CN111612815A - Infrared thermal imaging behavior intention analysis method and system - Google Patents
Infrared thermal imaging behavior intention analysis method and system Download PDFInfo
- Publication number
- CN111612815A CN111612815A CN202010297699.8A CN202010297699A CN111612815A CN 111612815 A CN111612815 A CN 111612815A CN 202010297699 A CN202010297699 A CN 202010297699A CN 111612815 A CN111612815 A CN 111612815A
- Authority
- CN
- China
- Prior art keywords
- signal
- thermal imaging
- infrared thermal
- tracking
- comparison
- 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
- 238000001931 thermography Methods 0.000 title claims abstract description 59
- 238000004458 analytical method Methods 0.000 title claims abstract description 34
- 238000001514 detection method Methods 0.000 claims abstract description 5
- 230000037406 food intake Effects 0.000 claims abstract description 3
- 230000006399 behavior Effects 0.000 claims description 60
- 238000012545 processing Methods 0.000 claims description 22
- 230000003542 behavioural effect Effects 0.000 claims description 20
- 230000033001 locomotion Effects 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 14
- 230000009471 action Effects 0.000 claims description 13
- 230000002776 aggregation Effects 0.000 claims description 13
- 238000004220 aggregation Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 230000010354 integration Effects 0.000 claims description 4
- 230000004931 aggregating effect Effects 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 230000005855 radiation Effects 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 description 24
- 238000005516 engineering process Methods 0.000 description 8
- 238000000605 extraction Methods 0.000 description 8
- 230000000007 visual effect Effects 0.000 description 3
- 238000010191 image analysis Methods 0.000 description 2
- 230000000391 smoking effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 241000282376 Panthera tigris Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 208000023504 respiratory system disease Diseases 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Abstract
The invention discloses an infrared thermal imaging behavior intention analysis method, which comprises the following steps: starting an infrared thermal imaging camera to acquire video data; distinguishing the background from the monitored person via profile temperature; tracking of video material ingestion is performed for the monitored person. The invention also discloses an infrared thermal imaging behavior intention analysis system, a video source is obtained through an infrared thermal imaging camera, an image is constructed by the thermal radiation difference generated by the temperature of an object, and the system can be normally used in the occasions of night, haze and low visibility. The moving target detection and tracking based on the gray level image are analyzed, so that the system identification rate and the tracking success rate are ensured; the infrared thermal imaging camera only presents contour images for sensitive information such as the face and body characteristics of a monitored person, cannot directly distinguish the appearance, the skin color and the like, does not infringe the privacy of citizens, and can be used in many sensitive occasions.
Description
Technical Field
The invention relates to a monitoring technology, in particular to an infrared thermal imaging behavior intention analysis method and system with higher image recognition capability.
Background
Nowadays, monitoring technologies are developed, especially image analysis technologies and network communication technologies, and in most towns, numerous monitoring systems are already provided for monitoring and protecting the security of private or public areas, finding out anomalies in time, and retaining monitoring data for later adjustment and evidence collection. Although image recognition technology has reached a high level, there are many factors that influence the accuracy and suitability of real-time video monitoring. At present, for the existing monitoring equipment, a visible light camera is mainly used for collecting a color image as an analysis source, and the following disadvantages exist:
1. a color image is acquired by a visible light camera and is used as an analysis source, and the monitored place is irradiated by sunlight and lamplight very much. Even if the night and the dark light field can be irradiated by the infrared lamp, the irradiation distance is very limited. Systems often identify errors and track failures.
2. Under the condition that the surface color of the monitored person or object is extremely close to the background color, the system is easy to extract the moving points wrongly, so that the probability of tracking failure is high.
3. The visible light camera can directly acquire sensitive image information such as face and body characteristics of a monitored person, and invasion is caused to citizen privacy. Meanwhile, the application range of the security camera is limited.
In addition, the existing intelligent tracking ball machine system for monitoring the visible light security is combined with an intelligent identification technology, mainly by carrying out differential calculation on images, the moving direction of a target in a visual range is automatically identified, a cloud platform is automatically controlled to track the moving target, and all actions are clearly transmitted to a monitoring center through a configured high-definition automatic zoom lens in the period from the time when the moving target enters the intelligent high-speed ball to the time when the moving target leaves the intelligent high-speed ball. And once a certain area gives an alarm, other related intelligent high-speed balls automatically rotate to an alarm point to start tracking, so that the monitoring picture can record the moving track of the actor. For the monitoring high-speed dome camera, the mobile tracking can be controlled without a back-end software platform, and the tracking task can be automatically completed only at the front end. When the target agent enters the visual range of the dome camera, the dome camera can automatically recognize the moving direction of the target, control the rotation of the holder, track the target, enable the target to be positioned in the center of a monitoring picture, automatically track the target continuously until the target leaves the visual range of the dome camera, and return to the original observation point to prepare for next tracking. The realization of the automatic tracking technology is completely based on a special intelligent analysis module, and the intelligent monitoring camera can quickly and accurately carry out automatic tracking, arbitrary positioning and all-weather full-range automatic cruising non-blind-spot monitoring by matching with a high-precision holder and a high-definition pixel variable-focus camera.
The moving target detection and tracking comprises the steps of background extraction, moving point group position extraction and moving object tracking. In the background extraction step, an improved mean-based background extraction algorithm and a region-of-interest extraction method for reducing image pixels are provided. The shading step in the moving blob extraction is based on a shading algorithm in RGB space. An improved line coding algorithm is proposed in the motion blob position extraction step.
However, the dependence on visible light causes a series of problems, and it is difficult to ensure the monitoring accuracy and success rate under abnormal environmental factors. Meanwhile, the behavior intention of the monitored personnel cannot be accurately analyzed through video monitoring at present, manual later-stage searching or real-time checking are needed for analysis, timely early warning cannot be provided, and the efficiency is low.
Disclosure of Invention
The invention solves the technical problem of providing an infrared thermal imaging behavior intention analysis method and system, which do not depend on the imaging basis of visible light, improve the accuracy and success rate of image monitoring analysis and improve the monitoring adaptability.
The technical solution of the invention is as follows:
an infrared thermal imaging behavior intention analysis method comprises the following steps:
starting an infrared thermal imaging camera to acquire video data;
distinguishing the background from the monitored person via profile temperature;
tracking of video material ingestion is performed for the monitored person.
The invention also provides an infrared thermal imaging behavior intention analysis system, which comprises:
an infrared thermal imaging camera video data acquisition device,
a signal processing device in signal connection with the video data acquisition device, comprising
The input signal processing module is used for extracting and classifying the received infrared thermal imaging video signals;
the signal comparison module is used for comparing the proposed infrared thermal imaging video signals with the classified infrared thermal imaging video signals;
the signal aggregation module is used for aggregating different video signal images in the comparison result;
the signal feedback module is used for generating the aggregated video signal image into a tracking target;
the tracking module sends a rotation control signal to the video data acquisition device according to the movement of the tracking target so that the tracking target is positioned in the middle of the video image;
the behavior intention analysis module is used for judging the identification behavior of the locked actor according to the recorded human posture behavior integrated data;
the signal processing device performs signal processing by adopting the steps of the infrared thermal imaging behavior intention analysis method.
From the above description, it is clear that the present invention has the following advantages:
1. the video source is obtained through the infrared thermal imaging camera, infrared thermal imaging does not depend on sunlight and lamplight to irradiate a monitored place, and an image is formed completely by means of the thermal radiation difference generated by the temperature of an object. Can be normally used in the occasions of dark night, haze and low visibility.
2. Under the condition that the surface color of the monitored person or object is extremely close to the background color, the system distinguishes the background from the monitored person through the contour temperature, and a motion tracking algorithm based on prediction is provided for the step of tracking the monitored person. Through analysis and comparison, the moving target detection and tracking based on the gray image sequence and the moving target detection and tracking based on the edge image sequence are realized, and the system identification rate and the tracking success rate are ensured.
3. The infrared thermal imaging camera only presents contour images for sensitive information such as the face and body characteristics of a monitored person. The appearance, skin color, etc. cannot be directly identified. The privacy of citizens is not infringed. Can be used in many sensitive occasions.
4. When the locked agent is too close to the monitoring camera, the system will take a snapshot of the agent's face. And identifying according to the heat effect of the human face, and matching infrared thermal imaging with a common visible light picture through a public security system database. The test conclusion shows that the recognition success rate reaches 80%, wherein 55% of recognition can be completed by only one picture. The mask or the veil can also identify the human face no matter in weak light or even in dark environment.
Drawings
Fig. 1 is a schematic flow chart of an infrared thermal imaging behavior intention analysis method according to the present invention.
Detailed Description
In order to more clearly understand the technical features, objects, and effects of the present invention, embodiments of the present invention will now be described with reference to the accompanying drawings.
In a preferred embodiment of the present invention, referring to fig. 1, the method for analyzing behavior intention by infrared thermal imaging includes the steps of:
s101, starting an infrared thermal imaging camera to acquire video data;
s102, distinguishing a background from a monitored person through contour temperature;
s103, the video material shooting tracking is carried out on the monitored person.
In the above-mentioned method for analyzing behavioral intention of infrared thermal imaging according to the present invention, in a preferred embodiment, the tracking of video data capturing performed by the monitored person is a prediction-based motion tracking, which includes comparing:
and the method comprises the following steps of detecting and tracking a moving object based on a gray image sequence and detecting and tracking a moving object based on an edge image sequence.
As described above, in the preferred embodiment of the present invention, the infrared thermal imaging behavior intention analyzing method includes the following steps:
the camera comprises an input signal, a first processing unit and a second processing unit, wherein the input signal is used for receiving a first video signal with a set frequency of a camera and extracting a static object image in a video picture generated by the first video signal in a first time interval as a background image;
processing signals, wherein video pictures of the first video signal at the same time interval after the first time interval are moving pictures, and extracting a specified number of comparison moving pictures in the same time interval in sequence;
signal comparison, namely performing active comparison on the comparison moving picture and a background image;
signal aggregation, wherein different moving pictures exist in comparison results, different parts of the moving pictures larger than the designated pixel area are aggregated, different parts smaller than the designated pixel area are ignored, and the aggregation larger than the designated pixel area generates edges by the square of the designated pixel area; the area of the appointed pixel is set according to the shooting distance;
and signal feedback, namely superposing a label on the aggregate moving in the background image and being larger than the specified pixel distance according to the sequential advance of the time interval to generate a tracking target visible on the screen.
In the above-mentioned method for analyzing behavioral intention of infrared thermal imaging according to the present invention, in a preferred embodiment of the method, the motion tracking based on prediction further comprises, after the signal feedback step, the steps of:
and signal tracking, namely taking the tracking target as a screen center, and sending a rotation instruction to the camera holder when the target exceeds the displacement of the specified pixel distance so as to keep the tracking target at the screen center.
In the above-mentioned method for analyzing behavioral intent of infrared thermal imaging according to the present invention, in a preferred embodiment, the video signal is 1280 × 1024 pixels, the specified frequency is 50 hz, the time interval is 1 second, and 50 pictures are included in 1 second;
in the signal processing step, the specified number of moving pictures are five comparison moving pictures of the 1 st, 10 th, 20 th, 30 th and 40 th frames at each second time interval;
in the signal comparison step, the activity comparison comprises the step of checking all pixel points of the comparison moving picture and the background image once, wherein each pixel point is defined by a gray value of 0 to 100, all pixel gray values of the background image are listed into a comparison test set, pixel point gray values on the comparison moving picture are listed into a comparison training set, and the comparison moving picture and the background image are compared point by point;
in the signal assembling step, the area of the designated pixel is 32 multiplied by 32 pixels;
in the signal feedback step, the specified pixel distance is 32 pixels, the superimposed label is a rectangular frame with a specified color, and the rectangular frame is continuously displayed at 5 frames per second.
In the above-mentioned infrared thermal imaging behavioral intention analysis method according to the present invention, in a preferred embodiment of the method, the signal tracking step includes ignoring the tracking target with an area smaller than 128 × 128 pixels.
The method for analyzing behavioral intent of infrared thermal imaging according to the present invention as described above, in a preferred embodiment, comprises the steps of:
after the agent is locked, the action intention of the agent is analyzed through the continuous motion track of the body limb.
In the above-mentioned method for analyzing behavioral intention of infrared thermography according to the present invention, in a preferred embodiment, the step of analyzing behavioral intention of an agent comprises:
behavior definition, namely acquiring signal aggregation of different human body posture actions by shooting multiple angles of the human body posture actions; compiling serial number codes for signal sets of human posture actions, obtaining a posture behavior integration data packet and storing the posture behavior integration data packet in a database;
comparing behaviors, namely comparing each pixel point of the contour set generated by the locked actor with the human body posture set in the database in a multi-angle and multi-time manner to obtain the behavior posture with the highest coincidence degree percentage as an identification behavior;
and behavior record, namely behavior record of the locked actor according to the identification behavior record.
In the above-mentioned method for analyzing behavioral intention of infrared thermography according to the present invention, in a preferred embodiment, the step of analyzing behavioral intention of an agent comprises:
and setting special identification behavior data, and giving an alarm if the identification behavior of the locked actor belongs to the identification behaviors in the special identification behavior data.
The invention also provides an infrared thermal imaging behavior intention analysis system, which comprises the following components in a preferred embodiment:
an infrared thermal imaging camera video data acquisition device,
a signal processing device in signal connection with the video data acquisition device, comprising
The input signal processing module is used for extracting and classifying the received infrared thermal imaging video signals;
the signal comparison module is used for comparing the proposed infrared thermal imaging video signals with the classified infrared thermal imaging video signals;
the signal aggregation module is used for aggregating different video signal images in the comparison result;
the signal feedback module is used for generating the aggregated video signal image into a tracking target;
the tracking module sends a rotation control signal to the video data acquisition device according to the movement of the tracking target so that the tracking target is positioned in the middle of the video image;
the behavior intention analysis module is used for judging the identification behavior of the locked actor according to the recorded human posture behavior integrated data;
the signal processing device performs signal processing by adopting the steps of the infrared thermal imaging behavior intention analysis method.
The infrared thermal imaging behavior intention analysis method is established on a software system platform of a visible light security monitoring intelligent tracking ball machine, and utilizes a gray image formed by infrared thermal imaging to replace a color common image formed by visible light as an analysis source. Because the image is a gray image, the invention can avoid the image noise caused by the fact that the color of the gum is close to the dark light, and simultaneously avoid the original complicated and unreliable color background extraction and shadow processing operation of a software system, and enhance the gray image analysis, thereby obtaining the quick and accurate target discovery operation.
For a better example, the motion tracking algorithm based on prediction may be implemented as follows:
input signal-receiving camera video signal, which is 1280x1024 pixels 50 hz video. And extracting a static object in 50 frames of pictures within 1 second as a background picture.
Signal processing-five images of 1 st, 10 th, 20 th, 30 th, 40 th frames in the video signal one second after the background data are extracted.
Comparison algorithm-the video signal makes active comparison with the background picture according to the frequency of five pictures per second, and all pixel points of 1280x1024 are checked once. The color of each pixel is simplified to different gray levels, and the gray level is defined to be 0 to 100 (for example, 256 numbers are used to increase the operation burden by using the visible color value). All pixel gray values of the background picture can be listed in the test set, the pixel point gray values of the moving picture are listed in the training set, the moving picture and the background picture are compared point by point to obtain a result of 0, and otherwise, the result of the comparison is 1.
Clustering-all comparisons concluded to be 1 are clustered, clustering less than 32x32 will be ignored. Clusters larger than 32x32 will be marked and listed as targets for inspection, and clusters larger than 32x32 in large scale will produce edges in squares of 32x 32.
Signal feedback-moving clusters larger than 32x32 in the background picture are superimposed with colored square borders, and the 5 frames per second picture is displayed continuously, generating a visible tracking target on the screen.
And a reasonable aggregation range (such as 128x128, the size of which is related to the lens parameters of the monitoring equipment according to the monitoring range and avoids judging the small animals and the floating objects as tracking targets) is set in the tracking-system to serve as the tracking targets. And tracking a target aggregation center as a screen center, and when the target exceeds 32 pixel point displacements. The system sends 485 mobile signals to the holder to enable the holder to rotate, and the target is located at the center of the screen again to achieve tracking.
The infrared thermal imaging camera only presents contour images for sensitive information such as the face and body characteristics of a monitored person. The appearance, skin color, etc. cannot be directly identified. The privacy of citizens is not infringed. Can be used in many sensitive occasions.
After the action is locked, the action intention and the like of the locked action can be very accurately identified and recorded through comparison of the identification action data, and real-time early warning is convenient to send out. Specifically, the behavior recognition may be:
and (3) behavior definition: all people need to shoot as many human body gestures as possible through multiple angles in the early stage, and aggregation is obtained through system comparison. Manually integrating the aggregation serial number codes and integrating various behaviors into a data packet and storing the data packet in a database.
And (4) behavior comparison: and each pixel point of the contour set generated by the locked agent is compared with the data of the database by multiple times in multiple angles, and the code with the highest percentage of hundreds is judged as the identification behavior.
The special behavior algorithm is as follows: for the behavior that the agent holds a high heat source, the thermal imaging can directly measure the temperature value. The system directly alarms on a variety of high-heat-source-sustaining behaviors (smoking, open-fire behavior).
When the locked agent is too close to the monitoring camera, the system will take a snapshot of the agent's face. And identifying according to the heat effect of the human face, and matching infrared thermal imaging with a common visible light picture through a public security system database. The test conclusion shows that the recognition success rate reaches 80%, wherein 55% of recognition can be completed by only one picture. The mask or the veil can also identify the human face no matter in weak light or even dark environment
The method is used for identifying according to the thermal effect of the human trunk/face, and processing images to realize tracking by adopting the infrared thermal imaging to generate gray level images and extracting the background, the moving point cluster and the moving point cluster position. Then, the human face is recognized in a weak light or even dark environment, and the gray level picture generated by infrared thermal imaging is matched with the common visible light picture of the database.
Furthermore, after the actor is locked, the intention of the actor is analyzed through the coherent motion track of the body limb, the obtained data can be compared through a database to obtain the conventional behaviors of sitting, standing, walking, running, smoking, eating, using electronic equipment and the like, and the analysis on the attack intentions of holding a knife, holding a gun, throwing, drawing a bow and an arrow and the like is more accurate. When the locked agent is too close to the monitoring camera, the system will take a snapshot of the agent's face.
And identifying according to the heat effect of the human face, and matching infrared thermal imaging with a common visible light picture through a public security system database. The test conclusion shows that the recognition success rate reaches 80%, wherein 55% of recognition can be completed by only one picture. The mask or the veil can also identify the human face no matter in weak light or even in dark environment. The invention can fundamentally eliminate the influence of the mask on shielding the face, and can also identify the identity of an actor during the transverse movement of respiratory diseases. The technology is applied to a monitoring ball machine/single-soldier shooting and recording evidence obtaining instrument, and can play a role like tiger in the field of public security frontier defense and security management.
The above description is only an exemplary embodiment of the present invention, and is not intended to limit the scope of the present invention. Any equivalent changes and modifications that can be made by one skilled in the art without departing from the spirit and principles of the invention should fall within the protection scope of the invention.
Claims (10)
1. An infrared thermal imaging behavior intention analysis method is characterized by comprising the following steps:
starting an infrared thermal imaging camera to acquire video data;
distinguishing the background from the monitored person via profile temperature;
tracking of video material ingestion is performed for the monitored person.
2. The infrared thermal imaging behavioral intent analysis method according to claim 1, characterized in that said tracking of video material intake performed by the monitored person is a prediction-based motion tracking comprising by analytical comparison:
moving object detection and tracking based on a sequence of grayscale images, an
And detecting and tracking the moving target based on the edge image sequence.
3. The infrared thermal imaging behavioral intent analysis method according to claim 2, characterized in that said prediction-based motion tracking comprises the steps of:
the camera comprises an input signal, a first processing unit and a second processing unit, wherein the input signal is used for receiving a first video signal with a set frequency of a camera and extracting a static object image in a video picture generated by the first video signal in a first time interval as a background image;
processing signals, wherein video pictures of the first video signal at the same time interval after the first time interval are moving pictures, and extracting a specified number of comparison moving pictures in the same time interval in sequence;
signal comparison, namely performing active comparison on the comparison moving picture and a background image;
signal aggregation, wherein different moving pictures exist in comparison results, different parts of the moving pictures larger than the designated pixel area are aggregated, different parts smaller than the designated pixel area are ignored, and the aggregation larger than the designated pixel area generates edges by the square of the designated pixel area; the area of the appointed pixel is set according to the shooting distance;
and signal feedback, namely superposing a label on the aggregate moving in the background image and being larger than the specified pixel distance according to the sequential advance of the time interval to generate a tracking target visible on the screen.
4. The infrared thermographic behavioral intent analysis method according to claim 3, wherein said prediction-based motion tracking further comprises, after the signal feedback step, the steps of:
and signal tracking, namely taking the tracking target as a screen center, and sending a rotation instruction to the camera holder when the target exceeds the displacement of the specified pixel distance so as to keep the tracking target at the screen center.
5. The infrared thermal imaging behavioral intention analysis method according to claim 3, wherein the video signal is 1280x1024 pixels, the specified frequency is 50 hz, the time interval is 1 second, and 50 pictures are contained within 1 second;
in the signal processing step, the specified number of moving pictures are five comparison moving pictures of the 1 st, 10 th, 20 th, 30 th and 40 th frames at each second time interval;
in the signal comparison step, the activity comparison comprises the step of checking all pixel points of the comparison moving picture and the background image once, wherein each pixel point is defined by a gray value of 0 to 100, all pixel gray values of the background image are listed into a comparison test set, pixel point gray values on the comparison moving picture are listed into a comparison training set, and the comparison moving picture and the background image are compared point by point;
in the signal assembling step, the area of the designated pixel is 32 multiplied by 32 pixels;
in the signal feedback step, the specified pixel distance is 32 pixels, the superimposed label is a rectangular frame with a specified color, and the rectangular frame is continuously displayed at 5 frames per second.
6. The infrared thermal imaging behavioral intention analysis method according to claim 4, characterized in that in the signal tracking step, it includes ignoring tracking targets smaller than 128x128 pixel area.
7. The infrared thermal imaging behavioral intention analysis method according to claim 5 or 6, characterized by comprising the steps of:
after the agent is locked, the action intention of the agent is analyzed through the continuous motion track of the body limb.
8. The infrared thermal imaging behavioral intention analysis method according to claim 7, wherein said analyzing behavioral intention of an agent includes the steps of:
behavior definition, namely acquiring signal aggregation of different human body posture actions by shooting multiple angles of the human body posture actions; compiling serial number codes for signal sets of human posture actions, obtaining a posture behavior integration data packet and storing the posture behavior integration data packet in a database;
comparing behaviors, namely comparing each pixel point of the contour set generated by the locked actor with the human body posture set in the database in a multi-angle and multi-time manner to obtain the behavior posture with the highest coincidence degree percentage as an identification behavior;
and behavior record, namely behavior record of the locked actor according to the identification behavior record.
9. The infrared thermographic behavioral intention analyzing method according to claim 8, wherein said analyzing the behavioral intention of the agent comprises the steps of:
and setting special identification behavior data, and giving an alarm if the identification behavior of the locked actor belongs to the identification behaviors in the special identification behavior data.
10. An infrared thermographic behavioral intent analysis system, comprising:
an infrared thermal imaging camera video data acquisition device,
a signal processing device in signal connection with the video data acquisition device, comprising
The input signal processing module is used for extracting and classifying the received infrared thermal imaging video signals;
the signal comparison module is used for comparing the proposed infrared thermal imaging video signals with the classified infrared thermal imaging video signals;
the signal aggregation module is used for aggregating different video signal images in the comparison result;
the signal feedback module is used for generating the aggregated video signal image into a tracking target;
the tracking module sends a rotation control signal to the video data acquisition device according to the movement of the tracking target so that the tracking target is positioned in the middle of the video image;
the behavior intention analysis module is used for judging the identification behavior of the locked actor according to the recorded human posture behavior integrated data;
the signal processing device performs signal processing by using the infrared thermal imaging behavior intention analyzing method steps as claimed in any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010297699.8A CN111612815A (en) | 2020-04-16 | 2020-04-16 | Infrared thermal imaging behavior intention analysis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010297699.8A CN111612815A (en) | 2020-04-16 | 2020-04-16 | Infrared thermal imaging behavior intention analysis method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111612815A true CN111612815A (en) | 2020-09-01 |
Family
ID=72199716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010297699.8A Pending CN111612815A (en) | 2020-04-16 | 2020-04-16 | Infrared thermal imaging behavior intention analysis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111612815A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112399144A (en) * | 2020-11-05 | 2021-02-23 | 上海明略人工智能(集团)有限公司 | Thermal imaging monitoring early warning method and device and thermal imaging monitoring management system |
| CN113936405A (en) * | 2021-10-08 | 2022-01-14 | 国能榆林能源有限责任公司 | Alarm method, alarm system and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050128291A1 (en) * | 2002-04-17 | 2005-06-16 | Yoshishige Murakami | Video surveillance system |
| CN101226597A (en) * | 2007-01-18 | 2008-07-23 | 中国科学院自动化研究所 | Method and system for recognizing nights pedestrian based on thermal infrared gait |
| CN102799856A (en) * | 2012-06-15 | 2012-11-28 | 天津大学 | Human action recognition method based on two-channel infrared information fusion |
| CN103150575A (en) * | 2013-01-31 | 2013-06-12 | 广州中国科学院先进技术研究所 | Real-time three-dimensional unmarked human body gesture recognition method and system |
| CN104778726A (en) * | 2015-04-29 | 2015-07-15 | 深圳市保千里电子有限公司 | Motion trail tracing method and system based on human body characteristics |
| CN109948474A (en) * | 2019-03-04 | 2019-06-28 | 成都理工大学 | AI thermal imaging all-weather intelligent monitoring method |
| CN110874905A (en) * | 2018-08-31 | 2020-03-10 | 杭州海康威视数字技术股份有限公司 | Monitoring method and device |
-
2020
- 2020-04-16 CN CN202010297699.8A patent/CN111612815A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050128291A1 (en) * | 2002-04-17 | 2005-06-16 | Yoshishige Murakami | Video surveillance system |
| CN101226597A (en) * | 2007-01-18 | 2008-07-23 | 中国科学院自动化研究所 | Method and system for recognizing nights pedestrian based on thermal infrared gait |
| CN102799856A (en) * | 2012-06-15 | 2012-11-28 | 天津大学 | Human action recognition method based on two-channel infrared information fusion |
| CN103150575A (en) * | 2013-01-31 | 2013-06-12 | 广州中国科学院先进技术研究所 | Real-time three-dimensional unmarked human body gesture recognition method and system |
| CN104778726A (en) * | 2015-04-29 | 2015-07-15 | 深圳市保千里电子有限公司 | Motion trail tracing method and system based on human body characteristics |
| CN110874905A (en) * | 2018-08-31 | 2020-03-10 | 杭州海康威视数字技术股份有限公司 | Monitoring method and device |
| CN109948474A (en) * | 2019-03-04 | 2019-06-28 | 成都理工大学 | AI thermal imaging all-weather intelligent monitoring method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112399144A (en) * | 2020-11-05 | 2021-02-23 | 上海明略人工智能(集团)有限公司 | Thermal imaging monitoring early warning method and device and thermal imaging monitoring management system |
| CN113936405A (en) * | 2021-10-08 | 2022-01-14 | 国能榆林能源有限责任公司 | Alarm method, alarm system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109819208B (en) | Intensive population security monitoring management method based on artificial intelligence dynamic monitoring | |
| CN107093171B (en) | Image processing method, device and system | |
| WO2020151084A1 (en) | Target object monitoring method, apparatus, and system | |
| Meivel et al. | Mask detection and social distance identification using internet of things and faster R-CNN algorithm | |
| US8254633B1 (en) | Method and system for finding correspondence between face camera views and behavior camera views | |
| US9224278B2 (en) | Automated method and system for detecting the presence of a lit cigarette | |
| Kalra et al. | Dronesurf: Benchmark dataset for drone-based face recognition | |
| TWI416068B (en) | Object tracking method and apparatus for a non-overlapping-sensor network | |
| CN108319926A (en) | A kind of the safety cap wearing detecting system and detection method of building-site | |
| CN111341461A (en) | Infrared epidemic situation intelligent comprehensive detection and identification system | |
| CN110543867A (en) | crowd density estimation system and method under condition of multiple cameras | |
| CN105930822A (en) | Human face snapshot method and system | |
| CN109218667B (en) | Public place safety early warning system and method | |
| US20200394384A1 (en) | Real-time Aerial Suspicious Analysis (ASANA) System and Method for Identification of Suspicious individuals in public areas | |
| CN111723633B (en) | Personnel behavior pattern analysis method and system based on depth data | |
| CN112396658A (en) | Indoor personnel positioning method and positioning system based on video | |
| CN111144252B (en) | Monitoring and early warning method for people stream analysis | |
| CN112733690A (en) | High-altitude parabolic detection method and device and electronic equipment | |
| US20220122360A1 (en) | Identification of suspicious individuals during night in public areas using a video brightening network system | |
| US20190096066A1 (en) | System and Method for Segmenting Out Multiple Body Parts | |
| CN112528817A (en) | Patrol robot visual detection and tracking method based on neural network | |
| CN111612815A (en) | Infrared thermal imaging behavior intention analysis method and system | |
| CN112270253A (en) | High-altitude parabolic detection method and device | |
| CN111539348A (en) | Face living body detection method based on polarization imaging | |
| CN112613568A (en) | Target identification method and device based on visible light and infrared multispectral image sequence |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20200901 |