CN101321270A - Monitoring system and method for real-time optimized image - Google Patents
Monitoring system and method for real-time optimized image Download PDFInfo
- Publication number
- CN101321270A CN101321270A CNA2008100317868A CN200810031786A CN101321270A CN 101321270 A CN101321270 A CN 101321270A CN A2008100317868 A CNA2008100317868 A CN A2008100317868A CN 200810031786 A CN200810031786 A CN 200810031786A CN 101321270 A CN101321270 A CN 101321270A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- real
- supervising
- time
- 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.)
- Granted
Links
Images
Abstract
The invention provides a monitoring system of real time optimizing image and method thereof, including the steps of: pre-processing the collected monitoring images by various processing modes, selecting optimal processing mode optimally or optimal processing mode predetermined according to image quality, then outputting and displaying image processed by the optimal processing mode. The invention further detects and traces position of a target in real time and outputs and displays the target region in an amplified manner. The invention can improve quality of the originally input image in real time, improve display effect of image and amplify the target region with obvious target details.
Description
Technical field
The present invention relates to fields such as Digital Image Processing, photogrammetry and computer vision, further be meant by the original digital image in input (commander) the monitoring Room is done various preliminary treatment, improve the display effect of image, and by detecting tracking target, the amplification target zone is to highlight the system and method for target detail.
Background technology
An important function in (commander) monitoring Room is by the input picture that receives outside picture pick-up device the incident outside the Room to be observed and analyzed, and is beneficial to make in time and judges and decision-making.Therefore, guarantee that (commander) monitoring Room is imported and the formedness of the picture quality of demonstration is a very important problem.
Much under the situations, the image that obtains from outside picture pick-up device just directly is transferred to the output demonstration of (commander) monitoring Room without handling, as shown in Figure 1.In the application process of reality, because the influence of image-forming conditions such as illumination, background, the quality of the image of taking often is difficult to be guaranteed in real time, for example contrast is too poor, image cross bright or dark excessively, target is less etc., so just, reduced the Real Time Observation ability of (commander) monitoring Room to target, be unfavorable for making right judgement and decision-making, influenced (commander) monitoring Room operating efficiency.In some cases, some systems also can handle output more earlier to the image of input, but present this processing mode generally is a single-mode, can not obtain the optimal image effect at any time, and can not adapt to the variation of imaging circumstances.
For example, most widely used tracking measurement equipment---the digital theodolite in the target range, it can realize the Real Time Observation of target and analysis, but its resulting digital picture just is directly inputted to the monitoring Room and shows without handling basically.Be subjected to the influence of target range image-forming condition, the quality of these images is many times poor, and is too little etc. as smudgy, poor contrast, target, is unfavorable for Real Time Observation and analysis to target.The rig camera defendance that tightens security all is equipped with in for example a lot of local (as bank, school etc.) again, but rig camera generally is difficult to adapt to the influence that image-forming condition changes (as the variation of illumination), and image quality is difficult to assurance, is unfavorable for security monitoring.
Summary of the invention
The objective of the invention is, the present situation that is difficult to guarantee at the picture quality in present input (commander) the monitoring Room, supervisory control system and method that one cover improves quality of input image, improves the Real Time Observation ability are proposed, making its output image effect that can guarantee (commander) monitoring Room substantially is in following getable optimized image effect of prevailing condition, simultaneously can amplification target, highlight target detail, be beneficial to Real Time Observation and analysis.
In order to reach the foregoing invention purpose, the technical solution used in the present invention is: a kind of supervisory control system of real-time optimized image, comprise camera head, display unit, pretreatment unit and controller, wherein the signal output part of camera head is connected with the signal input part of pretreatment unit, the signal output part access controller input of pretreatment unit, controller output end inserts the signal input part of display unit at last.
Accordingly, the present invention also provides a kind of method for supervising of real-time optimized image, and it comprises the steps:
1) image in acquisition monitoring zone;
2) from the image of gathering, extract a width of cloth and carry out multiple preliminary treatment, obtain comparing with former figure behind the multiple treatment effect figure;
3) select wherein a kind of pattern or select various modes to be combined into composite mode according to the image pretreating effect as optimum tupe;
4) image of subsequent acquisition all carries out exporting demonstration more in real time after the preliminary treatment according to above-mentioned selected optimum tupe;
5) when detecting imaging circumstances and change, repeating step 2, reselect optimum tupe after, continue monitoring.Or setting refresh time that can be artificial, repeating step 2.
Further, when needs amplify demonstration to the target area, carry out following step before output shows in the step 4: obtain the target location by detecting tracking technique, export in real time with the target area amplification and with enlarged image then.
As preferred version, in step 2, when the image of collection is big, carry out preliminary treatment again after dwindling processing earlier;
In the step 4, the image that output shows can show the switching of full images and target enlarged drawing, also the two stack can be shown.
Below the course of work of the present invention and principle are further specified:
1) image in utilization camera head acquisition monitoring zone, and with the image importing processing unit of gathering;
2) carrying out the image multi-mode handles and the multi-mode demonstration
This step is made multiple basic handling with multiple different image processing method to a certain width of cloth digital picture in the sequence image that collects, and for example contrast strengthens, level and smooth, sharpening, Gao Guang, histogram equalization, pseudo color strengthen, brightness is adjusted or the like.The digital picture of input can be colored, also can be gray scale.For bigger digital picture,, do various preliminary treatment again after image can being dwindled earlier for the speed that guarantees to handle.
After multiple basic handling is finished, show simultaneously with the input picture original graph with through the design sketch after the various processing, for you to choose.
3) select the image processing pattern
Preliminary treatment is chosen tupe according to qualifications according to the quality of image processing effect after finishing.
Requiring the full-automatic occasion of using, automatically every kind of pretreated picture quality is estimated according to the image quality evaluation function, and selected the image processing pattern according to qualifications or just appoint which kind of tupe of use in advance.In the occasion that allows manual intervention, can select the best tupe of effect according to subjective sensation.Be subjected to parameter influence bigger on the image processing pattern algorithm that has, also can carry out manual adjustment, so that treatment effect reaches best to the parameter of this algorithm.
Whether single according to the image processing pattern of selecting, can be divided into monotype selection and composite mode and select.Monotype is selected to be meant and select the final tupe of using of a kind of direct conduct from multiple default tupe; Composite mode is selected to be meant from default tupe and is selected various modes, this various modes is combined sequentially use the tupe of using as final.
4) by institute's lectotype image is handled
After the image processing pattern is selected, after the image of subsequent acquisition all carries out preliminary treatment with this processing pattern earlier, output in real time again.In the process of observing, when the selection of image processing pattern changed, the image of output also had corresponding variation.In addition, can be on the image of output some information of stack, as the speed of target, attitude etc.
5) processing and amplifying of target area
Need amplify the target area under the situation of demonstration, obtain the target location by target being detected follow the tracks of earlier, export in real time with the target area amplification and with enlarged image according to the target location then.
Require the occasion of full automatic treatment, adopting the motion target detection tracking technique that target is detected tracking automatically.Can adopt automatic testing method comprise difference image method, optical flow method, based on the method for active contour model, based on the method for statistical model, based on the method for higher order statistical, based on the method for small echo or based on neural network method or the like.
Occasion allowing manual intervention can manually be provided with trace point on initial image.
Detect target or be provided with after the trace point, the utilization tracking technique carries out real-time tracking to target, to determine the position of target on every two field picture.Adoptable track algorithm comprises gray scale correlation method, least square method, average drifting method or particle filter method or the like.
Goal-orientation is selected a suitably zone of size automatically in image, then this zone amplified to highlight target detail, if original image has passed through preliminary treatment, and can be with directly output in real time of enlarged image; If original image does not pass through preliminary treatment, can carry out exporting in real time again after the preliminary treatment to enlarged image, be beneficial to observation and analysis to target.
6) will on display unit, show through pretreated full images or target area enlarged image by controller.The wherein changeable demonstration of full images and target enlarged drawing also can show the two stack.
7) in the process of observing, various image preliminary treatment results can refresh automatically, with the contingent variations such as middle illumination, background that conform, for the selection and the change of pattern provides foundation.When environment changed, the image processing pattern of the best of selecting last time was not necessarily best concerning present environment, so be necessary the tupe that upgrades in time, makes image output effect at any time can both reach best.Refresh interval can artificially be set, and also can judge according to the variation size of the adjacent a little interframe variances of original image, brightness etc. whether imaging circumstances changes automatically, thereby decision refreshes constantly.
As known from the above, the invention provides a kind of supervisory control system and method thereof of image real-time optimization, it can improve the quality of original input picture in real time, improves the effect of image, and scalable target area, highlights target detail.Require the occasion of full automatic treatment, using the evaluation of picture appraisal function also to select tupe according to qualifications or just appoint which kind of tupe of use in advance through the picture quality after the various mode treatment respectively.Simultaneously, in the time need amplifying demonstration, detect tracking target earlier automatically, also output in real time of amplification target zone then to the target in the image.Allowing the occasion of manual intervention, but artificial selection image processing pattern is subjected to parameter value to influence bigger Processing Algorithm to some, but the manual adjustment parameter so that reaching best, effect and also can manually be provided with trace point on target, target is followed the tracks of.Also can increase advanced image compression algorithm in the system, keep accelerating transfer speed of data under the high-quality prerequisite of image.
Description of drawings
Fig. 1 is the structural representation of supervisory control system in the prior art;
Fig. 2 is the structural representation of supervisory control system of the present invention;
Fig. 3 is the FB(flow block) of method for supervising of the present invention;
In the above-mentioned accompanying drawing:
1-camera head 2-image processing apparatus 3-controller 4-display unit
The input unit of 5-pattern process computer main frame 6-pattern process computer
The display unit of 7-pattern process computer
Embodiment
2-3 is observed example with video camera to the scene of launching a guided missile on the sea realization technology of the present invention and process thereof is described in conjunction with the accompanying drawings.
As shown in Figure 3, its method for supervising specifically comprises the steps:
1) with theodolite with the frame frequency acquisition monitoring area images of 25 frame/seconds, obtain comprising the sequence image of missile target, and in the main frame with the pattern process computer of the image introducing equipment front end of collection, as Fig. 2.
2) carrying out multi-mode processing and multi-mode in the main frame of pattern process computer shows.From the image of gathering, extract a width of cloth original image and do various preliminary treatment, then preliminary treatment result and former figure are simultaneously displayed on the display unit 7 of this pattern process computer, for you to choose.Present embodiment adopts multiple image Preprocessing Algorithm, comprises that contrast of the prior art strengthens, level and smooth, sharpening, Gao Guang, histogram equalization, pseudo color strengthen, brightness is adjusted or the like.
For the image Preprocessing Algorithm, be that example describes with the contrast enhancing: the contrast enhancing is the gray scale by certain regular each pixel of pointwise modification input picture, thereby changes the dynamic range of gradation of image.It can make the gray scale dynamic range expansion, also can make its compression, or compresses in certain dynamic area and expanding in the interval in addition.Suitably choose picture point greyscale transformation rule, can make the raising of the contrast between the interested image-region of people in the image.
If input picture be f (x, y), the image after the processing be g (x, y), then contrast strengthen can be expressed as g (x, y)=T[f (x, y)], wherein T represents input picture and output image corresponding points grey scale mapping relation.
3) carrying out image display mode in image processing apparatus selects.The full-automatic enhancement process that present embodiment adopts, utilization picture appraisal function is estimated automatically to various preliminary treatment results in pattern process computer, selects the best tupe of effect, and the optimum tupe of selecting at last is a contrast enhancement patterns.Also can go out optimum tupe by artificial selection by the input unit of pattern process computer.
4) image processing apparatus is handled image by selected optimization model.Be that follow-up digital picture is all monitored control computer in the Room with carrying out being input to after the preliminary treatment (commander) by above-mentioned selected contrast enhancement patterns earlier again.When needs amplify demonstration to the target area, forward 5 to), otherwise, forward 6 to).
5) in image processing apparatus, carry out the target area amplifieroperation: at first automatically target is detected tracking, obtain the target location, then the target area is amplified and enlarged image outputed to control computer in (commander) monitoring Room.
Adopt the detection of optical flow method realization in the present embodiment to guided missile.The basic principle that optical flow method detects moving target is: give a velocity for each pixel in the image, this has just formed an image motion field, in a particular moment of motion, point on the image and the point on the three-dimensional body are corresponding one by one, according to the velocity feature of each pixel, can carry out dynamic analysis to image.If there is not moving target in the image, then light stream vector is a continually varying in the entire image zone, when there were relative motion in object and image background, the formed velocity of moving object was inevitable different with the domain background velocity, thereby detected the position of moving object.
Utilization gray scale related algorithm is realized the tracking to guided missile among the embodiment.Wherein the basic principle of related algorithm tracking is: compare the coefficient correlation of two images in each position, the point of coefficient correlation maximum is exactly a best match position.
6) will control computer and be connected, will on LCDs, show through pretreated full images or target area enlarged image by the control computer with the display unit-LCDs in (commander) monitoring Room.The wherein changeable demonstration of full images and target enlarged drawing also can show the two stack.
7) refresh automatically once every 10 seconds by clock module control pretreating effect figure, promptly return 2), repeat aforesaid operations, to adapt to the variation of imaging circumstances.
Claims (8)
1, a kind of supervisory control system of real-time optimized image, comprise camera head and display unit, it is characterized in that, also comprise pretreatment unit and controller, wherein the signal output part of camera head is connected with the signal input part of pretreatment unit, the signal output part access controller input of pretreatment unit, controller output end inserts the signal input part of display unit at last.
2, a kind of method for supervising of real-time optimized image is characterized in that, comprises the steps:
1) image in acquisition monitoring zone;
2) from the image of gathering, extract a width of cloth and carry out multiple preliminary treatment, obtain comparing with former figure behind the multiple treatment effect figure;
3) select wherein a kind of pattern or select various modes to be combined into composite mode according to the image pretreating effect as optimum tupe;
4) image of subsequent acquisition all carries out exporting demonstration more in real time after the preliminary treatment according to selected tupe;
5) when detecting imaging circumstances and change, repeating step 2, reselect optimum tupe after, continue monitoring.
3, according to the method for supervising of the described real-time optimized image of claim 2, it is characterized in that, when needs amplify demonstration to the target area, carry out following step before output shows in the step 4: obtain the target location by detecting tracking technique, export in real time with the target area amplification and with enlarged image then.
4, according to the method for supervising of the described real-time optimized image of claim 2, it is characterized in that, in step 2, when the image of collection is big, carry out preliminary treatment again after dwindling processing earlier.
According to the method for supervising of the described real-time optimized image of claim 2, it is characterized in that 5, the image that output shows shows the switching of full images and target enlarged drawing, or the two stack is shown.
6, according to the method for supervising of the described real-time optimized image of claim 2, it is characterized in that, also can be in the step 5 according to the refresh time interval of artificial setting, repeating step 2 is reselected optimum tupe.
According to the method for supervising of the described real-time optimized image of claim 6, it is characterized in that 7, the refresh time of setting is spaced apart 10 seconds.
8, according to the method for supervising of the described real-time optimized image of claim 2, it is characterized in that the concrete grammar that is detected as in the step 5 as environment change is: judge according to the adjacent a little interframe variances of original image of camera collection, the variation size of brightness whether imaging circumstances changes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100317868A CN101321270B (en) | 2008-07-16 | 2008-07-16 | Monitoring system and method for real-time optimized image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008100317868A CN101321270B (en) | 2008-07-16 | 2008-07-16 | Monitoring system and method for real-time optimized image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101321270A true CN101321270A (en) | 2008-12-10 |
| CN101321270B CN101321270B (en) | 2011-06-29 |
Family
ID=40181055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008100317868A Active CN101321270B (en) | 2008-07-16 | 2008-07-16 | Monitoring system and method for real-time optimized image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101321270B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102724386A (en) * | 2012-05-18 | 2012-10-10 | 浙江捷尚视觉科技有限公司 | Real-time video intelligent system and processing method thereof |
| CN105589458A (en) * | 2015-05-19 | 2016-05-18 | 中国人民解放军国防科学技术大学 | Guidance parameter display method and device |
| CN105589460A (en) * | 2015-05-19 | 2016-05-18 | 中国人民解放军国防科学技术大学 | Method and device for controlling mobile vehicle body |
| CN103079047B (en) * | 2012-12-25 | 2016-07-20 | 华为技术有限公司 | A kind of method of parameter adjustment and terminal |
| CN106851105A (en) * | 2017-03-08 | 2017-06-13 | 吕盼稂 | It is a kind of to record the method and apparatus that high-resolution moves facial image |
| CN107248262A (en) * | 2017-06-07 | 2017-10-13 | 上海储翔信息科技有限公司 | Driver management platform for vehicle in use |
| CN107645633A (en) * | 2016-07-21 | 2018-01-30 | 三菱电机大楼技术服务株式会社 | Image processing apparatus |
| CN109118434A (en) * | 2017-06-26 | 2019-01-01 | 南京东大智能化系统有限公司 | A kind of image pre-processing method |
| WO2020124374A1 (en) * | 2018-12-18 | 2020-06-25 | 深圳市大疆创新科技有限公司 | Image processing method, terminal device and storage medium |
| WO2022063321A1 (en) * | 2020-09-28 | 2022-03-31 | 影石创新科技股份有限公司 | Image processing method and apparatus, device and storage medium |
-
2008
- 2008-07-16 CN CN2008100317868A patent/CN101321270B/en active Active
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102724386A (en) * | 2012-05-18 | 2012-10-10 | 浙江捷尚视觉科技有限公司 | Real-time video intelligent system and processing method thereof |
| CN103079047B (en) * | 2012-12-25 | 2016-07-20 | 华为技术有限公司 | A kind of method of parameter adjustment and terminal |
| CN105589458A (en) * | 2015-05-19 | 2016-05-18 | 中国人民解放军国防科学技术大学 | Guidance parameter display method and device |
| CN105589460A (en) * | 2015-05-19 | 2016-05-18 | 中国人民解放军国防科学技术大学 | Method and device for controlling mobile vehicle body |
| CN105589460B (en) * | 2015-05-19 | 2019-07-12 | 中国人民解放军国防科学技术大学 | A kind of method and device of control movement car body |
| CN107645633A (en) * | 2016-07-21 | 2018-01-30 | 三菱电机大楼技术服务株式会社 | Image processing apparatus |
| CN106851105A (en) * | 2017-03-08 | 2017-06-13 | 吕盼稂 | It is a kind of to record the method and apparatus that high-resolution moves facial image |
| CN107248262A (en) * | 2017-06-07 | 2017-10-13 | 上海储翔信息科技有限公司 | Driver management platform for vehicle in use |
| CN109118434A (en) * | 2017-06-26 | 2019-01-01 | 南京东大智能化系统有限公司 | A kind of image pre-processing method |
| WO2020124374A1 (en) * | 2018-12-18 | 2020-06-25 | 深圳市大疆创新科技有限公司 | Image processing method, terminal device and storage medium |
| WO2022063321A1 (en) * | 2020-09-28 | 2022-03-31 | 影石创新科技股份有限公司 | Image processing method and apparatus, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101321270B (en) | 2011-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101321270B (en) | Monitoring system and method for real-time optimized image | |
| CN102221559B (en) | Online automatic detection method of fabric defects based on machine vision and device thereof | |
| CN101828201B (en) | Image processing device and method, and learning device, method | |
| CN109902633A (en) | Accident detection method and device based on the camera supervised video of fixed bit | |
| US7982774B2 (en) | Image processing apparatus and image processing method | |
| CN111242025B (en) | Real-time action monitoring method based on YOLO | |
| CN103093458B (en) | The detection method of key frame and device | |
| CN105828065B (en) | A kind of video pictures overexposure detection method and device | |
| CN104700430A (en) | Method for detecting movement of airborne displays | |
| CN113160123B (en) | Leather defect detection method, system and device based on YOLOv5 | |
| CN104573704A (en) | Eye part detection apparatus and method | |
| CN111339832A (en) | Method and device for detecting face synthetic image | |
| CN111401246A (en) | Smoke concentration detection method, device, equipment and storage medium | |
| CN111161160B (en) | Foggy weather obstacle detection method and device, electronic equipment and storage medium | |
| CN104700405A (en) | Foreground detection method and system | |
| CN116311079A (en) | Civil security engineering monitoring method based on computer vision | |
| CN112419261A (en) | Visual acquisition method and device with abnormal point removing function | |
| KR101077312B1 (en) | Humman detection appartus using Haar-like fearture and method thereof | |
| CN112488165A (en) | Infrared pedestrian identification method and system based on deep learning model | |
| CN107944424A (en) | Front end human image collecting and Multi-angle human are distributed as comparison method | |
| CN110324605B (en) | Intelligent naked eye 3D display system based on neural network | |
| CN103473753A (en) | Target detection method based on multi-scale wavelet threshold denoising | |
| CN105844671A (en) | Rapid background subtraction method under changing illumination conditions | |
| CN115862128A (en) | Human body skeleton-based customer abnormal behavior identification method | |
| CN101447083B (en) | Beaconing-free vision measuring-technique for moving target based on time-space correlative characteristics |
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 |