CN108647365A - A kind of camouflage of target effect evaluation method and system based on artificial interpretation - Google Patents
A kind of camouflage of target effect evaluation method and system based on artificial interpretation Download PDFInfo
- Publication number
- CN108647365A CN108647365A CN201810527666.0A CN201810527666A CN108647365A CN 108647365 A CN108647365 A CN 108647365A CN 201810527666 A CN201810527666 A CN 201810527666A CN 108647365 A CN108647365 A CN 108647365A
- Authority
- CN
- China
- Prior art keywords
- target
- reader
- camouflage
- image
- area
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/14—Digital output to display device ; Cooperation and interconnection of the display device with other functional units
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The present invention provides a kind of camouflage of target effect evaluation method and system based on artificial interpretation, and required best interpretation environment is tested in setting interpretation to this method in systems first, is easier to identify the target after in image pretending to sentence reader.Next, system loads all images to be identified one by one, if sentencing reader identifies camouflaged target in image, the target area identified in image is marked out, system compares the degree of overlapping of the area of the area and camouflaged target of identification division, and stores comparison result;If sentence reader it is unidentified go out target area, store accordingly result information.Final system prompt sentences reader and continues to identify the next camouflage test to be identified for carrying target, and after the completion of the identification of all images to be recognized, system exports discrimination, which is the probability sentenced reader and correctly identify target in image according to the comparison result of storage.Camouflage of target recruitment evaluation is rapidly completed in this system in a manner of human-computer interaction, improves testing efficiency.
Description
Technical field
The invention belongs to computer assisted image processing fields, and in particular to a kind of camouflage of target effect based on artificial interpretation is commented
Estimate method and system.
Background technology
Image interpretation experiment is a kind of mode to camouflage of target effect subjective evaluation, it is by interpretation personnel certain
Rule and under the conditions of the target image after camouflage is observed, the process that camouflage effectiveness is assessed according to score.One
As for, interpretation process choosing carries out indoors, selects a certain number of interpretation personnel, carries out diagnostic rule to them first
Study, on the basis of understanding interpretation flow, observes each image within a certain period of time, provide whether there is or not " it was found that " knot of target
By if it find that just further giving a mark to image detail by the requirement in rule.
Under normal circumstances, it carries out field trial and needs a large amount of manpower and materials, repeatability is poor, and is influenced by external field environment
It is larger, and build emulation platform using computer quickly can carry out recruitment evaluation to the camouflaged target under different scenes.
It is published in by its etc. of going into business in May, 2011《Applied science journal》On article " human eye vision attention mechanism is used for
In camouflage evaluation ", it is proposed that establish camouflage of target Evaluation Model on Effectiveness using human eye vision attention mechanism.This method is first
It extracts the brightness of input picture and towards feature, fusion color difference Butut obtains corresponding characteristic remarkable picture, according to competition and prohibit
Only return mechanisms calculate focus-of-attention, are then based on BP neural network and establish visual focus transfer characteristic and Methods for Target Detection Probability
Correspondence model, to assessing the camouflage effectiveness of target.This method can accurately quantify in same type image
The camouflage effectiveness of target.But this method has some shortcomings:First, this method needs 40 sights after special training
Personnel are examined, manpower expends big;Second is that this method when carrying out a camouflage of target recruitment evaluation, is directed to a type of figure every time
It is less efficient as carrying out camouflage effectiveness assessment.
In conclusion it is a kind of urgently to be resolved hurrily ask to seek a kind of low consumption and efficient camouflage of target effect evaluation method
Topic.
Invention content
The present invention provides a kind of camouflage of target effect evaluation method based on artificial interpretation, includes the following steps:
Step S1:Then all images to be recognized of system loads check or adjust the resolution ratio of display to sentence reader
Images to be recognized is preferably observed, and reader is sentenced according to the prompt of the practical imaging size of equipment and is adjusted to most preferably see by position is observed
Cha Qu finally adjusts display brightness to adapt to sentence the visual effect of reader
Step S2:If sentencing reader identifies target in image, mark out in image that identification division, system compare
The degree of overlapping of the area of identification division and the real area of camouflaged target, and comparison result is stored, the mesh in unidentified image
When mark, unidentified information is stored.
Sentence reader to differentiate the image of load, if identifying camouflaged target in the picture, marks target and scheming
The target area of matching identification and true camouflaged target area are recorded comparison knot by the shared surface area as in, system
Fruit data;If it is unidentified go out target area, record result data be " unidentified ";
Step S3:System guides are sentenced reader and are differentiated one by one to remaining all images to be recognized, when all to be identified
After image completes differentiation process, the discrimination of current camouflaged target is obtained by calculation according to the result data of record in system.
Preferably, step S1 further includes sentencing reader to learn diagnostic rule, and the diagnostic rule is obtained by system, and is sentenced
Read rule includes how operating and completing before the deadline the interpretation experiment of image when sentencing reader and finding target.
Preferably, in step sl, the viewing distance of interpretation personnel is 6 times of height.
Preferably, in step s 2, if sentence reader it is unidentified go out camouflaged target, notify lower of system loads to be identified
Image.
Preferably, in step S2 and S3, if the area of identification division is Simg_1, the area S of camouflaged targetimg_2, then
Work as 2*Simg_2< Simg_1∩Simg_2< Simg_2When/2, system judgement identifies successfully.
The camouflage of target recruitment evaluation system based on artificial interpretation that the present invention also provides a kind of, including presetting module, storage
Module and judgment module, the presetting module connect memory module, the memory module connection judgment module.Presetting module is used for
The best screen resolution of interpretation experiment is automatically determined, prompts to sentence reader's adjustment viewing distance and screen intensity, and prompt interpretation
Person learns diagnostic rule.Memory module sentences reader to the corresponding sentence read result of every image for storing, and is identified simultaneously when sentencing reader
After marking out in image identification division, the degree of overlapping of the area of the area and camouflaged target of identification division is compared, and is stored
Comparison result, when sentence reader it is unidentified go out image in target after, store corresponding result.Judgment module sentences reader for prompting
Continue to identify the next images to be recognized for carrying camouflaged target, after the completion of the identification of all images to be recognized, according to storage
As a result, output discrimination.
Preferably, presetting module obtains by calling system function and adjusts display resolution automatically, and prompts to sentence reader
Adjust viewing distance and screen intensity.
According to camouflage of target effect evaluation method and system provided by the invention based on artificial interpretation, this method is first certainly
The dynamic best screen resolution determined needed for interpretation experiment, prompt sentence reader and adjust viewing distance and screen intensity.Next, sentencing
If reader identifies the camouflaged target in image, identification division is marked out in image, system compares the face of identification division
The degree of overlapping of product and the area of camouflaged target, and store comparison result, if sentence reader it is unidentified go out image in camouflaged target,
Store corresponding information.Final system prompt sentences reader and continues to identify the next images to be recognized for carrying camouflaged target, when all
After the completion of images to be recognized identification, system is according to storage as a result, output discrimination.Mesh is rapidly completed in a manner of human-computer interaction
Camouflage effectiveness assessment is marked, testing efficiency is improved.
After the detailed description of the application embodiment is read in conjunction with the figure, other features and advantages of the application will become more
Add clear.
Description of the drawings
Fig. 1 is the camouflage of target effect evaluation method flow chart based on artificial interpretation that present pre-ferred embodiments provide;
Fig. 2 is the camouflage of target recruitment evaluation system schematic based on artificial interpretation that present pre-ferred embodiments provide.
Specific implementation mode
Below in conjunction with drawings and examples, the invention will be further described.
As shown in Figure 1, the present invention provides a kind of camouflage of target effect evaluation method based on artificial interpretation, including following step
Suddenly.
Step S1:Then all images to be recognized of system loads check or adjust the resolution ratio of display to sentence reader
Images to be recognized is preferably observed, and reader is sentenced according to the prompt of the practical imaging size of equipment and is adjusted to most preferably see by position is observed
Cha Qu finally adjusts display brightness to adapt to sentence the visual effect of reader;
Specifically, the display environment condition used in reference image quality evaluation, indoor observation ambient brightness is 50 banks
De La/square metre room, indoor illumination intensity are 3~6 luxs.Using display or projecting apparatus, images to be recognized is according to its practical point
Resolution 1:1 ratio shows that the viewing distance of interpretation personnel is 6 times of height.In the present embodiment, display physical resolution
According to 4:3 displays, if sentencing reader checks that the resolution ratio of present displays is the optimum resolution of present displays, without adjusting
It is whole;If not the optimum resolution of present displays, then sentencing reader automatically will be current by way of interactive according to prompt
The resolution adjustment of display is optimum resolution.
In this, starting before identifying camouflaged target, sentences reader and learn diagnostic rule, diagnostic rule is obtained by system, including
The interpretation for how operating and completing before the deadline image when sentencing reader's discovery target is tested.
Step S2:Sentence reader to differentiate the image of load, if identifying camouflaged target in the picture, marks out mesh
Surface area shared in the picture is marked, system records the target area of matching identification and true camouflaged target area
Comparison result data;If it is unidentified go out target area, record result data be " unidentified ".
In this step, if sentence reader it is unidentified go out camouflaged target, notify system to be loaded into next images to be recognized.In
This, the residence time of every width images to be recognized is 1 minute, and the wherein interpretation time is 20 seconds, and residue sentences reader in 40 seconds and provides conclusion.
Specially:If sentencing reader finds camouflaged target, outlines target area using mouse and confirm in systems;If interpretation personnel
Target is not found, and selection in systems continues to be loaded into next image.
In this, the mode for comparing the degree of overlapping of the area of the area and camouflaged target of identification division is:If sentencing reader's mark
The area for the identification division known is Simg_1, the area S of camouflaged targetimg_2, then work as 2*Simg_2< Simg_1∩Simg_2< Simg_2/
When 2, system judgement identifies successfully;If other situations, then system judges recognition failures, and comparison result deposit each time is
System.
Step S3:System guides are sentenced reader and are differentiated one by one to remaining all images to be recognized, when all to be identified
After image completes differentiation process, the discrimination of current camouflaged target is obtained by calculation according to the result data of record in system.
Specifically, after interpretation experiment, which can provide discrimination and the interpretation experiment of interpretation experiment automatically
Detailed data.Sentence whether reader can preserve that interpretation is tested as a result, preservation result data can be by system according to demand
Prompt is realized in an interactive way.
As shown in Fig. 2, the present invention also provides a kind of camouflage of target recruitment evaluation system based on artificial interpretation, including it is presetting
Module 101, memory module 102 and judgment module 103, the presetting module 101 connect memory module 102, the memory module
102 connection judgment modules 103.Presetting module 101 tests best screen resolution for that will automatically determine interpretation, prompts to sentence reader
Adjust the study of viewing distance, screen intensity and interpretation experimental rules.Memory module 102 sentences reader to every image for storing
Corresponding sentence read result, after sentencing reader and identifying and mark out in image identification division, compare the area of identification division with
The degree of overlapping of the area of camouflaged target, and comparison result is stored, if it is the target found in image to sentence reader, storage is corresponding
Information.Judgment module 103 is used to prompt to sentence reader and continues to identify next the images to be recognized with camouflaged target, when being needed
After the completion of identifying image recognition, according to storage as a result, output discrimination.
Specifically, presetting module 101 obtains simultaneously adjust automatically display resolution by calling system function automatically, and prompts
Sentence reader and adjusts viewing distance and screen intensity.By calling system function, the display resolution of hardware is obtained, checks current point
Resolution whether be hardware supported optimum resolution.If the resolution ratio of present displays is not the resolution ratio of best image, can
It is automatically best image resolution ratio by the resolution adjustment of present displays.
In conclusion the camouflage of target effect evaluation method based on artificial interpretation that preferred embodiment provides through the invention
And system, this method system first will automatically determine interpretation and test best screen resolution, prompt to sentence reader's adjustment viewing distance
And screen intensity.Next, if sentencing reader identifies camouflaged target, mark out in image that identification division, system compare
The degree of overlapping of the area of identification division and the area of camouflaged target, and comparison result is stored, when the target in unidentified image,
Store unidentified information.Final system prompt sentences reader and continues to identify the next images to be recognized for carrying camouflaged target, works as institute
After the completion of having image recognition to be identified, system is according to storage as a result, output discrimination.It is rapidly completed in a manner of human-computer interaction
Camouflage of target recruitment evaluation, improves testing efficiency.
The above is only the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications should also regard
For protection scope of the present invention.
Claims (7)
1. a kind of camouflage of target effect evaluation method based on artificial interpretation, which is characterized in that include the following steps:
All images to be recognized of S1, system loads, the resolution ratio for then checking or adjusting display are preferably seen to sentence reader
Images to be recognized is examined, and reader is sentenced according to the prompt of the practical imaging size of equipment and is adjusted to optimal viewing area by position is observed, most
Adjustment display brightness is to adapt to sentence the visual effect of reader afterwards;
S2, sentence reader the image of load is differentiated, if identifying camouflaged target in the picture, mark target in image
In shared surface area, the target area of matching identification and true camouflaged target area record comparison result by system
Data;If it is unidentified go out target area, record result data be " unidentified ";
S3, System guides are sentenced reader and are differentiated one by one to remaining all images to be recognized, when all images to be recognized are completed
After differentiation process, the discrimination of current camouflaged target is obtained by calculation according to the result data of record in system.
2. the camouflage of target effect evaluation method according to claim 1 based on artificial interpretation, which is characterized in that step S1
Further include sentencing reader to learn diagnostic rule, the diagnostic rule obtained by system, including when sentencing reader and finding target how
Operation and the interpretation experiment for completing image before the deadline.
3. the camouflage of target effect evaluation method according to claim 1 based on artificial interpretation, which is characterized in that in step
In S1, the viewing distance of interpretation personnel is 6 times of height.
4. the camouflage of target effect evaluation method according to claim 1 based on artificial interpretation, which is characterized in that in step
In S2, if sentence reader it is unidentified go out camouflaged target, notify lower images to be recognized of system loads.
5. the camouflage of target effect evaluation method according to claim 1 based on artificial interpretation, which is characterized in that in step
In S2 and S3, if the area of identification division is Simg_1, the area S of camouflaged targetimg_2, then work as Simg_2/ 2 < Simg_1∩Simg_2
< 2*Simg_2When, system judgement identifies successfully.
6. a kind of camouflage of target recruitment evaluation system based on artificial interpretation, which is characterized in that including presetting module, memory module
And judgment module, the presetting module connection memory module, the memory module connection judgment module, wherein:
Presetting module prompts to sentence reader's adjustment viewing distance for automatically determining the best screen resolution needed for interpretation experiment
And screen intensity, and prompt to sentence reader's study diagnostic rule;
Memory module sentences reader to the corresponding sentence read result of every image for storing, and identifies sentencing reader and marks out image
In after identification division, compare the degree of overlapping of the area of the area and camouflaged target of identification division, and store comparison result, when
When target in unidentified image, accordingly result information is stored;
Judgment module continues to identify the next images to be recognized for carrying camouflaged target for prompting to sentence reader, when needing to be known
After the completion of other image recognition, according to system storage as a result, output discrimination.
7. the camouflage of target recruitment evaluation system according to claim 6 based on artificial interpretation, which is characterized in that pre- mode transfer
Block obtains by calling system function and adjusts display resolution automatically, and reader's adjustment viewing distance is sentenced in prompt and screen is bright
Degree.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810527666.0A CN108647365A (en) | 2018-05-29 | 2018-05-29 | A kind of camouflage of target effect evaluation method and system based on artificial interpretation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810527666.0A CN108647365A (en) | 2018-05-29 | 2018-05-29 | A kind of camouflage of target effect evaluation method and system based on artificial interpretation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108647365A true CN108647365A (en) | 2018-10-12 |
Family
ID=63758475
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810527666.0A Pending CN108647365A (en) | 2018-05-29 | 2018-05-29 | A kind of camouflage of target effect evaluation method and system based on artificial interpretation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108647365A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109683416A (en) * | 2019-01-28 | 2019-04-26 | 青岛九维华盾科技研究院有限公司 | A kind of double-colored adjustable electrochromism thin-film device |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101162489A (en) * | 2007-11-26 | 2008-04-16 | 唐荣华 | Personate photograph producing method used for keep input secret and method of use thereof |
CN102622588A (en) * | 2012-03-08 | 2012-08-01 | 无锡数字奥森科技有限公司 | Dual-certification face anti-counterfeit method and device |
CN103440389A (en) * | 2013-08-31 | 2013-12-11 | 西安电子科技大学 | Infrared camouflage effect simulation system and infrared camouflage effect evaluation method based on full link |
CN104835104A (en) * | 2014-02-07 | 2015-08-12 | 湖南第一师范学院 | Digital image active camouflage scheme based on fractal |
CN106295694A (en) * | 2016-08-05 | 2017-01-04 | 浙江工业大学 | Face recognition method for iterative re-constrained group sparse representation classification |
CN107729830A (en) * | 2017-10-09 | 2018-02-23 | 西安工业大学 | Camouflage painting effect detection computational methods based on background characteristics |
-
2018
- 2018-05-29 CN CN201810527666.0A patent/CN108647365A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101162489A (en) * | 2007-11-26 | 2008-04-16 | 唐荣华 | Personate photograph producing method used for keep input secret and method of use thereof |
CN102622588A (en) * | 2012-03-08 | 2012-08-01 | 无锡数字奥森科技有限公司 | Dual-certification face anti-counterfeit method and device |
CN103440389A (en) * | 2013-08-31 | 2013-12-11 | 西安电子科技大学 | Infrared camouflage effect simulation system and infrared camouflage effect evaluation method based on full link |
CN104835104A (en) * | 2014-02-07 | 2015-08-12 | 湖南第一师范学院 | Digital image active camouflage scheme based on fractal |
CN106295694A (en) * | 2016-08-05 | 2017-01-04 | 浙江工业大学 | Face recognition method for iterative re-constrained group sparse representation classification |
CN107729830A (en) * | 2017-10-09 | 2018-02-23 | 西安工业大学 | Camouflage painting effect detection computational methods based on background characteristics |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109683416A (en) * | 2019-01-28 | 2019-04-26 | 青岛九维华盾科技研究院有限公司 | A kind of double-colored adjustable electrochromism thin-film device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106250870B (en) | A kind of pedestrian's recognition methods again of joint part and global similarity measurement study | |
CN107609493B (en) | Method and device for optimizing human face image quality evaluation model | |
CN105825511B (en) | A kind of picture background clarity detection method based on deep learning | |
CN104966070B (en) | Biopsy method and device based on recognition of face | |
CN108492272A (en) | Cardiovascular vulnerable plaque recognition methods based on attention model and multitask neural network and system | |
CN103632158B (en) | Forest fire prevention monitor method and forest fire prevention monitor system | |
CN110348312A (en) | A kind of area video human action behavior real-time identification method | |
CN106909220A (en) | A kind of sight line exchange method suitable for touch-control | |
CN110349145A (en) | Defect inspection method, device, electronic equipment and storage medium | |
CN110837784A (en) | Examination room peeping cheating detection system based on human head characteristics | |
CN106169071A (en) | A kind of Work attendance method based on dynamic human face and chest card recognition and system | |
CN106250825A (en) | A kind of at the medical insurance adaptive face identification system of applications fields scape | |
CN109805943A (en) | A kind of intelligent evaluating system and method based on micro- Expression Recognition | |
CN107845300A (en) | A kind of safe examination system monitored in real time | |
CN115273180B (en) | Online examination invigilating method based on random forest | |
CN109636846A (en) | Object localization method based on circulation attention convolutional neural networks | |
CN111144749A (en) | Intelligent labeling crowdsourcing platform for power grid images and working method | |
CN109997199A (en) | Tuberculosis inspection method based on deep learning | |
CN107818707A (en) | A kind of examination system of automatic problem building | |
CN114550270A (en) | Micro-expression identification method based on double-attention machine system | |
CN108647365A (en) | A kind of camouflage of target effect evaluation method and system based on artificial interpretation | |
TW202219494A (en) | A defect detection method and a defect detection device | |
CN110765953A (en) | Multimedia teaching teacher sign-in monitoring method and system | |
CN110427833A (en) | A kind of gesture tracking method, system and storage medium | |
CN113627393B (en) | Commodity identification method based on dual neural network and intelligent vending system |
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 |