shots;
drawing a track of the reference point, and outputting intrusion early warning information when an extension line of the track passes through the perimeter area;
the prediction module performs track drawing of the reference point, and the operations comprise the following steps:
constructing a motion model; extracting the head position and the gesture vector of a human body, the position and the gesture vector of limbs and the position of an obstacle from the snap shot picture;
Analyzing the head position and the gesture vector of the extracted human body, extracting the position and the gesture vector of limbs and predicting the position of the human body in the next snap shot picture by using a walking behavior database of the human body;
determining the position of the next reference point according to the predicted position of the human body, and fitting a track based on the position of the reference point;
the walking behavior database of the person comprises:
when the head position and the posture of the human body are opposite to the hands of the human body, predicting the position of the human body in the next snap-shot picture based on the last two snap-shot pictures;
when the head position and the posture of the human body are opposite to the obstacle, predicting the position of the human body in the next shot picture based on the last two shot pictures, and correcting the position of the human body in the next shot picture based on the position of the obstacle, so that the distance between the position of the human body in the next shot picture and the position of the obstacle is greater than the preset distance;
when the position and posture vector of the four limbs show the acceleration walking behavior, predicting the position of the human body in the next captured picture based on the last two captured pictures, and correcting the position of the human body in the next captured picture based on the preset acceleration.
7. An electronic device, comprising: the device comprises a display screen, a processor and a memory;
the processor is electrically connected with the memory and the display screen respectively;
the memory stores instructions that, when executed by the processor, cause the processor to perform the method of any of claims 1 to 5.
8. A computer readable storage medium, characterized in that a computer readable storage medium has stored therein a program code which is callable by a processor for performing the method according to any one of claims 1-5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010108343.5A CN111339901B (en) | 2020-02-21 | 2020-02-21 | Image-based intrusion detection method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010108343.5A CN111339901B (en) | 2020-02-21 | 2020-02-21 | Image-based intrusion detection method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111339901A CN111339901A (en) | 2020-06-26 |
CN111339901B true CN111339901B (en) | 2023-06-09 |
Family
ID=71184212
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010108343.5A Active CN111339901B (en) | 2020-02-21 | 2020-02-21 | Image-based intrusion detection method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111339901B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112150749A (en) * | 2020-09-08 | 2020-12-29 | 成都红云鼎科技有限公司 | Layered early warning method based on multiple defense areas |
US20220292281A1 (en) * | 2021-03-15 | 2022-09-15 | Toniya Anthonipillai | Loitering and Vagrancy Computer Vision Ai |
CN113343856B (en) * | 2021-06-09 | 2022-03-29 | 北京容联易通信息技术有限公司 | Image recognition method and system |
CN113888824A (en) * | 2021-06-30 | 2022-01-04 | 华北理工大学 | Mine violation border crossing behavior identification and alarm method and device based on topological spatial relationship |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005227854A (en) * | 2004-02-10 | 2005-08-25 | Sumitomo Electric Ind Ltd | Apparatus for sensing intrusion through entrance |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150009031A1 (en) * | 2013-07-03 | 2015-01-08 | Honeywell International Inc. | Multilayer perimeter instrusion detection system for multi-processor sensing |
CN108253958B (en) * | 2018-01-18 | 2020-08-11 | 亿嘉和科技股份有限公司 | Robot real-time positioning method in sparse environment |
CN108734091A (en) * | 2018-03-30 | 2018-11-02 | 暨南大学 | Compartment anomaly detection method, computer installation and computer readable storage medium |
CN109257569B (en) * | 2018-10-24 | 2020-06-05 | 广东佳鸿达科技股份有限公司 | Security video monitoring and analyzing method |
CN109618287B (en) * | 2018-12-29 | 2021-06-15 | 杭州平普智能科技有限公司 | Prisoner management method and device based on UWB positioning |
CN109828588A (en) * | 2019-03-11 | 2019-05-31 | 浙江工业大学 | Paths planning method in a kind of robot chamber based on Multi-sensor Fusion |
CN110110657A (en) * | 2019-05-07 | 2019-08-09 | 中冶赛迪重庆信息技术有限公司 | Method for early warning, device, equipment and the storage medium of visual identity danger |
CN110370273B (en) * | 2019-06-27 | 2021-04-09 | 华为技术有限公司 | Robot obstacle avoidance method, device and system |
CN110532852B (en) * | 2019-07-09 | 2022-10-18 | 长沙理工大学 | Subway station pedestrian abnormal event detection method based on deep learning |
CN110688987B (en) * | 2019-10-16 | 2022-03-25 | 山东建筑大学 | Pedestrian position detection and tracking method and system |
-
2020
- 2020-02-21 CN CN202010108343.5A patent/CN111339901B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005227854A (en) * | 2004-02-10 | 2005-08-25 | Sumitomo Electric Ind Ltd | Apparatus for sensing intrusion through entrance |
Also Published As
Publication number | Publication date |
---|---|
CN111339901A (en) | 2020-06-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111339901B (en) | Image-based intrusion detection method and device, electronic equipment and storage medium | |
CN109657533B (en) | Pedestrian re-identification method and related product | |
JP5432983B2 (en) | Entrance / exit detection device, monitoring device, and entrance / exit detection method | |
CN103942850B (en) | Based on medical personnel's monitoring method on duty of video analysis and RFID technique | |
US20190347486A1 (en) | Method and apparatus for detecting a garbage dumping action in real time on video surveillance system | |
CN106803083B (en) | Pedestrian detection method and device | |
JP6385419B2 (en) | Object detection device | |
KR101972356B1 (en) | An apparatus and a method for detecting upper body | |
CN111738240A (en) | Region monitoring method, device, equipment and storage medium | |
CN111753724A (en) | Abnormal behavior identification method and device | |
CN110490171B (en) | Dangerous posture recognition method and device, computer equipment and storage medium | |
WO2018179202A1 (en) | Information processing device, control method, and program | |
CN112163470A (en) | Fatigue state identification method, system and storage medium based on deep learning | |
US20230135400A1 (en) | Method and system for facial recognition | |
US20110280442A1 (en) | Object monitoring system and method | |
CN113223046A (en) | Method and system for identifying prisoner behaviors | |
US9064172B2 (en) | System and method for object detection | |
CN116311084B (en) | Crowd gathering detection method and video monitoring equipment | |
CN111489380A (en) | Target object track analysis method | |
CN113646803A (en) | Human detection device and human detection method | |
CN111325937B (en) | Method, device and electronic system for detecting crossing behavior | |
CN112104838B (en) | Image distinguishing method, monitoring camera and monitoring camera system thereof | |
CN114764895A (en) | Abnormal behavior detection device and method | |
CN116229502A (en) | Image-based tumbling behavior identification method and equipment | |
CN113989914B (en) | Security monitoring method and system based on face recognition |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |