CN117671576A - Imaging processing method for identifying dangerous target - Google Patents
Imaging processing method for identifying dangerous target Download PDFInfo
- Publication number
- CN117671576A CN117671576A CN202310789151.9A CN202310789151A CN117671576A CN 117671576 A CN117671576 A CN 117671576A CN 202310789151 A CN202310789151 A CN 202310789151A CN 117671576 A CN117671576 A CN 117671576A
- Authority
- CN
- China
- Prior art keywords
- target
- dangerous
- personnel
- processing unit
- image
- 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
- 238000003384 imaging method Methods 0.000 title claims abstract description 39
- 238000003672 processing method Methods 0.000 title claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 57
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000009471 action Effects 0.000 claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 11
- 230000006399 behavior Effects 0.000 claims description 41
- 238000005286 illumination Methods 0.000 claims description 8
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000002372 labelling Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 230000001815 facial effect Effects 0.000 claims description 4
- 230000019771 cognition Effects 0.000 claims description 3
- 230000004931 aggregating effect Effects 0.000 claims description 2
- 230000002776 aggregation Effects 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 claims description 2
- 238000004148 unit process Methods 0.000 claims description 2
- 238000013459 approach Methods 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000010009 beating Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 231100001261 hazardous Toxicity 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
Landscapes
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses an imaging processing method for identifying dangerous targets, which is based on dangerous personnel characteristics and behavior interpretation, does not depend on information such as photos of identified personnel stored in advance, but is based on imaging detection and image processing technology approaches, and obtains evidence of participation in dangerous actions by directly obtaining personnel video image information of an event scene, so as to realize identification of whether the personnel targets are dangerous targets; the behavior judgment of personnel can be automatically realized in a large range; can be used for accurately positioning personnel and taking evidence by photographing. The invention has accurate result; the dangerous behavior of any person can be timely judged without storing information such as photos of the identified person in advance; the method can utilize the automatic and intelligent results of image processing, is rapid in judgment and is convenient for taking pictures and evidence.
Description
Technical Field
The invention belongs to the technical field of photoelectric imaging, and relates to an imaging processing method for identifying dangerous targets.
Background
The current automatic personnel identification method generally captures facial images or fingerprint information of the identified personnel, and compares the facial images or fingerprint information with information such as a pre-stored personnel identity card image or a pre-acquired fingerprint to identify whether the identified personnel are purported persons.
The application number 201810379291.8 discloses a personnel identification method based on multiple gestures, which is used for assisting in confirming a specific user by mobile phone software. Application number 202010313368.9 discloses an intelligent personnel identification control cabinet, wherein a personnel identification device is arranged to adopt biological information such as a face, a fingerprint and the like to carry out identity authentication (compare with stored information) so as to confirm whether the intelligent personnel identification control cabinet has the authority of powering on operation equipment. Application number 201710726355.2 relates to a laboratory management system and method based on personnel identification, the system comprises an intelligent positioning tag, an intelligent tag card sender, an intelligent tag identification device, a personnel identification device, a data processing device, an alarm device and the like, and whether the personnel has authority to enter a laboratory is identified according to the detection of the intelligent tag. Application number 201810925887.3 relates to a method of grouping repeated scene images, running a repeated scene image grouping platform for grouping images, the grouping platform comprising an editorial staff identification device for determining the ethnicity of a human body in a captured image based on contours of the human body identified from the captured image. Application number 202110464477.5 discloses a method, a system, a device and a storage medium for identifying lost personnel, wherein an intelligent camera is utilized to perform control in various public places where personnel are easy to lose, and lost personnel and suspicious personnel are automatically identified by comparing stored lost personnel and mental patient data. Application number 202010541240.8 describes an indoor personnel identification and activity monitoring device comprising an infrared detection module and a personnel identification unit, identifying non-indoor personnel based on stored indoor personnel data. Application number 201810856757.9 describes a mall monitoring system including video monitoring devices, smoke detection devices, etc. to identify potential thieves by comparing whether personnel clothing is consistent with mall work clothes.
These devices and methods are related to the comparison of existing stored information (staff face, fingerprint, clothing, or patient and lost person information), or the identification of staff targets by a label detection method, respectively, and do not involve the problem of identifying target dangerous attributes based on images.
Disclosure of Invention
Object of the invention
The purpose of the invention is that: the imaging processing method for identifying the dangerous target is based on the characteristic and behavior interpretation of the dangerous target personnel, does not depend on the information such as the photo of the identified personnel stored in advance, but is based on the imaging detection and image processing technology approach, and the evidence of participation in dangerous actions is obtained by directly obtaining the personnel video image information of the event scene, so that the identification of whether the personnel target is the dangerous target is realized; the behavior judgment of personnel can be automatically realized in a large range; can be used for accurately positioning personnel and taking evidence by photographing.
(II) technical scheme
In order to solve the technical problems, the invention provides an imaging processing method for identifying dangerous targets, which is used for acquiring video images and target distance and position information based on a set of photoelectric equipment, wherein the photoelectric equipment comprises a continuous zooming camera, a continuous zooming thermal infrared imager, a laser range finder, a stable turntable, a control unit and a processing unit; the stable turntable is arranged on the bearing platform, so that the aiming line of the photoelectric sensor is stable, the camera and the thermal infrared imager can correspondingly obtain a scene and clear images of people in the scene, the laser range finder correspondingly obtains the accurate relative position of the target, the angle measuring mechanism and the positioning device in the stable turntable obtain the observation position of the bearing platform, the relative azimuth, pitching and other angles of the target, the processing unit processes the image and the data to obtain the attribute and the position coordinate of the target, and the control unit receives an external control instruction to enable the photoelectric equipment to select different working modes, observation angles, view field sizes and the like. The imaging processing method utilizes the target video image and the position information obtained by the photoelectric equipment, and identifies whether the target is in accordance with dangerous target characteristics or not and has dangerous behaviors by reflecting the characteristics of the target in the image and the target actions reflected by a plurality of images, collects evidence and implements tracking.
The imaging processing method mainly comprises the following steps:
step 1, defining characteristics and behavior combinations of personnel targets belonging to dangerous targets based on cognition of dangerous targets and dangerous behaviors, forming corresponding relations, and storing the corresponding relations in a processing unit of photoelectric equipment;
step 2, when working, the photoelectric equipment arrives at an event occurrence area along with the bearing platform, or is fixed on bearing devices such as buildings, telegraph poles and the like in the concerned area; an operator controls the photoelectric equipment to continuously observe a region of interest to obtain continuous images, and a camera is selected to observe a scene under the condition of good daytime or illumination, and a thermal infrared imager is selected to observe the scene under the condition of night or poor illumination; the processing unit detects personnel targets in the image, and white frame selection marking and tracking target display are adopted; the processing unit calculates the position coordinates of each target according to the data of the laser range finder, the stable platform positioning device and the angle measuring mechanism;
step 3, the processing unit compares the target image, the continuous multiple images and the stored characteristic combination, and judges whether the target belongs to dangerous target attributes and dangerous behaviors or not: when the resolution of the obtained target image is high enough, the processing unit compares whether the face of the target person is provided with a mask or not, and whether dangerous objects are held on the hand or not; detecting a person target with a mask and dangerous objects, and judging the person target as a potential dangerous object by the processing unit; the processing unit carries out yellow frame selection labeling and tracking display on the target, and the photoelectric equipment continuously tracks and displays potential attributes of the target;
step 4, when the resolution ratio of the obtained target image is insufficient to identify the characteristics of the face and the handhold object of the personnel, an operator sends a zooming instruction to a control unit, the control unit enlarges the focal length of an objective lens of a camera or a thermal infrared imager to enlarge the size of the target, the resolution ratio is improved, a processing unit identifies a mask or a dangerous object, yellow frame selection marking and tracking display are correspondingly carried out, and photoelectric equipment continuously tracks and displays potential attributes of the mask or dangerous object;
step 5, whether the image resolution can identify the target face mask or the handheld dangerous object, namely whether the detected personnel target is a potential dangerous target (white frame selection or yellow frame selection), the processing unit performs front-back contrast analysis on the multi-frame image with the continuous observation area, and detects and judges the action of the personnel target in the image:
(1) The personnel target moves to the event area and is interpreted as aggregation;
(2) The object is separated from the limb of the personnel target, and the object moves rapidly to other personnel, vehicles, buildings and the like, and the object throwing behavior is judged;
(3) And the personnel target handhelds move to other personnel, vehicles, buildings and the like rapidly, and the personnel target handhelds are judged to be the object holding and smashing actions.
Step 6, the processing unit judges the dangerous target attribute and the behavior of the personnel target according to the personnel target behavior obtained by personnel target feature detection comparison and front-back image comparison, judges the personnel with one of the behaviors in the step 5 as the dangerous target, and adopts red frame selection identification and tracking display; outputting a judging result;
and 7, an operator adjusts the focal length of the camera or the thermal infrared imager, enlarges the red frame to select a target, intercepts a photo of a dangerous target, and realizes photo evidence collection.
(III) beneficial effects
The imaging processing method for identifying dangerous objects provided by the technical scheme has the following steps
The beneficial effects are that:
firstly, based on cognition of police officers, photoelectric imaging detection and image processing technology approaches, whether personnel targets in images have dangerous target characteristics or not and whether the actions of the personnel targets reflected by a plurality of images are dangerous targets or not are automatically judged, and the result is accurate;
secondly, the information such as the photo of the identified person is not required to be stored in advance, the information of the existing criminals is not relied on, and the dangerous behavior of any person can be judged in time;
thirdly, the automatic and intelligent achievement of image processing can be utilized, judgment is quick, and photographing and evidence obtaining are facilitated.
Drawings
FIG. 1 is a schematic diagram of an imaging process flow for identifying a hazardous target.
FIG. 2 is a diagram of the interaction of imaging process information to identify a dangerous object.
Detailed Description
To make the objects, contents and advantages of the present invention more apparent, the following detailed description of the present invention will be given with reference to the accompanying drawings and examples.
As shown in fig. 1, the imaging processing method for identifying a dangerous target in this embodiment is described in an embodiment of onboard application and handling of dangerous events at a railway station, and images and videos processed in this embodiment are acquired through legal compliance paths and are known by image and video objects.
The airborne photoelectric equipment is a two-axis stable turntable, the weight of the equipment is 33kg, the equipment is borne by an unmanned helicopter, the stability precision is 0.05mrad, and the equipment is connected with a ground station through an unmanned airborne data chain; the control personnel observe the reconnaissance image and information of the photoelectric equipment at the ground station, and control the photoelectric equipment through the keyboard and the handle; the digital camera of the photoelectric equipment has the sensor resolution of 1920 multiplied by 1080, the frame frequency of 25Hz, and is provided with a continuous zoom objective lens, and the field of view is in the range of 1.2 degrees multiplied by 0.68-35.6 degrees multiplied by 20 degrees; the infrared thermal imager of the photoelectric equipment has the sensor resolution of 1280 multiplied by 1024 and the frame frequency of 25Hz, works in a medium wave infrared band, is provided with a continuous zoom objective lens, and has the field range of 1.8 degrees multiplied by 1.44 degrees multiplied by 20 degrees; a laser range finder of photoelectric equipment, the working wave band is 1.54 mu m, and the range finding precision is +/-1 m; the photoelectric equipment is provided with a Beidou receiver and a fiber-optic gyroscope strapdown inertial navigation device, the ground station is provided with Beidou foundation enhancement equipment, the airborne Beidou positioning precision is not lower than 1m, and the north-seeking precision of the inertial navigation device is 0.8mrad; the photoelectric equipment is observed in an air-to-ground mode, the range of not less than 1058mx1000m can be detected at the height of 3km from the ground, and the positioning accuracy of the system to a target at the position of 1km can be within 5m (CEP) by utilizing the foundation enhanced Beidou receiver.
The imaging processing method for identifying dangerous objects in the embodiment comprises the following steps:
step 1, defining characteristics and behavior combinations of personnel targets belonging to dangerous targets based on knowledge of dangerous targets and dangerous behaviors, including but not limited to:
(1) Wearing a mask to shield facial features;
(2) Aggregating to an event area;
(3) The object is hit by the object holding device.
Writing the characteristics and the behaviors into corresponding relations, and storing the corresponding relations in a processing unit of the photoelectric equipment;
step 2, when a task is executed, the photoelectric equipment reaches a railway station square along with the unmanned helicopter, hovers in the air of 260 m-3 km and is subjected to ground reconnaissance; an operator rides on the vehicle-mounted ground station to reach a safety guard area outside the railway station, controls photoelectric equipment on the control seat, observes the displayed image and continuously scouts the region of interest; under the condition of daytime or good illumination, a camera is selected to observe the scene, and under the condition of night or poor illumination, a thermal infrared imager is selected to observe the scene; the processing unit detects personnel targets in the image, and white frame selection marking and tracking target display are adopted; the processing unit calculates the position coordinates of each target according to the data of the laser range finder, the stable platform positioning device and the angle measuring mechanism;
step 3, the processing unit compares the target image, the continuous multiple images and the stored characteristic combination, and can automatically judge whether the target belongs to dangerous target attributes and dangerous behaviors or not: when the obtained target face size is not smaller than 10 pixels, the processing unit compares whether the face of the target face of the person is worn with a mask, and the processing unit judges that the face is potentially dangerous; comparing whether personnel hold dangerous objects, detecting personnel targets with dangerous objects, and judging the personnel targets as potential hazards by a processing unit; for the potential targets automatically judged by the processing unit, the processing unit carries out yellow frame selection labeling and tracking display on the targets, and the photoelectric equipment continuously tracks and displays potential attributes of the targets;
step 4, when the resolution of the obtained target image is insufficient to identify the characteristics of the face and the handhold object of the personnel, an operator sends a zooming instruction to a control unit, the control unit enlarges the focal length of an objective lens of a camera or a thermal infrared imager, so that the size of the target is enlarged, the resolution is improved to be more than 10 pixels of the judged characteristics, a processing unit identifies a mask or a dangerous object, yellow frame selection marking and tracking display are correspondingly carried out, and the photoelectric equipment continuously tracks and displays potential attributes of the mask or the dangerous object;
step 5, whether the detected personnel target is a potential dangerous target (white frame selection or yellow frame selection) or not, the processing unit performs front-back comparison analysis on the multi-frame image with the continuous observation area, and detects and judges the behavior of the personnel target in the image:
(1) The personnel target moves to the railway station square instead of escaping from the dangerous area, and is interpreted as gathering;
(2) The object is separated from the limb of the personnel target, and the object moves rapidly to other personnel, stores beside the railway station and restaurant buildings, and the object throwing behavior is judged;
(3) The personnel target handhelds move to quickly move to other personnel, vehicles, buildings and the like, and the action of holding the objects and beating the objects is judged;
when the resolution is insufficient and cannot be judged, the operator further increases the focal length of the objective lens of the imaging device, or controls the bearing platform to approach the target, so that the number of imaging pixels where the target is positioned is increased, the number of pixels occupied by the mask and dangerous goods is more than 10 pixels, and the number of pixels where the characters are positioned is not less than 10×10.
Step 6, the processing unit judges the dangerous target attribute and the behavior of the personnel target according to the personnel target behavior obtained by personnel target feature detection comparison and front-back image comparison, judges the personnel with one of the behaviors in the step 5 as the dangerous target, and adopts red frame selection identification and tracking display; outputting a judging result;
and 7, an operator adjusts the focal length of the camera or the thermal infrared imager, enlarges the red frame to select the target, intercepts the photo with the characteristics of the face and the whole body of the dangerous target, realizes photographing and evidence obtaining, and reports the target coordinates to related personnel.
The dangerous behavior can be further increased and defined by comparing front and rear images so that the processing unit can automatically detect the dangerous behavior;
the target mark can be in other colors, can be selected in a frame mode, can be selected in a circle mode or can be marked in other modes;
the scientific research of the stable turntable is an inertial stable turntable with a two-axis structure, and can also be an inertial stable turntable with a three-axis structure;
the imaging equipment of the camera and the thermal infrared imager can adopt one type; the infrared thermal imager is not limited to work in the middle-wave infrared, the long-wave infrared or the double-color infrared band; instead of one or both imaging devices, short-wave infrared cameras can also be used.
In the step 3 of the method, the operator can judge the potential dangerous target by himself, and the operator identifies the potential target (yellow box selection mark and tracking display);
in steps 5 and 6 of the method, the operator can determine whether the operator target has dangerous behavior (red frame selection identification and tracking display) by himself.
As shown in fig. 2, the imaging processing method for identifying the dangerous target mainly exchanges information between the control station and the photoelectric equipment, between the imaging equipment of the photoelectric equipment and other equipment, and between the control station and the action team;
the control station sends an instruction of a control person to the photoelectric equipment, controls the photoelectric equipment to switch the sensor, adjusts the focal length of the objective lens to change the observation field of view and searches the area; acquiring a scene, a target image, target position (coordinate) information and identification information which are acquired from photoelectric equipment;
the photoelectric equipment acquires a scene video image by utilizing imaging equipment (a television camera, a thermal infrared imager and the like) integrated with the photoelectric equipment, and calculates the distance of a target by utilizing a laser range finder, a measuring angle of a stable turntable and a positioning device; the control unit is used for processing and responding to the instruction of the control station, changing the sensor and the working mode, adjusting the field of view and rotating search; the processing unit detects the obtained images, compares the obtained personnel targets, dangerous target characteristics, dangerous behaviors and the like, judges the attribute of the targets, and captures a photo to obtain evidence; and sending a judgment result to the control station.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (10)
1. The imaging processing method for identifying the dangerous target is characterized by acquiring a target video image, a target distance and position information based on a set of photoelectric equipment, wherein the photoelectric equipment comprises a continuous zooming camera, a continuous zooming thermal infrared imager, a laser range finder, a stable turntable, a control unit and a processing unit; the stable turntable is arranged on the bearing platform, so that the aiming line of the photoelectric sensor is stable, the camera and the thermal infrared imager correspondingly acquire a scene and clear images of people in the scene, the laser range finder correspondingly acquires the accurate relative position of the target, the angle measuring mechanism and the positioning device in the stable turntable acquire the observation position, the relative azimuth and the pitching angle of the bearing platform, the processing unit processes the image and the data to acquire the attribute and the position coordinate of the target, and the control unit receives an external control instruction to enable the photoelectric equipment to select different working modes, observation angles and view field sizes; the imaging processing method utilizes the target video image and the position information obtained by the photoelectric equipment, and identifies whether the target is in accordance with dangerous target characteristics or not and has dangerous behaviors by reflecting the characteristics of the target in the image and the target actions reflected by a plurality of images, collects evidence and implements tracking.
2. The imaging processing method for identifying a dangerous object according to claim 1, wherein the imaging processing method comprises the steps of:
step 1, defining characteristics and behavior combinations of personnel targets belonging to dangerous targets based on cognition of dangerous targets and dangerous behaviors, forming corresponding relations, and storing the corresponding relations in a processing unit of photoelectric equipment;
step 2, when in operation, the photoelectric equipment reaches an event occurrence area along with the bearing platform or is fixed on a bearing device of the concerned area; an operator controls the photoelectric equipment to continuously observe a region of interest to obtain continuous images, and a camera is selected to observe a scene under the condition of good daytime or illumination, and a thermal infrared imager is selected to observe the scene under the condition of night or poor illumination; the processing unit detects personnel targets in the image, and white frame selection marking and tracking target display are adopted; the processing unit calculates the position coordinates of each target according to the data of the laser range finder, the stable platform positioning device and the angle measuring mechanism;
step 3, the processing unit compares the target image, the continuous multiple images and the stored characteristic combination, and judges whether the target belongs to dangerous attributes and dangerous behaviors or not: when the resolution of the obtained target image is higher than a set threshold value, the processing unit compares whether the face of the target person is worn with a mask or not, and whether dangerous objects are held on the hand or not; detecting a person target with a mask and dangerous objects, and judging the person target as a potential danger by the processing unit; the processing unit carries out yellow frame selection labeling and tracking display on the target, and the photoelectric equipment continuously tracks and displays potential attributes of the target;
step 4, when the resolution ratio of the obtained target image is insufficient to identify the characteristics of the face and the handhold object of the personnel, an operator sends a zooming instruction to a control unit, the control unit enlarges the focal length of an objective lens of a camera or a thermal infrared imager to enlarge the size of the target, the resolution ratio is improved, a processing unit identifies a mask or a dangerous object, yellow frame selection marking and tracking display are correspondingly carried out, and photoelectric equipment continuously tracks and displays potential attributes of the mask or dangerous object;
step 5, whether the image resolution can identify a target face mask or a handheld dangerous object, namely whether the detected personnel target is a potential dangerous target or not, the processing unit performs front-back contrast analysis on the multi-frame image with the continuous observation area, and detects and judges the behavior of the personnel target in the image;
step 6, the processing unit judges the dangerous attribute and the behavior of the personnel target according to the personnel target behavior obtained by personnel target feature detection comparison and front-back image comparison, judges the personnel with one of the behaviors in the step 5 as the dangerous target, and adopts red frame selection identification and tracking display; outputting a judging result;
and 7, an operator adjusts the focal length of the camera or the thermal infrared imager, enlarges the red frame to select a target, intercepts a photo of a dangerous target, and realizes photo evidence collection.
3. The method of claim 2, wherein in step 5, the behaviors of the person's target in the image are classified into the following categories:
(1) The personnel target moves to the event area and is interpreted as aggregation;
(2) The object is separated from the limb of the personnel target, and the object moves rapidly to the positions of other personnel, vehicles and buildings, and the object throwing behavior is judged;
(3) And the personnel target handhelds move to quickly move to the positions of other personnel, vehicles and buildings, and the personnel target handhelds are judged to be the object holding and smashing actions.
4. The imaging processing method for identifying dangerous objects according to claim 3, wherein the imaging processing method is used for handling dangerous events of railway stations, and the airborne photoelectric equipment adopts a two-axis stable turntable, is carried by an unmanned helicopter and is connected with a ground station through a data chain of the unmanned plane; the control personnel observe the reconnaissance image and information of the photoelectric equipment at the ground station, and control the photoelectric equipment through the keyboard and the handle; the digital camera of the photoelectric equipment has the sensor resolution of 1920 multiplied by 1080, the frame frequency of 25Hz, and is provided with a continuous zoom objective lens, and the field of view is in the range of 1.2 degrees multiplied by 0.68-35.6 degrees multiplied by 20 degrees; the infrared thermal imager of the photoelectric equipment has the sensor resolution of 1280 multiplied by 1024 and the frame frequency of 25Hz, works in a medium wave infrared band, is provided with a continuous zoom objective lens, and has the field range of 1.8 degrees multiplied by 1.44 degrees multiplied by 20 degrees; a laser range finder of photoelectric equipment, the working wave band is 1.54 mu m, and the range finding precision is +/-1 m; the photoelectric equipment is provided with a Beidou receiver and a fiber-optic gyroscope strapdown inertial navigation device, the ground station is provided with Beidou foundation enhancement equipment, the airborne Beidou positioning precision is not lower than 1m, and the north-seeking precision of the inertial navigation device is 0.8mrad; the photoelectric equipment is observed in an air-to-ground mode, the range of not less than 1058mx1000m can be detected at the height of 3km from the ground, the ground is utilized to enhance the Beidou receiver, and the positioning accuracy of the system to the target at the position of 1km can be within 5 m.
5. The imaging processing method for identifying dangerous objects according to claim 4, wherein the imaging processing method is used for defining characteristics and behavior combinations of the dangerous objects of personnel in step 1 when handling dangerous events of railway stations, including but not limited to:
(1) Wearing a mask to shield facial features;
(2) Aggregating to an event area;
(3) The object is hit by the object holding device.
6. The imaging processing method for identifying dangerous objects according to claim 5, wherein the imaging processing method is used for performing ground reconnaissance when an unmanned helicopter arrives at a railway station square, hovers in the air of 260 m-3 km and is used for performing task in step 2; an operator rides on the vehicle-mounted ground station to reach a safety guard area outside the railway station, controls photoelectric equipment on the control seat, observes the displayed image and continuously scouts the region of interest; under the condition of daytime or good illumination, a camera is selected to observe the scene, and under the condition of night or poor illumination, a thermal infrared imager is selected to observe the scene; the processing unit detects personnel targets in the image, and white frame selection marking and tracking target display are adopted; and the processing unit calculates the position coordinates of each target according to the data of the laser range finder, the stabilized platform positioning device and the angle measuring mechanism.
7. The method of claim 6, wherein when the method is used for handling dangerous events at a railway station, in step 3, the processing unit compares the target image, the continuous multiple images and the stored feature combinations, and automatically interprets whether the target belongs to dangerous target attributes and dangerous behaviors: when the obtained target face size is not smaller than 10 pixels, the processing unit compares whether the face of the target face of the person is worn with a mask, and the processing unit judges that the target face is a potential dangerous target; comparing whether personnel hold dangerous objects, detecting personnel targets with dangerous objects, and judging the personnel targets as potential dangerous targets by a processing unit; and for the potential targets automatically judged by the processing unit, the processing unit carries out yellow frame selection labeling and tracking display on the targets, and the photoelectric equipment continuously tracks and displays potential attributes of the targets.
8. The imaging processing method for recognizing dangerous objects according to claim 7, wherein when the imaging processing method is used for handling dangerous events of a railway station, in step 4, when the obtained resolution of the object image is insufficient for recognizing the face of the person object and the characteristics of the hand-held object, an operator sends a zooming instruction to a control unit, the control unit enlarges the focal length of an objective lens of a camera or a thermal infrared imager to enlarge the size and the resolution of the object to be recognized by not less than 10 pixels, the processing unit recognizes a mask or a dangerous object, and accordingly carries out yellow frame selection labeling and tracking display, and the photoelectric device continuously tracks and displays the potential attributes thereof.
9. The method for imaging and processing dangerous objects according to claim 8, wherein in step 5, the processing unit performs a front-back contrast analysis on the multi-frame image with the continuous observation area, and detects and determines the behavior of the person object in the image, regardless of whether the detected person object is a potential dangerous object, when the method is used for handling dangerous events at a railway station:
(1) The personnel target moves to the railway station square instead of escaping from the dangerous area, and is interpreted as gathering;
(2) The object is separated from the limb of the personnel target, and the object moves rapidly to other personnel, stores beside the railway station and restaurant buildings, and the object throwing behavior is judged;
(3) And the personnel target handhelds move to quickly move to the positions of other personnel, vehicles and buildings, and the personnel target handhelds are judged to be the object holding and smashing actions.
10. The imaging processing method for identifying dangerous objects according to claim 9, wherein when the imaging processing method is used for handling dangerous events of a railway station, in step 6, the processing unit judges dangerous attributes and behaviors of the personnel objects according to personnel object behaviors obtained by personnel object feature detection comparison and front-rear image comparison, judges the personnel having one of the behaviors in step 5 as a dangerous object, and adopts red frame selection identification and tracking display; outputting a judging result; in the step 7, an operator adjusts the focal length of a camera or a thermal infrared imager, enlarges a red frame to select a target, intercepts a photo with the characteristics of the face and the whole body of a dangerous target, and realizes photographing and evidence obtaining; reporting the target coordinates to the relevant personnel.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310789151.9A CN117671576A (en) | 2023-06-30 | 2023-06-30 | Imaging processing method for identifying dangerous target |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310789151.9A CN117671576A (en) | 2023-06-30 | 2023-06-30 | Imaging processing method for identifying dangerous target |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117671576A true CN117671576A (en) | 2024-03-08 |
Family
ID=90065092
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310789151.9A Pending CN117671576A (en) | 2023-06-30 | 2023-06-30 | Imaging processing method for identifying dangerous target |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117671576A (en) |
-
2023
- 2023-06-30 CN CN202310789151.9A patent/CN117671576A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11905000B2 (en) | Flying device, electronic device, and program | |
JP6275285B2 (en) | Logistics system, luggage transport method, and program | |
US7961906B2 (en) | Human detection with imaging sensors | |
JP6128468B2 (en) | Person tracking system and person tracking method | |
US20090268030A1 (en) | Integrated video surveillance and cell phone tracking system | |
CN106878666A (en) | The methods, devices and systems of destination object are searched based on CCTV camera | |
CN110142785A (en) | A kind of crusing robot visual servo method based on target detection | |
CN106341661B (en) | Patrol robot | |
CN205232319U (en) | Camera and monitored control system | |
CN102436738A (en) | Traffic monitoring device based on unmanned aerial vehicle (UAV) | |
CN112068111A (en) | Unmanned aerial vehicle target detection method based on multi-sensor information fusion | |
CN104902233B (en) | Comprehensive safety monitor system | |
JP6827399B2 (en) | Unmanned aerial vehicle control system, logistics system, unmanned aerial vehicle control method, luggage transportation method, and program | |
US8483567B2 (en) | Infrared communication system and method | |
US20210025675A1 (en) | Vehicle-mounted device with network-connected scopes for allowing a target to be simultaneously tracked by multiple other devices | |
CN105637322A (en) | A method of determining the location of a point of interest and the system thereof | |
CN104255022B (en) | Server, client terminal, system and the readable medium of virtual zoom capabilities are added for camera | |
CN107451527A (en) | A kind of large-scale public place hunting system based on unmanned plane | |
CN115346256A (en) | Robot searching method and system | |
US11875080B2 (en) | Object sharing method and apparatus | |
CN105070204A (en) | Miniature AMOLED optical display | |
CN113505704B (en) | Personnel safety detection method, system, equipment and storage medium for image recognition | |
CN112528699B (en) | Method and system for obtaining identification information of devices or users thereof in a scene | |
KR101579275B1 (en) | Security system using real-time monitoring with location-trace for dangerous-object | |
KR102480424B1 (en) | Personal mobility having local monitogring function |
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 |