CN111063148A - Remote night vision target video detection method - Google Patents
Remote night vision target video detection method Download PDFInfo
- Publication number
- CN111063148A CN111063148A CN201911402390.4A CN201911402390A CN111063148A CN 111063148 A CN111063148 A CN 111063148A CN 201911402390 A CN201911402390 A CN 201911402390A CN 111063148 A CN111063148 A CN 111063148A
- Authority
- CN
- China
- Prior art keywords
- early warning
- module
- target
- laser
- alarm
- 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
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19617—Surveillance camera constructional details
- G08B13/1963—Arrangements allowing camera rotation to change view, e.g. pivoting camera, pan-tilt and zoom [PTZ]
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19639—Details of the system layout
- G08B13/19645—Multiple cameras, each having view on one of a plurality of scenes, e.g. multiple cameras for multi-room surveillance or for tracking an object by view hand-over
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Studio Devices (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention belongs to the field of intelligent video monitoring, and particularly relates to a remote night vision target video detection method, wherein < A. a thermal imaging acquisition device > sends acquired video stream data to < E2. an early warning module > in real time; the control module converts (x, y, h) of the early warning signal into corresponding instructions and respectively sends the instructions to a laser emitter, a holder and a visible light/infrared camera; sending the collected video stream to an identification module of < E3. > by a visible light/infrared camera; < E3. discernment module > carries out analysis, discernment, classification to the video stream of gathering, obtains alarm information. The high-precision automatic detection and identification target under the long-distance and large-range night vision environment is realized, and the system has the advantages of being modular in function, all-weather, long-distance, highly integrated, light in weight, easy to deploy and the like.
Description
Technical Field
The invention belongs to the field of intelligent video monitoring, and particularly relates to a remote night vision target video detection method.
Background
At present, the traditional video analysis technology is difficult to overcome multi-factor interference under complex scenes, and a large number of false alarms are easily caused particularly under the conditions of bad weather, low imaging quality under night vision conditions, unclear video pictures and the like. Meanwhile, the traditional video analysis technology generally adopts a back-end server to carry out operation, so that the server has high calculation pressure and heavy transmission load.
Patent "railway foreign matter invasion safety perception and identification system based on distributed optical fiber sensing" (application publication number CN 109541715A) discloses a railway foreign matter invasion safety perception and identification system based on distributed optical fiber sensing, based on two kinds of distributed optical fiber sensing technologies and video linkage technologies: monitoring a vibration invasion signal generated when illegal personnel, vehicles and construction invade a high-speed railway by utilizing a phase optical time domain analysis technology, and giving information such as the position, the event and the like of the invasion signal; monitoring a fence deformation signal of the high-speed railway generated when large-scale rockfall and debris flow invade the high-speed railway by utilizing a Brillouin optical time domain analysis technology, and giving information such as the position of the invasion signal; when any one or two of the two optical fiber sensing systems gives an alarm, the optical fiber sensing systems can be linked with cameras arranged on a high-speed railway tower pole to call monitoring pictures of the intrusion event. The three subsystems work synchronously, study mutually, monitor invasion layer by layer, and realize early warning and identification of foreign matter invasion of the high-speed railway together. The patent mainly adopts the optical fiber vibration mode to detect abnormal vibration, and can not actively analyze abnormal types, and the camera is used for providing monitoring pictures for managers to check when abnormality occurs. Meanwhile, the problem that the monitoring effect is influenced by low imaging quality and unclear video picture under bad weather and night vision conditions is not solved.
The patent "an all-weather monitoring image intelligent analysis and alarm system and method" (application publication No. CN 105321289A) provides an all-weather monitoring image intelligent analysis and alarm system, the system is provided with three working modes, namely an offshore water surface oil spilling video analysis and alarm mode, a crowd density monitoring and alarm mode and a fire detection and alarm mode; the system comprises an infrared intelligent monitoring module, a holder and an upper computer: the infrared intelligent monitoring module comprises an infrared imaging module, an information processing module and a power supply and is used for image acquisition and processing: the cradle head is used for bearing the follow-up platform and the line transmission channel, adjusting the monitoring observation position, automatically adjusting the angle and ensuring that the target is positioned in the monitoring area; the upper computer is set to be communicated with the infrared intelligent monitoring module, so that the all-weather monitoring image intelligent analysis and alarm system is controlled and configured, data obtained by the system is stored and displayed, and a man-machine interaction interface is provided. The patent mainly adopts an infrared camera and a threshold segmentation method to detect oil spilling, crowd density and fire in a monitored area. The effective visible distance of the infrared camera does not exceed 50 meters under the condition of no light source at night, and meanwhile, a detection method of threshold segmentation is adopted, so that serious false alarm and false alarm can be generated when the background is changed violently, so that the patent cannot solve the problem of automatic analysis and alarm of a monitoring system under the conditions of long distance and complex scenes, particularly poor imaging quality under bad weather and night vision conditions, unclear video pictures and the like.
The patent "an all-weather video monitoring method based on deep learning" (application publication No. CN 104320617A) discloses an all-weather video monitoring method based on deep learning, which comprises the following steps: acquiring a video stream in real time, and acquiring a plurality of original sampling chart samples and speed sampling chart samples based on the acquired video stream through line sampling; performing space-time correction on the obtained speed sampling pattern book; on the basis of the original sampling diagram and the speed sampling diagram, performing off-line training to obtain a deep learning model, wherein the deep learning model comprises a classification model and a statistical model; and carrying out pedestrian group state analysis on the real-time video stream by using the obtained deep learning model. The invention has good adaptability to different environments, illumination intensity, weather conditions and camera angles; the high accuracy rate can be ensured for crowd environments such as the crowd with large flow and the like; the method has small calculated amount, can meet the requirement of real-time video processing, and can be widely applied to monitoring and management of public places such as buses, subways and squares where detained people are dense. The patent mainly solves the problem of crowd state analysis, does not provide a solution that the imaging quality is not high and the video picture is not clear under the condition of no illumination at night, and is not suitable for long-distance and large-range application scenes.
Problem of the invention
The patent provides a remote night vision target video detection method, which solves the problem of automatic detection and identification of a large-range remote night vision target of video camera equipment in the field of safety monitoring. The method provides a complete solution and a system device product aiming at the problems and weak links in the aspects of insufficient identification capability of the current night vision automatic detection, insufficient accuracy of a real-time video target identification and analysis technology, insufficient performance of an embedded video analysis technology, large demand of video monitoring equipment, difficulty in large-scale deployment and the like.
The invention discloses a remote night vision target video detection method,
[0000] the intelligent monitoring system comprises a thermal imaging acquisition device, a laser emitter, a tripod head, a visible light/infrared camera, a main control unit and a terminal platform, wherein the main control unit is divided into four small components, namely an E1 control module, an E2 early warning module, an E3. identification module and a E4. power supply;
a, a thermal imaging acquisition device sends acquired video stream data to an E2 early warning module in real time;
the early warning module analyzes the received video stream, identifies early warning signals, and obtains data sets { (x1, y1, h1), (x2, y2, h2) … … (xn, yn, hn) } of early warning target coordinate points and height information;
the early warning module sends an early warning signal target coordinate point and a height information data set to the control module E1;
the method comprises the following steps that (E1) a pre-warning signal is selected from a pre-warning signal data set according to a certain strategy, and (x, y, h) of the pre-warning signal is converted into corresponding instructions (HLVX, HLVY, W, F), wherein x is converted into HLVX, y is converted into HLVY, h is converted into W and F, and the W and F are respectively sent to a laser transmitter >, < C. cloud platform >, < D. visible light/infrared camera >;
transmitting a laser instruction to the < B. laser emitter > according to the [0200], < E1. control module >;
according to [0210], the laser transmitter automatically judges whether laser needs to be transmitted or not according to the current light signal, and ignores the early warning signal if the current light signal is strong; otherwise, calculating and determining the laser emission position and the aperture size according to the early warning information, and emitting laser at a fixed point;
according to the (0210) and (B) laser transmitters, after receiving the instruction, automatically calculating the laser emission position and the size of a light spot according to specific instruction parameters and early warning target position and height information, realizing intelligent light supplement at different distances and improving the night vision light supplement level;
sending horizontal rotation and pitching rotation instructions to the < C. pan-tilt > according to the [0200], < E1. control module >;
after receiving the instruction according to [0220], < C. cloud deck >, automatically calculating the horizontal and pitching rotation angles according to the instruction parameters and the early warning target position information, and reaching the specified position;
sending a zooming instruction to the < D. visible light/infrared camera > according to the [0200], < E1. control module >;
after receiving the instruction according to [0231] and < D. visible light/infrared camera >, automatically adjusting the focal length according to the instruction information parameters and the early warning target position and height information to obtain a proper video acquisition angle;
sending the collected video stream to an identification module of < E3. > by a visible light/infrared camera;
the identification module E3. analyzes, identifies and classifies the collected video stream to obtain alarm information;
organizing packaging alarm information according to [0310], < E3. identification module ], and sending the alarm information to < F. terminal platform >, so as to realize a monitoring alarm function;
and according to [0310], < E3. identification module > simultaneously sending confirmation signals to < E1. control module >, confirming that the alarm position is automatically analyzed and identified, and avoiding repeated identification.
In the above method for detecting a remote night vision target video, preferably, the early warning signal in step [0200] is converted by the following method:
according to [0200], the early warning signal selection strategy is a random strategy or a head priority strategy; the first priority strategy is that when the early warning signal set is { (x1, y1, h1), (x2, y2, h2) … … (xn, yn, hn) }, the first early warning signal (x1, y1, h1) is selected, and the rest signals are discarded; the random strategy is to randomly select any one signal and discard the rest signals;
according to [0200], before system deployment and operation, firstly selecting n calibration points in a monitoring area of a thermal imaging acquisition device, wherein n is a natural number greater than 2, recording coordinates of the calibration points and height { (X1, Y1, H1), (X2, Y2, H1) … … (Xn, Yn, H1) } sets of standard size targets, and simultaneously recording < C-holder > coordinates, < B-laser emitter > spot size and < D-visible light/infrared camera > focal length { (HLVX1, HLVY1, W1, F1), (HLVX2, HLVY2, W2, F2) … … (HLVXn, HLYVn, Wn, Fn) } sets of corresponding calibration points;
according to [0202], sorting the X coordinate and the Y coordinate of the index point and the target height H from big to small or from small to big respectively, and sorting { Xn-1, X2, X3, Xn, … … X1}, { Yn, Y1, Y2, Yn-1, … …, Y3}, { H2, Hn, H1 … … H3 }; simultaneously arranging corresponding < C. pan-tilt > coordinates, < B. laser transmitter > light spot and < D. visible/infrared camera > focal length in a corresponding order { HLVXn-1, HLVX2, HLVX3, … … HLVX1, HLVXn }, { HLVYn, HLVY1, HLVYn-1, … …, HLVY2, HLVY3}, { W2, Wn, W1 … … W3}, { F2, Fn, F1 … … F3 };
according to [0203]The conversion mode of (x, y, h) of the early warning signal and the control command (HLVX, HLVY, W, F) is as follows, wherein the technical scheme for converting x into HLVX is as follows: firstly, searching whether a value in { Xn-1, X2, X3, Xn, … … X1} is equal to X, and if X2 ═ X, HLVX ═ HLVX 2; if not, find between which X X lies, e.g. Xa＜x＜XbThen adopt the formula
Calculating corresponding HLVX;
according to [0204], y is converted into HLVY, h is converted into W and F in the same way as x is converted into HLVX, and the calculation formula is as follows:
in the above method for detecting a remote night vision target video, preferably, the alarm information in step [0310] includes: alarm device IP, alarm target class line, alarm target number, alarm time.
In the above remote night vision target video detection method, preferably, in [0312], the < e1. control module > stores the confirmation signal sent from the < E3. identification module >, compares the confirmation signal with the subsequent warning signal of [0200], and when the confirmation signal is consistent with the warning signal, no instruction is sent to the < b. laser emitter >, < c. pan-tilt >, < d. visible light/infrared camera >.
This patent has adopted leading embedded analysis processing unit, with analysis processing core device < E. the main control unit > preposition, integrates, realizes the simplification of whole monitoring analysis device, has reduced the hardware cost, is favorable to implementing the deployment, falls to the ground to use on a large scale for providing the guarantee.
Through the scheme, when related products and technologies of laser and thermal imaging cameras are introduced, a front-end embedded intelligent analysis module technology is creatively adopted, and a new generation algorithm technology based on deep learning is combined, so that the automatic detection and identification functions of a large-range and long-distance night vision target in a complex scene are realized.
The invention has the advantages of
1. The high-performance embedded technology is utilized, the video target analysis algorithm is arranged in front of the front-end equipment of the video analysis system, and after the front end finds the alarm target, the alarm information is sent to the rear-end server, so that the high integration of the monitoring device is realized, and the large-scale floor application deployment is facilitated.
2. This patent is based on the degree of depth learning frame, and the sample is handled in the collection and the degree of depth learning refines the model, has realized the classification and the quick discernment of the target under the complicated control background condition.
3. The thermal imaging and infrared acquisition device is combined with a whole, excellent performance of each device is integrated, and a related method strategy is adopted, so that the problem of large-range and long-distance target tracking and positioning is well solved. The all-weather, all-section and automatic video monitoring capability is realized.
4. The video target type definition is provided, a humanoid target and a fire target entering a protection area can be identified, and the capability of expanding to other types of targets is achieved.
5. The patent utilizes modern computer vision technology and artificial intelligence technology, and realizes the positioning and identification of the target concerned in the video scene by analyzing the video sequence shot by the camera in real time without human intervention, and analyzes and judges the behavior of the target on the basis to detect a certain event mode or acquire the information content of a certain data type. Meanwhile, the warning information reminds related personnel of paying attention in a client or mobile phone short message mode, so that the practical response speed is improved, and the burden of the personnel is reduced.
In conclusion, the patent provides an all-weather, especially under the night vision environment, method and device for automatically detecting and identifying the target in a large range and in a long distance, which effectively solve the problems of insufficient automatic detection and identification capability, insufficient accuracy of real-time target analysis and identification technology, difficult large-scale practical application and deployment of monitoring equipment and the like under the complex dangerous environment, especially under the environment, realize the high-precision automatic detection and identification target under the long-distance and large-range night vision environment, and have the characteristics of function modularization, all weather, long distance, high integration, light weight, easy deployment and the like.
Drawings
FIG. 1 is a schematic flow chart of the present invention
FIG. 2 is a schematic view of the thermal imaging camera installation deployment of the present invention
Fig. 3 is a schematic illustration of the laser camera installation deployment of the present invention.
Detailed Description
Fig. 2-3 are schematic diagrams of installation and deployment of the present invention.
As shown in fig. 2, the front-end hardware system includes 3 thermal imaging cameras installed on one side of the GSM-R tower. Wherein the range of the 70 m thermal imaging monitoring is 0-70 m (the installation height is 20 m), the range of the 500 m thermal imaging monitoring is 70-500 m (the installation height is 20 m), and the range of the 1500 m thermal imaging monitoring is 500-1500 m (the installation height is 30 m). To ensure complete coverage of the surveillance zone, adjacent surveillance zones overlap by about 20%.
As shown in fig. 3, the front end hardware system includes 2 laser cameras with pan/tilt heads, and is installed on one side of the GSM-R communication tower. Wherein the 500 m thermal laser camera monitors the range of 0-500 m (the installation height is 25 m), and the 1500 m laser camera monitors the range of 500-1500 m (the installation height is 30 m). To ensure complete coverage of the surveillance zone, adjacent surveillance zones overlap by about 20%.
The present invention will be described in detail with reference to fig. 1.
[0000] The method and the device are composed of a thermal imaging acquisition device, a laser emitter, a tripod head, a visible light/infrared camera, a main control unit and a terminal platform, wherein the main control unit is divided into four small components, namely an E1 control module, an E2 early warning module, an E3. identification module and a E4. power supply;
a, a thermal imaging acquisition device sends acquired video stream data to an E2 early warning module in real time;
the early warning module analyzes the received video stream, identifies early warning signals, and obtains data sets { (x1, y1, h1), (x2, y2, h2) … … (xn, yn, hn) } of early warning target coordinate points and height information;
the early warning module sends an early warning signal target coordinate point and a height information data set to the control module E1;
the method comprises the following steps that (E1) a pre-warning signal is selected from a pre-warning signal data set according to a certain strategy, and (x, y, h) of the pre-warning signal is converted into corresponding instructions (HLVX, HLVY, W, F), wherein x is converted into HLVX, y is converted into HLVY, h is converted into W and F, and the W and F are respectively sent to a laser transmitter >, < C. cloud platform >, < D. visible light/infrared camera >;
according to [0200], the early warning signal selection strategy can be a random strategy or a first-order priority strategy; when the early warning signal set is { (x1, y1, h1), (x2, y2, h2) … … (xn, yn, hn) }, if a first priority strategy is adopted, the early warning signal (x1, y1, h1) is selected, and the rest signals are discarded; if a random strategy is adopted, any one of the signal sets is selected, for example, (x2, y2, h2), and the rest signals are discarded;
according to [0200], before system deployment and operation, firstly selecting n (n >2) calibration points in a monitoring area of a thermal imaging acquisition device, recording coordinates of the calibration points and height { (X1, Y1, H1), (X2, Y2, H1) … … (Xn, Yn, H1) } sets of standard-size targets, and simultaneously recording < C-holder > coordinates, < B-laser emitter > spot size and < D-visible light/infrared camera > focal length { (HLVX1, HLVY1, W1, F1), (HLVX2, HLVY2, W2, F2) … … (HLVVXYn, VHLYn, Wn, Fn) } sets of corresponding calibration points;
according to [0202], the X and Y coordinates of the index point and the target height H are respectively sorted from large to small or from small to large in worth of size, { Xn-1, X2, X3, Xn, … … X1}, { Yn, Y1, Y2, Yn-1, … …, Y3}, { H2, Hn, H1 … … H3 }; simultaneously arranging corresponding < C. pan-tilt > coordinates, < B. laser transmitter > light spot and < D. visible/infrared camera > focal length in a corresponding order { HLVXn-1, HLVX2, HLVX3, … … HLVX1, HLVXn }, { HLVYn, HLVY1, HLVYn-1, … …, HLVY2, HLVY3}, { W2, Wn, W1 … … W3}, { F2, Fn, F1 … … F3 };
according to [0203]First, taking the conversion of X into HLVX, taking X into the example of the conversion of (X, y, h) of the warning signal and the control command (HLVX, HLVY, W, F), it is searched whether a value in { Xn-1, X2, X3, Xn, … … X1} is equal to X, and if X2 is equal to X, HLVX is equal to HLVX 2; if not, find between which X X lies, e.g. Xa＜x＜XbThen adopt the formula
Calculating corresponding HLVX;
according to [0204], y is converted into HLVY, h is converted into W and F in the same way as x is converted into HLVX, and the calculation formula is as follows:
transmitting a laser instruction to the < B. laser emitter > according to the [0200], < E1. control module >;
according to [0210], the patent provides a laser automatic control device mechanism, which automatically judges whether laser needs to be emitted or not according to the current light signal, and ignores the early warning signal if the current light signal is strong; otherwise, calculating and determining the laser emission position and the aperture size according to the early warning information, and emitting laser at a fixed point;
according to the (0210) and (B) laser transmitters, after receiving the instruction, automatically calculating the laser emission position and the size of a light spot according to specific instruction parameters and early warning target position and height information, realizing intelligent light supplement at different distances and improving the night vision light supplement level;
sending horizontal rotation and pitching rotation instructions to the < C. pan-tilt > according to the [0200], < E1. control module >;
after receiving the instruction according to [0220], < C. cloud deck >, automatically calculating the horizontal and pitching rotation angles according to the instruction parameters and the early warning target position information, and reaching the specified position;
sending a zooming instruction to the < D. visible light/infrared camera > according to the [0200], < E1. control module >;
after receiving the instruction according to [0231] and < D. visible light/infrared camera >, automatically adjusting the focal length according to the instruction information parameters and the early warning target position and height information to obtain a proper video acquisition angle;
sending the collected video stream to an identification module of < E3. > by a visible light/infrared camera;
the identification module E3. analyzes, identifies and classifies the collected video stream to obtain alarm information;
organizing encapsulation alarm information according to [0310], < E3. identification module ], wherein the alarm information comprises contents such as alarm equipment IP, alarm target class lines, alarm target number, alarm time and the like, and sending the alarm information to < F. terminal platform > to realize a monitoring alarm function;
and according to [0310], < E3. identification module > simultaneously sending confirmation signals to < E1. control module >, confirming that the alarm position is automatically analyzed and identified, and avoiding repeated identification.
This patent has adopted leading embedded analysis processing unit, with analysis processing core device < E. the main control unit > preposition, integrates, realizes the simplification of whole monitoring analysis device, has reduced the hardware cost, is favorable to implementing the deployment, falls to the ground to use on a large scale for providing the guarantee.
Claims (4)
1. A remote night vision target video detection method is characterized by comprising the following steps:
[0000] the intelligent monitoring system comprises a thermal imaging acquisition device, a laser emitter, a tripod head, a visible light/infrared camera, a main control unit and a terminal platform, wherein the main control unit is divided into four small components, namely an E1 control module, an E2 early warning module, an E3. identification module and a E4. power supply;
[0100] a, a thermal imaging acquisition device sends acquired video stream data to an E2 early warning module in real time;
[0110] the early warning module analyzes the received video stream, identifies early warning signals, and obtains data sets { (x1, y1, h1), (x2, y2, h2) … … (xn, yn, hn) } of early warning target coordinate points and height information;
[0120] the early warning module sends an early warning signal target coordinate point and a height information data set to the control module E1;
[0200] the method comprises the following steps that (E1) a pre-warning signal is selected from a pre-warning signal data set according to a certain strategy, and (x, y, h) of the pre-warning signal is converted into corresponding instructions (HLVX, HLVY, W, F), wherein x is converted into HLVX, y is converted into HLVY, h is converted into W and F, and the W and F are respectively sent to a laser transmitter >, < C. cloud platform >, < D. visible light/infrared camera >;
[0210] transmitting a laser instruction to the < B. laser emitter > according to the [0200], < E1. control module >;
[0211] according to [0210], the laser transmitter automatically judges whether laser needs to be transmitted or not according to the current light signal, and ignores the early warning signal if the current light signal is strong; otherwise, calculating and determining the laser emission position and the aperture size according to the early warning information, and emitting laser at a fixed point;
[0212] according to the (0210) and (B) laser transmitters, after receiving the instruction, automatically calculating the laser emission position and the size of a light spot according to specific instruction parameters and early warning target position and height information, realizing intelligent light supplement at different distances and improving the night vision light supplement level;
[0220] sending horizontal rotation and pitching rotation instructions to the < C. pan-tilt > according to the [0200], < E1. control module >;
[0221] after receiving the instruction according to [0220], < C. cloud deck >, automatically calculating the horizontal and pitching rotation angles according to the instruction parameters and the early warning target position information, and reaching the specified position;
[0230] sending a zooming instruction to the < D. visible light/infrared camera > according to the [0200], < E1. control module >;
[0230] after receiving the instruction according to [0231] and < D. visible light/infrared camera >, automatically adjusting the focal length according to the instruction information parameters and the early warning target position and height information to obtain a proper video acquisition angle;
[0300] sending the collected video stream to an identification module of < E3. > by a visible light/infrared camera;
[0310] the identification module E3. analyzes, identifies and classifies the collected video stream to obtain alarm information;
[0311] organizing packaging alarm information according to [0310], < E3. identification module ], and sending the alarm information to < F. terminal platform >, so as to realize a monitoring alarm function;
[0312] and according to [0310], < E3. identification module > simultaneously sending confirmation signals to < E1. control module >, confirming that the alarm position is automatically analyzed and identified, and avoiding repeated identification.
2. The remote night vision target video detection method as claimed in claim 1, wherein said early warning signal of step [0200] is converted by:
[0201] according to [0200], the early warning signal selection strategy is a random strategy or a head priority strategy; the first priority strategy is that when the early warning signal set is { (x1, y1, h1), (x2, y2, h2) … … (xn, yn, hn) }, the first early warning signal (x1, y1, h1) is selected, and the rest signals are discarded; the random strategy is to randomly select any one signal and discard the rest signals;
[0202] according to [0200], before system deployment and operation, firstly selecting n calibration points in a monitoring area of a thermal imaging acquisition device, wherein n is a natural number greater than 2, recording coordinates of the calibration points and height { (X1, Y1, H1), (X2, Y2, H1) … … (Xn, Yn, H1) } sets of standard size targets, and simultaneously recording < C-holder > coordinates, < B-laser emitter > spot size and < D-visible light/infrared camera > focal length { (HLVX1, HLVY1, W1, F1), (HLVX2, HLVY2, W2, F2) … … (HLVXn, HLYVn, Wn, Fn) } sets of corresponding calibration points;
[0203] according to [0202], sorting the X coordinate and the Y coordinate of the index point and the target height H from big to small or from small to big respectively, and sorting { Xn-1, X2, X3, Xn, … … X1}, { Yn, Y1, Y2, Yn-1, … …, Y3}, { H2, Hn, H1 … … H3 }; simultaneously arranging corresponding < C. pan-tilt > coordinates, < B. laser transmitter > light spot and < D. visible/infrared camera > focal length in a corresponding order { HLVXn-1, HLVX2, HLVX3, … … HLVX1, HLVXn }, { HLVYn, HLVY1, HLVYn-1, … …, HLVY2, HLVY3}, { W2, Wn, W1 … … W3}, { F2, Fn, F1 … … F3 };
[0204]according to [0203]The conversion mode of (x, y, h) of the early warning signal and the control command (HLVX, HLVY, W, F) is as follows, wherein the technical scheme for converting x into HLVX is as follows: firstly, searching whether a value in { Xn-1, X2, X3, Xn, … … X1} is equal to X, and if X2 ═ X, HLVX ═ HLVX 2; if not, find between which X X lies, e.g. Xa＜x＜XbThen adopt the formula
Calculating corresponding HLVX;
[0205] according to [0204], y is converted into HLVY, h is converted into W and F in the same way as x is converted into HLVX, and the calculation formula is as follows:
3. the remote night vision target video detection method of claim 1, wherein the alarm information in step [0310] includes: alarm device IP, alarm target class line, alarm target number, alarm time.
4. The remote night vision target video detection method of claim 1, comprising: in [0312], the < e1. control module > stores the confirmation signal sent by the < E3. identification module >, compares the confirmation signal with the subsequent early warning signal of [0200], and does not send instructions to the < b. laser emitter >, < c. pan-tilt >, < d. visible light/infrared camera > when the confirmation signal is consistent with the early warning signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911402390.4A CN111063148A (en) | 2019-12-30 | 2019-12-30 | Remote night vision target video detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911402390.4A CN111063148A (en) | 2019-12-30 | 2019-12-30 | Remote night vision target video detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111063148A true CN111063148A (en) | 2020-04-24 |
Family
ID=70305051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911402390.4A Pending CN111063148A (en) | 2019-12-30 | 2019-12-30 | Remote night vision target video detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111063148A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113011313A (en) * | 2021-03-18 | 2021-06-22 | 杭州海宴科技有限公司 | Automatic goods shelf channel safety detection method and device |
| WO2022126375A1 (en) * | 2020-12-15 | 2022-06-23 | 深圳市大疆创新科技有限公司 | Zooming method, imaging device, gimbal and movable platform |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101933058A (en) * | 2008-01-31 | 2010-12-29 | 传感电子有限责任公司 | Video sensor and alarm system and method with object and event classification |
| US20110043806A1 (en) * | 2008-04-17 | 2011-02-24 | Avishay Guetta | Intrusion warning system |
| CN103473926A (en) * | 2013-09-11 | 2013-12-25 | 无锡加视诚智能科技有限公司 | Gun-ball linkage road traffic parameter collection and rule breaking snapshooting system |
| CN205160670U (en) * | 2015-07-28 | 2016-04-13 | 杭州振讯科技有限公司 | Laser thermal imaging automatic tracking system |
| CN106571014A (en) * | 2016-10-24 | 2017-04-19 | 上海伟赛智能科技有限公司 | Method for identifying abnormal motion in video and system thereof |
| CN107253485A (en) * | 2017-05-16 | 2017-10-17 | 北京交通大学 | Foreign matter invades detection method and foreign matter intrusion detection means |
| CN207117799U (en) * | 2017-07-03 | 2018-03-16 | 深圳市保千里电子有限公司 | A kind of long-distance infrared laser night-vision devices |
| CN207150735U (en) * | 2017-05-31 | 2018-03-27 | 北京世纪瑞尔软件有限公司 | A kind of round-the-clock auto-alarming tracks of device |
| CN108447075A (en) * | 2018-02-08 | 2018-08-24 | 烟台欣飞智能系统有限公司 | A kind of unmanned plane monitoring system and its monitoring method |
-
2019
- 2019-12-30 CN CN201911402390.4A patent/CN111063148A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101933058A (en) * | 2008-01-31 | 2010-12-29 | 传感电子有限责任公司 | Video sensor and alarm system and method with object and event classification |
| US20110043806A1 (en) * | 2008-04-17 | 2011-02-24 | Avishay Guetta | Intrusion warning system |
| CN103473926A (en) * | 2013-09-11 | 2013-12-25 | 无锡加视诚智能科技有限公司 | Gun-ball linkage road traffic parameter collection and rule breaking snapshooting system |
| CN205160670U (en) * | 2015-07-28 | 2016-04-13 | 杭州振讯科技有限公司 | Laser thermal imaging automatic tracking system |
| CN106571014A (en) * | 2016-10-24 | 2017-04-19 | 上海伟赛智能科技有限公司 | Method for identifying abnormal motion in video and system thereof |
| CN107253485A (en) * | 2017-05-16 | 2017-10-17 | 北京交通大学 | Foreign matter invades detection method and foreign matter intrusion detection means |
| CN207150735U (en) * | 2017-05-31 | 2018-03-27 | 北京世纪瑞尔软件有限公司 | A kind of round-the-clock auto-alarming tracks of device |
| CN207117799U (en) * | 2017-07-03 | 2018-03-16 | 深圳市保千里电子有限公司 | A kind of long-distance infrared laser night-vision devices |
| CN108447075A (en) * | 2018-02-08 | 2018-08-24 | 烟台欣飞智能系统有限公司 | A kind of unmanned plane monitoring system and its monitoring method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022126375A1 (en) * | 2020-12-15 | 2022-06-23 | 深圳市大疆创新科技有限公司 | Zooming method, imaging device, gimbal and movable platform |
| CN113011313A (en) * | 2021-03-18 | 2021-06-22 | 杭州海宴科技有限公司 | Automatic goods shelf channel safety detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101942491B1 (en) | Hybrid ai cctv mediation module device consisting of road traffic situation monitoring and real time traffic information analysis | |
| CN109709589B (en) | Sea and air area three-dimensional perception prevention and control system | |
| CN110738846B (en) | Vehicle behavior monitoring system based on radar and video group and implementation method thereof | |
| CN111770266B (en) | Intelligent visual perception system | |
| CN110133573A (en) | A kind of autonomous low latitude unmanned plane system of defense based on the fusion of multielement bar information | |
| CN108227606B (en) | Ship security intelligent management system based on multi-source perception | |
| CN109333504A (en) | A kind of patrol robot and patrol robot management system | |
| KR100492148B1 (en) | The Artificial Intelligence Image Security System using the distance and direction of Moving Object | |
| CN108802758B (en) | Intelligent security monitoring device, method and system based on laser radar | |
| CN106657921A (en) | Portable radar perimeter security and protection system | |
| CN104394361A (en) | Pedestrian crossing intelligent monitoring device and detection method | |
| CN103929592A (en) | All-dimensional intelligent monitoring equipment and method | |
| CN103295396A (en) | Method and system for speedy ex-situ evidence collection of traffic accident | |
| CN112449093A (en) | Three-dimensional panoramic video fusion monitoring platform | |
| KR102509570B1 (en) | Control device using artificial intelligence learning images and electrical signals and intrusion alert systme including the same | |
| CN101330595A (en) | Gang control system for ultra long range radar and video | |
| CN106548251A (en) | A kind of electronic monitoring and control system and method based on main passive fusion | |
| CN111063148A (en) | Remote night vision target video detection method | |
| CN104038733A (en) | Sentinel early warning monitoring system | |
| CN109917364A (en) | A kind of warning system of integrated radar guiding and photoelectric tracking function | |
| CN113034828A (en) | System for realizing target detection and identification based on embedded computing terminal and layout method | |
| CN112257683A (en) | Cross-mirror tracking method for vehicle running track monitoring | |
| CN112349055A (en) | Target monitoring device and method based on radar and video linkage | |
| CN115035470A (en) | Low, small and slow target identification and positioning method and system based on mixed vision | |
| CN114283544B (en) | Railway platform intrusion monitoring system and method based on artificial intelligence |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200424 |
|
| RJ01 | Rejection of invention patent application after publication |