CN102693632B - Monitoring device and monitoring method based on visual image - Google Patents
Monitoring device and monitoring method based on visual image Download PDFInfo
- Publication number
- CN102693632B CN102693632B CN201210127924.9A CN201210127924A CN102693632B CN 102693632 B CN102693632 B CN 102693632B CN 201210127924 A CN201210127924 A CN 201210127924A CN 102693632 B CN102693632 B CN 102693632B
- Authority
- CN
- China
- Prior art keywords
- mentioned
- video
- image
- monitoring processor
- target
- 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.)
- Expired - Fee Related
Links
Images
Abstract
The invention provides a monitoring device and a monitoring method based on a visual image. The device comprises a focusing camera, an electric rotational station for controlling omni-directional rotation of the camera, and a monitoring processor for controlling the camera and the electric rotational station and receiving and processing data information fed back by the camera and the electric rotational station. The monitoring processor comprises a human-computer interaction module, a camera control module, a target detection module, an image calibration module and a video conversion module. The method for applying the device to monitor a specific target can be applied in two working modes: ifa target image in a panoramic monitoring video is not distinct,a monitoring system employs real-time monitoring, detection, tracking and automatic focusing to capturea high definition image of the target; and if a target image in the panoramic monitoring video is distinct, video recording and then analyzing are employed for monitoring, and video can be played backwards. Different working modes are employed according to different situations to guarantee that image and the video containingthe distinct features of the specific image target can be finally obtained.
Description
Technical field
The invention belongs to security protection and intelligent transportation field, be specifically related to a kind of supervising device and method for supervising based on visual pattern.
Background technology
Along with the high speed development of Chinese society, security protection and traffic problems are day by day remarkable, this trip to people, and even life security has all produced great impact.
Based on the supervising device of visual pattern, with unique advantage, obtained very wide application, be for example arranged on road monitoring whether have vehicle etc. more, pressed double amber lines to travel in violation of rules and regulations, whether there is vehicle etc. to ride on stop line, also or whether monitoring has vehicle etc. to be parked in territory, prohibition parking area.For some application scenarios, for example, at freeway toll station, should monitor and whether have vehicle to get over, pressed the middle double amber lines of road to travel in violation of rules and regulations, whether monitor again apart from charge station a distance has rule-breaking vehicle to stop, and need to capture evidence obtaining to those vehicles in violation of rules and regulations, need to ensure that the image forensic or the video that obtain are clearly simultaneously.But the mode of operation of existing supervising device is more single, a set of supervising device can not meet above-mentioned multiple requirement simultaneously, can not realize above-mentioned several functions simultaneously, often needs the different supervising device of two covers to carry out respectively function separately.So shortcoming such as existing supervising device ubiquity that monitoring range is little, function singleness, specific aim are poor.
Summary of the invention
For solving the deficiencies in the prior art, the object of the present invention is to provide one to there is multi-functional supervising device and method for supervising, according to the different mode of operation of the whether clear employing of specific image target, final image and the video recording that obtains the clear feature that contains specific image target.
In order to realize above-mentioned target, the present invention adopts following technical scheme:
Based on a supervising device for visual pattern, it is characterized in that, comprising:
The video camera that also can focus by programmed control,
The electric platform of the aforementioned video camera all-direction rotation of variable speed controllable Electric Machine Control being provided with by self,
For controlling aforementioned video camera and electric platform and receiving and process aforementioned video camera and the monitoring processor of the data message that electric platform feeds back;
The wire communication that aforementioned video camera, electric platform and monitoring processor form energy transfer data information is connected or wireless telecommunications connection.
Aforesaid a kind of supervising device based on visual pattern, it is characterized in that, aforementioned electric platform is provided with the angular encoder for video camera rotation information being fed back to aforementioned monitoring processor, and aforementioned angular encoder is connected with aforementioned monitoring processor formation wire communication or wireless telecommunications connect.
Aforesaid a kind of supervising device based on visual pattern, is characterized in that, aforementioned monitoring processor comprises:
For the human-computer interaction module of data message input, feedback,
For controlling the camera control module of aforementioned video camera and electric platform,
The module of target detection detecting for the target of view data that aforementioned video camera is obtained,
Realize the video conversion module of subsequent treatment and Reverse Play for the video conversion that aforementioned video camera is obtained,
For set up the image calibration module of the corresponding relation of image and actual scene according to camera parameters.
Aforesaid a kind of supervising device based on visual pattern, it is characterized in that, aforementioned resolution of video camera more than 1080 × 1920, the high definition high-speed camera that can rotate within the scope of horizontal 360-degree, vertical 180 degree, the image visual field of aforementioned high definition high-speed camera is that long and narrow bearing of trend square and narrow rectangular long limit is consistent with road direction.
The method for supervising of aforesaid device, is characterized in that, comprises the steps:
One, supervising device initialization:
(1) input the initial parameter for carrying out image calibration by above-mentioned human-computer interaction module, the initial parameter of input comprises: the setting height(from bottom) of aforementioned video camera, actual monitored scope, optical parametric and image parameter in general monitor state video camera of lower time, and aforementioned initial parameter is kept in monitoring processor;
(2) earlier figures is set up the corresponding relation of image and actual scene according to aforementioned initial parameter as demarcating module, set the parameter of the realistic objective that needs monitoring, set up with reference to aforementioned corresponding relation the parameter corresponding in image and assert condition, qualified image object is specific image target;
(3) the aforementioned video camera of aforementioned monitoring processor control starts to monitor aforementioned actual monitored scope, within the scope of aforementioned actual monitored, draw to have and judge whether in violation of rules and regulations actual surveyed area of target, image range corresponding to aforementioned actual monitored scope is picture control scope, and the image that aforementioned actual surveyed area is corresponding is virtual detection region;
Two, the triggering of image object and the selection of mode of operation: occur after aforementioned specific image target in aforementioned monitoring processor is found the video single-frame images of its reception, the video having received is detected, whether judgement wherein contains the high-resolution single frames picture that can show specific image target identification feature, and aforementioned specific image target identification feature refers to can be according to the characteristics of image of the physical features of its corresponding realistic objective of the image identifying of specific image target; If do not contained, adopt the first evidence obtaining mode of operation; If contained, adopt the second evidence obtaining mode of operation;
Three, collect evidence, obtain and preserve image forensic data message: aforementioned monitoring processor and video camera match and adopt aforementioned the first evidence obtaining mode of operation or the second evidence obtaining mode of operation processing of collecting evidence, and obtain and preserve image forensic data message.
Aforesaid method for supervising, is characterized in that, the detailed process of aforementioned the first evidence obtaining mode of operation is:
(1) video camera obtains the monitor video of current scene by gathering continuous single frames;
(2) monitoring processor obtains the single frames of a current scene at regular intervals from aforementioned monitor video, whether utilize background subtraction point-score to detect in the single frames of aforementioned current scene has specific image target to enter virtual detection region, if specific image target does not enter aforementioned virtual detection region, aforementioned monitoring processor repeating step (2); If specific image target has entered aforementioned virtual detection region, aforementioned monitoring processor starts timing, and carries out step (3), and this timing time is specific objective acknowledging time;
(3) monitoring processor obtains the single frames of a current scene at regular intervals from aforementioned monitor video, whether the virtual detection region that utilizes background subtraction point-score to detect in aforementioned single frames also has specific image target, if no, aforementioned monitoring processor repeating step (2); If had, and aforementioned specific objective acknowledging time does not reach predefined threshold value, repeating step (3); If had, and aforementioned specific objective acknowledging time reaches predefined threshold value, and the aforementioned specific objective of the aforementioned camera alignment of aforementioned monitoring processor control is focused and captures.
Aforesaid method for supervising, is characterized in that, the detailed process of aforementioned the second evidence obtaining mode of operation is:
(1) monitoring processor control video camera is recorded a video to guarded region, and every video of set time camera switching, the video of recording end enters step (2) as pending video;
(2) whether monitoring processor enters aforementioned virtual detection region to the specific image target in the single frames of aforementioned pending video and detects, if do not entered, changes single frames and proceeds step (2); If entered, monitoring processor predicts that aforementioned specific image target arrives the needed time of image forensics position, this time is the pre-evidence obtaining time, aforementioned image forensics position is the position in the correspondence image of target evidence obtaining position, aforementioned target evidence obtaining position is that video camera obtains the specific objective position at specific objective place in actual scene when image forensic data message clearly, and this pending video enters as step (3) as evidence obtaining video;
(3) aforementioned monitoring processor carries out video direction finding and starts timing aforementioned specific image target, if aforementioned specific image target is moved to the direction near video camera, aforementioned monitoring processor still reads the single frames of aforementioned evidence obtaining video with normal sequence; If to the direction motion away from video camera, aforementioned monitoring processor starts inverted order and reads the single frames of aforementioned evidence obtaining video;
(4), when monitoring processor reads the time aforementioned pre-evidence obtaining time of arrival of aforementioned evidence obtaining video, preserve evidence obtaining single frames and the aforementioned evidence obtaining video of current time.
Usefulness of the present invention is: user manually delimit virtual detection region as required, have usable range widely; Within the scope of picture control, for the corresponding specific image target of specific objective different mode of operation of clear employing whether in single frames, final image and the video recording that obtains the clear feature that contains specific image target.
Brief description of the drawings
Fig. 1 be a kind of supervising device based on visual pattern of the present invention a specific embodiment install and use schematic diagram;
Fig. 2 is the schematic top plan view of a concrete application scenarios of a kind of supervising device based on visual pattern of the present invention;
Fig. 3 be in a kind of method for supervising based on visual pattern of the present invention spotting apart from the demarcation schematic diagram of true horizon distance and the actual size of video camera;
Fig. 4 is the main flow schematic diagram of two kinds of mode of operations in a specific embodiment of a kind of method for supervising based on visual pattern of the present invention;
Fig. 5 is the idiographic flow schematic diagram of the second mode of operation in Fig. 4;
The implication of Reference numeral in figure:
1-video camera, 2-electric platform, 3-monitoring processor, 4-the first surveyed area, 5-the second surveyed area, 6-image forensics position.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is done to concrete introduction.
With reference to Fig. 1, a kind of supervising device based on visual pattern of the present invention, comprising: video camera 1, electric platform 2 and monitoring processor 3, the wire communication that forms each other energy transfer data information connects or wireless telecommunications connection.
As the preferred scheme of one, video camera 1 is for having the video camera of electronic Tri-Chargable lens.
As the preferred scheme of one, for capturing and record out the image of high-resolution, video camera 1 be resolution more than 1920 × 1080, and within the scope of 360 ° of levels, vertical 180 °, all can do the high definition high-speed camera that speed change is rotated.
When the zone length of needs monitoring and the ratio of width are when larger, as the preferred scheme of one, the image visual field of video camera 1 is long and narrow square, and the bearing of trend on narrow rectangular long limit is consistent with road direction, not only expand monitoring range, and do not collect image for the region video camera 1 that does not need monitoring beyond road, so also saved the time of monitoring processor 3 image data processing information, improve operation efficiency.
Electric platform 2(is not shown) control video camera 1 all-direction rotation, self be provided with variable speed controllable motor (not shown).
As the preferred scheme of one, electric platform 2 is provided with angular encoder (not shown), and angular encoder and monitoring processor 3 form that wire communication is connected or wireless telecommunications connect, for video camera 1 rotation information is fed back to monitoring processor 3.There is the electric platform 2 of variable speed controllable motor and angular encoder, ensured that video camera 1 can, rotating speed slow miss candid photograph target little because of corner.
As the preferred scheme of one, monitoring processor 3 comprises: human-computer interaction module, camera control module, module of target detection, video conversion module and image calibration module.
Human-computer interaction module: for input and the feedback of data message and data message is kept to monitoring processor, data message comprises: the setting height(from bottom) of video camera 1, actual monitored scope, optical parametric and image parameter in general monitor state video camera of lower time 1, the such as pixel dimension of the focal length of video camera 1, CCD etc.
Camera control module: the output control video camera 1 according to monitoring processor 1 to processing data information and the work of electric platform 2.
Whether module of target detection: the target in the view data that video camera 1 is obtained detects, namely have target to occur in detected image.
Video conversion module: the video conversion that video camera 1 is obtained is realized and facilitated subsequent treatment and Reverse Play, be mainly reflected in interval between the key frame in the video that reduction camera 1 obtains, video is carried out to Reverse Play, key frame refers to by module of target detection and detects the frame for containing specific image target.
Image calibration module: set up the corresponding relation of image and actual scene according to the parameter of video camera 1, parameter comprises: the setting height(from bottom) of video camera 1, actual monitored scope, optical parametric and image parameter etc. in general monitor state video camera of lower time 1, by the corresponding relation of image and actual scene, after being processed, view data just can obtain the clarification of objective information in actual scene, for example size of the position at specific objective place, a certain Place object.
Utilize above-mentioned supervising device to monitor, for the ease of understanding, with reference to Fig. 2, be applied in freeway toll station describe as example taking supervising device, details are as follows for method for supervising:
After watch-dog installs and switches on, user is by " demarcation information " function button on the guidance panel (not shown) of click monitoring processor, the demarcation information exchanges such as the pixel dimension of the focal length of video camera 1, field angle, CCD, setting height(from bottom), actual monitored scope are crossed to human-computer interaction module and be kept in monitoring processor 3, analyze, judge the basis of the big or small corresponding relation of image and actual scene as monitoring processor 3, user is by " surveyed area " function button on clicking operation panel, need on the picture of image, draw virtual detection region to be detected with mouse according to different monitoring, this virtual detection region is corresponding with the tested region of reality in actual scene, in the present embodiment, actual tested region is double amber lines and the charge station's certain distance territory, vehicle prohibition parking area that is located at road both sides in addition in the middle of freeway toll station road nearby, be denoted as respectively the first surveyed area 4, the second surveyed area 5, in image respectively to there being the first virtual detection region, the second virtual detection region, human-computer interaction module is preserved the first virtual detection region and the second virtual detection region delimited, wait for that monitored processor 3 calls in follow-up processing procedure.
Then, image calibration module in monitoring processor 3 is according to the corresponding relation that is kept at parameter in human-computer interaction module and sets up image and actual scene, and with reference to Fig. 3, circular is as follows:
The field angle of supposing video camera 1 is θ
fov, focal length is that f, setting height(from bottom) are that the pixel dimension of h, CCD is l.The p if the vertical curve at video camera 1 place and ground intersect at a point, a ' is target position on the ground, and b ', c ' they are ground two points crossing with field of view edge, and the distance that b ' is ordered to p is less than the distance that c ' is ordered to p, and a ' is L with the horizontal range of video camera 1
pa ', some a ', b ' become location of pixels corresponding on image to be respectively a, b and c with c ' at CCD, and oo ' is the optical axis of video camera 1, can obtain the location point a ' at target place in actual scene and the horizontal range L of video camera 1 according to above-mentioned data message
pa ':
If the actual width of target is w ', it becomes the imaging width on image at CCD is w, and its horizontal range apart from video camera 1 is L, has:
Wherein, l
acand l
bcfor the pixel distance that in image, a point, b point and c are ordered, L
pa 'and L
pc 'the distance of ordering for a ' point in actual scene, c ' and p.
User sets the parameter of the realistic objective that needs monitoring according to actual conditions, just can set up the parameter corresponding in image assert condition with reference to above-mentioned corresponding relation, and qualified image object is specific image target.
Next, video camera 1 and monitoring processor 3 carry out self-inspection and initialize routine, then monitoring processor 3 is controlled video camera 1 and is started to monitor within the scope of actual monitored, and the target starting entering the first virtual detection region, the second virtual detection region detects, be exactly that the double amber lines in the middle of freeway toll station road nearby and the corresponding image section in territory, vehicle prohibition parking area that is located at road both sides beyond charge station's certain distance are detected in the present embodiment.
Due to vehicle be must be by charge station gateway, and not necessarily stop in territory, prohibition parking area, so the focal length of video camera 1 is located near of gateway, so, the image that near target video camera 1 through gateway time obtains is all clearly, certainly, because the set focal length of video camera 1 has been aimed at nearby, so take in the time of the target in territory, prohibition parking area relatively at a distance, the image that video camera 1 obtains is all unsharp.
Certainly, it should be noted that, user can also, according to the needs of actual conditions, be arranged on the distant place of camera lens from video camera 1 by focal length, and so the image of the camera lens of video camera 1 target is more nearby unsharp.So the setting of focal length is not limited to shown in the present embodiment.
In the present embodiment, if detect that vehicle occurs in charge station at a distance, beyond the focal range at video camera 1, so the image of the virtual vehicle obtaining is unsharp, virtual vehicle refers to image section corresponding with actual vehicle in image, monitoring processor 3 adopts the first evidence obtaining mode of operation to carry out the processing of image data information, that is:
(1) the continuous single frames of video camera 1 Real-time Collection obtains the monitor video of current scene;
(2) monitoring processor 3 obtains the single frames of a current scene at regular intervals from video camera 1, whether utilize background subtraction point-score to detect in this single frames has virtual vehicle to enter the second virtual detection region, if virtual vehicle does not enter the second virtual detection region, monitoring processor 3 repeating steps (2), if virtual vehicle has entered the second virtual detection region, monitoring processor 3 starts timing, and carry out step (3), this timing time is particular vehicle acknowledging time, the object of setting this particular vehicle acknowledging time is in order to prevent wrong judgement target, be presented as in the present embodiment, prevent the vehicle that crosses territory, prohibition parking area to be judged in this region and to stop, if the particular vehicle acknowledging time of vehicle has exceeded the threshold value of prior setting, and this vehicle is also in this region, determine its certain violation, collect evidence, leave this region if also do not arrive this vehicle of threshold value at particular vehicle acknowledging time, do not assert that it in violation of rules and regulations, without collecting evidence.
(3) monitoring processor 3 obtains the single frames of a current scene at regular intervals from above-mentioned monitor video, whether the second virtual detection region that utilizes background subtraction point-score to detect in this single frames also has virtual vehicle, if no, monitoring processor 3 repeating steps (2); If had, and particular vehicle acknowledging time does not reach predefined threshold value, repeating step (3); If had, and particular vehicle acknowledging time reaches predefined threshold value, illustrate that this vehicle has rested on this territory, prohibition parking area really, need to collect evidence, so the vehicle that monitoring processor 3 is controlled in video camera 1 this territory, prohibition parking area of aligning is focused, is captured, and preserves this image forensic data message.
If detect that vehicle occurs in charge station more nearby, due in the focal range of video camera 1, so the image of the virtual vehicle obtaining is clearly, monitoring processor 3 adopts the second evidence obtaining mode of operation to carry out the processing of image data information, that is:
(1) monitoring processor 3 is controlled video camera 1 guarded region is recorded a video, and switches a video every set time video camera 1, and the video of recording end enters step (2) as pending video;
(2) whether monitoring processor 3 enters the first virtual detection region to the virtual vehicle in the single frames of pending video and detects, namely whether get over, pressed the double amber lines in the middle of road to detect to vehicle, if do not entered, change single frames and proceed step (2), if entered, as the preferred scheme of one, monitoring processor 3 carries out video frequency speed-measuring to virtual vehicle, obtain this virtual vehicle and arrive the needed time of image forensics position, this time is the pre-evidence obtaining time, this image forensics position is the position in the correspondence image of target evidence obtaining position, this target evidence obtaining position is that video camera 1 obtains the vehicle position at vehicle place in actual scene when image forensic data message clearly, so target evidence obtaining position is than the camera lens of the more close video camera 1 in the position in the first virtual detection region, predict that by video frequency speed-measuring virtual vehicle arrives the needed time of image forensics position more accurate, next, this pending video enters as step (3) as evidence obtaining video,
(3) monitoring processor 3 carries out video direction finding and starts timing virtual vehicle, if this virtual vehicle is to the direction motion near video camera 1, illustrate and also do not arrive target evidence obtaining position, monitoring processor 3 still reads the single frames of this evidence obtaining video with normal sequence, in the time that monitoring processor 3 reads the time above-mentioned pre-evidence obtaining time of arrival of evidence obtaining video, illustrate that in violation of rules and regulations vehicle has arrived near target evidence obtaining position or its, the image of this violation vehicle that now video camera 1 obtains is clearly, monitoring processor 3 is preserved evidence obtaining single frames and the above-mentioned evidence obtaining video of current time.If find that through video direction finding virtual vehicle is to the direction motion away from video camera 1, detect that at monitoring processor 3 vehicle more, while pressing double amber lines, this vehicle has passed through target evidence obtaining position, in order to obtain the clear single frames of vehicle in target evidence obtaining position, monitoring processor 3 calls video conversion module to carry out inverted order and reads its single frames evidence obtaining video, in the time that inverted order reads the time above-mentioned pre-evidence obtaining time of arrival of evidence obtaining video, illustrate that in violation of rules and regulations vehicle has arrived near target evidence obtaining position or its, the image of this violation vehicle that now video camera 1 obtains is clearly, monitoring processor 3 is preserved evidence obtaining single frames and the above-mentioned evidence obtaining video of current time.
As the preferred scheme of one, if it is little to enter the changes in vehicle speed of the second surveyed area 5, the vehicle of for example running on expressway, monitoring processor 3 is just without first passing through the video frequency speed-measuring prediction prefetch card time more so, as long as collect evidence the time in advance according to collect evidence in advance time and a minimum of maximum of the ultimate value prediction of speed, conventionally the ultimate value of speed is: average velocity × (1 ± 20%), monitoring processor 3 selects normal sequence to read evidence video or reverse read evidence video according to video direction finding again, and be kept at multiple single frames and this evidence obtaining video of the evidence video between minimum pre-evidence obtaining time and maximum pre-evidence obtaining time.In the method, the speed of target travel is known, has saved video frequency speed-measuring step, has greatly simplified computation process.
As the preferred scheme of one, if it is obvious to enter the target signature difference of the second surveyed area 5, the size of for example target has larger difference, monitoring processor 3 is also without the prediction prefetch card time so, as long as extract this clarification of objective, then select normal sequence to read evidence video or reverse read evidence video according to video direction finding, according to the feature of extracting, this target is followed the tracks of again, in the time that target arrives target evidence obtaining position, be kept at the single frames that contains the clear feature of target and this evidence video that obtain herein.
The video to be checked obtaining for video camera 1, only preserve and contain the target evidence obtaining video of evidence in violation of rules and regulations, do not preserve for wherein there is no target video to be checked in violation of rules and regulations, that is to say, the video that monitoring processor 3 preserves is all to contain target evidence in violation of rules and regulations, has so greatly improved the speed of obtain evidence video, has reduced workload, improve work efficiency, also optimized the configuration of resource simultaneously.
It should be noted that, above-described embodiment does not limit the present invention in any form, and all employings are equal to replaces or technical scheme that the mode of equivalent transformation obtains, all drops in protection scope of the present invention.
Claims (6)
1. the supervising device based on visual pattern, is characterized in that, comprising:
The video camera that also can focus by programmed control,
The electric platform of the above-mentioned video camera all-direction rotation of variable speed controllable Electric Machine Control being provided with by self,
For controlling above-mentioned video camera and electric platform and receiving and process above-mentioned video camera and the monitoring processor of the data message that electric platform feeds back;
The wire communication that above-mentioned video camera, electric platform and monitoring processor form energy transfer data information is connected or wireless telecommunications connection;
Above-mentioned monitoring processor comprises:
For the human-computer interaction module of data message input, feedback,
For controlling the camera control module of above-mentioned video camera and electric platform,
The module of target detection detecting for the target of view data that above-mentioned video camera is obtained,
Realize the video conversion module of subsequent treatment and Reverse Play for the video conversion that above-mentioned video camera is obtained,
For set up the image calibration module of the corresponding relation of image and actual scene according to camera parameters.
2. a kind of supervising device based on visual pattern according to claim 1, it is characterized in that, above-mentioned electric platform is provided with the angular encoder for video camera rotation information being fed back to above-mentioned monitoring processor, and above-mentioned angular encoder is connected with above-mentioned monitoring processor formation wire communication or wireless telecommunications connect.
3. a kind of supervising device based on visual pattern according to claim 1 and 2, it is characterized in that, above-mentioned video camera is the high definition high-speed camera that resolution is rotated more than 1080 × 1920, within the scope of horizontal 360-degree, vertical 180 degree, and the image visual field of above-mentioned high definition high-speed camera is that long and narrow bearing of trend square and narrow rectangular long limit is consistent with road direction.
4. the method for supervising that utilizes a kind of supervising device based on visual pattern as claimed in claim 1, is characterized in that, comprises the steps:
One, supervising device initialization:
(1) input the initial parameter for carrying out image calibration by above-mentioned human-computer interaction module, the initial parameter of input comprises: the setting height(from bottom) of above-mentioned video camera, actual monitored scope, optical parametric and image parameter in general monitor state video camera of lower time, and above-mentioned initial parameter is kept in monitoring processor;
(2) above-mentioned image calibration module is set up the corresponding relation of image and actual scene according to above-mentioned initial parameter, set the parameter of the realistic objective that needs monitoring, set up with reference to above-mentioned corresponding relation the parameter corresponding in image and assert condition, qualified image object is specific image target;
(3) the above-mentioned video camera of above-mentioned monitoring processor control starts to monitor above-mentioned actual monitored scope, within the scope of above-mentioned actual monitored, draw to have and judge whether in violation of rules and regulations actual surveyed area of target, image range corresponding to above-mentioned actual monitored scope is picture control scope, and the image that above-mentioned actual surveyed area is corresponding is virtual detection region;
Two, the triggering of image object and the selection of mode of operation: occur after above-mentioned specific image target in above-mentioned monitoring processor is found the video single-frame images of its reception, the video having received is detected, whether judgement wherein contains the high-resolution single frames picture that can show specific image target identification feature, and above-mentioned specific image target identification feature refers to can be according to the characteristics of image of the physical features of its corresponding realistic objective of the image identifying of specific image target; If do not contained, adopt the first evidence obtaining mode of operation; If contained, adopt the second evidence obtaining mode of operation;
Three, collect evidence, obtain and preserve image forensic data message: above-mentioned monitoring processor and video camera match and adopt above-mentioned the first evidence obtaining mode of operation or the second evidence obtaining mode of operation processing of collecting evidence, and obtain and preserve image forensic data message.
5. method for supervising according to claim 4, is characterized in that, the detailed process of above-mentioned the first evidence obtaining mode of operation is:
(1) video camera obtains the monitor video of current scene by gathering continuous single frames;
(2) monitoring processor obtains the single frames of a current scene at regular intervals from above-mentioned monitor video, whether utilize background subtraction point-score to detect in the single frames of above-mentioned current scene has specific image target to enter virtual detection region, if specific image target does not enter above-mentioned virtual detection region, above-mentioned monitoring processor repeating step (2); If specific image target has entered above-mentioned virtual detection region, above-mentioned monitoring processor starts timing, and carries out step (3), and this timing time is specific objective acknowledging time;
(3) monitoring processor obtains the single frames of a current scene at regular intervals from above-mentioned monitor video, whether the virtual detection region that utilizes background subtraction point-score to detect in above-mentioned single frames also has specific image target, if no, above-mentioned monitoring processor repeating step (2); If had, and above-mentioned specific objective acknowledging time does not reach predefined threshold value, repeating step (3); If had, and above-mentioned specific objective acknowledging time reaches predefined threshold value, and the above-mentioned specific objective of the above-mentioned camera alignment of above-mentioned monitoring processor control is focused and captures.
6. method for supervising according to claim 4, is characterized in that, the detailed process of above-mentioned the second evidence obtaining mode of operation is:
(1) monitoring processor control video camera is recorded a video to guarded region, and every video of set time camera switching, the video of recording end enters step (2) as pending video;
(2) whether monitoring processor enters above-mentioned virtual detection region to the specific image target in the single frames of above-mentioned pending video and detects, if do not entered, changes single frames and proceeds step (2); If entered, monitoring processor predicts that above-mentioned specific image target arrives the needed time of image forensics position, this time is the pre-evidence obtaining time, above-mentioned image forensics position is the position in the correspondence image of target evidence obtaining position, above-mentioned target evidence obtaining position is that video camera obtains the specific objective position at specific objective place in actual scene when image forensic data message clearly, and this pending video enters as step (3) as evidence obtaining video;
(3) above-mentioned monitoring processor carries out video direction finding and starts timing above-mentioned specific image target, if above-mentioned specific image target is to the direction motion near video camera, above-mentioned monitoring processor still reads the single frames of above-mentioned evidence obtaining video with normal sequence; If to the direction motion away from video camera, above-mentioned monitoring processor starts inverted order and reads the single frames of above-mentioned evidence obtaining video;
(4), when monitoring processor reads the time above-mentioned pre-evidence obtaining time of arrival of above-mentioned evidence obtaining video, preserve evidence obtaining single frames and the above-mentioned evidence obtaining video of current time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210127924.9A CN102693632B (en) | 2012-04-27 | 2012-04-27 | Monitoring device and monitoring method based on visual image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210127924.9A CN102693632B (en) | 2012-04-27 | 2012-04-27 | Monitoring device and monitoring method based on visual image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102693632A CN102693632A (en) | 2012-09-26 |
| CN102693632B true CN102693632B (en) | 2014-06-11 |
Family
ID=46859026
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210127924.9A Expired - Fee Related CN102693632B (en) | 2012-04-27 | 2012-04-27 | Monitoring device and monitoring method based on visual image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102693632B (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104038727B (en) * | 2013-03-05 | 2017-11-03 | 北京计算机技术及应用研究所 | A kind of method that video monitoring system and its video camera are accurately controlled |
| CN103325256A (en) * | 2013-06-29 | 2013-09-25 | 佘若凡 | Binocular video road traffic incident detection system |
| CN103400500B (en) * | 2013-08-09 | 2016-01-20 | 江苏中科易正电子科技有限公司 | Vehicle information data acquisition device and method |
| CN104469227B (en) * | 2013-09-22 | 2017-06-30 | 杭州海康威视系统技术有限公司 | The kinescope method and device of vehicle behavior |
| CN103777954B (en) * | 2014-01-26 | 2017-04-05 | 国网浙江象山县供电公司 | A kind of information processing method and device |
| CN104363412B (en) * | 2014-10-29 | 2018-02-13 | 南京茵默维电子科技有限公司 | The big visual field supervisory systems and monitoring and managing method of elongated zones |
| CN105989639A (en) * | 2015-02-16 | 2016-10-05 | 径卫视觉科技（上海）有限公司 | Life auxiliary type intelligent driving recording system |
| CN106560861A (en) * | 2015-09-30 | 2017-04-12 | 徐贵力 | Intelligent supervision method based on computer vision |
| CN105338317A (en) * | 2015-10-29 | 2016-02-17 | 中国邮政集团公司广州市分公司 | Monitoring method and monitoring device for mailboxes |
| CN105635672A (en) * | 2015-12-26 | 2016-06-01 | 余姚市绿波机械科技有限公司 | Safety reminding system of power transmission line and reminding method thereof |
| CN107306348A (en) * | 2016-04-19 | 2017-10-31 | 中兴通讯股份有限公司 | Video frequency monitoring method, apparatus and system |
| CN105957340B (en) * | 2016-04-27 | 2019-08-13 | 电子科技大学 | A kind of traffic monitoring method based on adaptive virtual coil |
| CN107687932B (en) * | 2016-08-05 | 2020-08-04 | 成都理想境界科技有限公司 | Method and device for detecting delay of head-mounted display equipment |
| CN107293136A (en) * | 2017-06-30 | 2017-10-24 | 中山市程博工业产品设计有限公司 | A kind of Intelligent shutting-down device |
| CN107481291B (en) * | 2017-08-16 | 2020-04-03 | 长安大学 | Traffic monitoring model calibration method and system based on physical coordinates of marked dotted lines |
| CN109561249A (en) * | 2017-09-26 | 2019-04-02 | 北京小米移动软件有限公司 | Adjust the method and device of focal length |
| CN107844772A (en) * | 2017-11-09 | 2018-03-27 | 汕头职业技术学院 | A kind of motor vehicle automatic testing method based on movable object tracking |
| CN108093218A (en) * | 2017-12-25 | 2018-05-29 | 浙江东南建设管理有限公司 | A kind of construction quality video Supervision and system |
| CN109756620B (en) * | 2018-12-15 | 2021-07-23 | 深圳壹账通智能科技有限公司 | Virtual camera identification method based on mobile phone and related equipment |
| CN111010549A (en) * | 2019-03-11 | 2020-04-14 | 南昌工程学院 | Video information heterogeneous computing system and working method thereof |
| CN110274612A (en) * | 2019-06-21 | 2019-09-24 | 大庆安瑞达科技开发有限公司 | A kind of pitching of remote optical turntable and orientation calibration method based on ground remote sensing photomap |
| CN112015231B (en) * | 2020-07-31 | 2021-06-22 | 中标慧安信息技术股份有限公司 | Method and system for processing surveillance video partition |
| CN112418026B (en) * | 2020-11-11 | 2023-06-02 | 宁波职业技术学院 | Vehicle violation detection method, system and device based on video detection |
| CN114359875B (en) * | 2022-03-17 | 2022-05-17 | 成都考拉悠然科技有限公司 | Vehicle effective snapshot method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1440005A (en) * | 2002-12-30 | 2003-09-03 | 北京市振隆科技发展公司 | Electronic V.F. network police monitoring systems |
| EP1916642A1 (en) * | 2006-10-24 | 2008-04-30 | Dambach-Werke GmbH | Traffic detecting device |
| CN101789177A (en) * | 2010-02-05 | 2010-07-28 | 南京航空航天大学 | Device and method for detecting and tracking vehicles crossing and pressing the yellow line and for capturing vehicle information |
| CN101917595A (en) * | 2010-07-01 | 2010-12-15 | 李志恒 | Video monitoring system for automatically marking road section information |
| CN102394012A (en) * | 2011-06-21 | 2012-03-28 | 陆化普 | Monitoring system having preset bit function for vehicle crossing over markings and method thereof |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100812765B1 (en) * | 2006-06-21 | 2008-03-12 | 럭스테크 주식회사 | Apparatus for detect overspeed vehicle |
-
2012
- 2012-04-27 CN CN201210127924.9A patent/CN102693632B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1440005A (en) * | 2002-12-30 | 2003-09-03 | 北京市振隆科技发展公司 | Electronic V.F. network police monitoring systems |
| EP1916642A1 (en) * | 2006-10-24 | 2008-04-30 | Dambach-Werke GmbH | Traffic detecting device |
| CN101789177A (en) * | 2010-02-05 | 2010-07-28 | 南京航空航天大学 | Device and method for detecting and tracking vehicles crossing and pressing the yellow line and for capturing vehicle information |
| CN101917595A (en) * | 2010-07-01 | 2010-12-15 | 李志恒 | Video monitoring system for automatically marking road section information |
| CN102394012A (en) * | 2011-06-21 | 2012-03-28 | 陆化普 | Monitoring system having preset bit function for vehicle crossing over markings and method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102693632A (en) | 2012-09-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102693632B (en) | Monitoring device and monitoring method based on visual image | |
| CN101789177B (en) | Device and method for detecting and tracking vehicles crossing and pressing the yellow line and for capturing vehicle information | |
| CN102110376B (en) | Roadside parking space detection device based on computer vision | |
| Grassi et al. | Parkmaster: An in-vehicle, edge-based video analytics service for detecting open parking spaces in urban environments | |
| CN105809972B (en) | A kind of parking management method and device, system | |
| CN102708685B (en) | Device and method for detecting and snapshotting violation vehicles | |
| CN100557657C (en) | A kind of vehicle checking method based on video image characteristic | |
| CN102622895B (en) | Video-based vehicle speed detecting method | |
| CN103841313A (en) | Pan-tilt camera control method, system and device | |
| US20050237390A1 (en) | Multiple camera system for obtaining high resolution images of objects | |
| KR20200025247A (en) | Managemnet system of parking by moving vehicle tracking | |
| CN106558224B (en) | A kind of traffic intelligent monitoring and managing method based on computer vision | |
| CN102426801A (en) | Automatic parking lot berth management method and system based on visual analysis | |
| KR100820952B1 (en) | Detecting method at automatic police enforcement system of illegal-stopping and parking vehicle using single camera and system thereof | |
| CN101320048A (en) | Multi-charge coupled device image sensor large visual field vehicle speed measuring device arranged in sector form | |
| CN107767687A (en) | Free parking space detection method and system based on binocular stereo vision | |
| CN105303826A (en) | Violating side parking evidence obtaining device and method | |
| CN113850137A (en) | Power transmission line image online monitoring method, system and equipment | |
| CN205230349U (en) | Traffic speed of a motor vehicle detects and snapshot system based on TOF camera | |
| RU120270U1 (en) | PEDESTRIAN CROSSING CONTROL COMPLEX | |
| CN112884803B (en) | Real-time intelligent monitoring target detection method and device based on DSP | |
| CN109815925A (en) | A kind of intelligent infrared road image identifying system | |
| CN201266436Y (en) | Multi-video united detection device for mobile vehicle | |
| CN111225182B (en) | Image acquisition equipment, method and device | |
| CN202404111U (en) | DSP-based video velocity-measuring system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CB03 | Change of inventor or designer information |
Inventor after: Xu Guili Inventor after: Zhang Meixuan Inventor after: Yang Xiaowei Inventor after: Li Xu Inventor after: Lv Wen Inventor before: Xu Guili Inventor before: Yang Xiaowei Inventor before: Li Xu Inventor before: Lv Wen |
|
| COR | Change of bibliographic data | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140611 Termination date: 20170427 |