CN105139404B

CN105139404B - A kind of the license camera and shooting quality detection method of detectable shooting quality

Info

Publication number: CN105139404B
Application number: CN201510553102.0A
Authority: CN
Inventors: 不公告发明人
Original assignee: GUANGZHOU XINGFU NETWORK TECHNOLOGY Co Ltd
Current assignee: GUANGZHOU XINGFU NETWORK TECHNOLOGY Co Ltd
Priority date: 2015-08-31
Filing date: 2015-08-31
Publication date: 2018-12-21
Anticipated expiration: 2035-08-31
Also published as: CN105139404A

Abstract

The invention discloses a kind of license camera of detectable shooting quality and shooting quality detection method, the license camera includes data obtaining module, quality detection module, grading module, quality overall score computing module and shooting suggestion module；The described method includes: obtaining file attribute information, portrait characteristic information, background information, luminance information, chrominance information and the sharpness information of certificate photograph；According to the file attribute information of acquisition, portrait characteristic information, background information, luminance information, chrominance information and sharpness information, quality testing item by item is carried out to every dimensions；It scores the quality measurements of every dimensions, obtains the score value of every dimensions；According to the score value and scoring weight of every dimensions, the quality overall score of certificate photograph is calculated；According to the quality overall score of certificate photograph, the shooting suggestion of profession is provided.Present invention can assure that the requirement of certificate photograph relevant criterion can be met by the certificate photograph of detection.

Description

A kind of the license camera and shooting quality detection method of detectable shooting quality

Technical field

The present invention relates to a kind of license camera of detectable shooting quality and shooting quality detection methods, belong to certificate photograph Shooting and processing technology field.

Background technique

Certificate photograph refer to as identity card, passport, Hong Kong pass, Macao's pass, the Taiwan pass, exit permit, The photo that the legal certificate making such as residence permit, social security card uses has the requirement of its standard, such as:

No.2 residence card digital photo technical standard: " GA 461-2004 ID cards digital photo Technical requirements "；

Motor vehicle driving license digital photo technical standard: " GA482-2012 People's Republic of China (PRC) motor vehicle driving license Part "；

Exit and entry certificates digital photo technical standard: " GA/T 1180-2014 exit and entry certificates digital photo technology is wanted It asks ".

The project of these standard requirements is numerous and fine crushing, and people are generally difficult to that these standards are understood completely, and execution is got up Even more very difficult, shooting effect is even more to vary with each individual, and photographic quality is irregular.However, at present all only by the people of profession Member manually checks on to picture quality, can be automatically performed the quality inspection to these photos there has been no a software on the market It surveys, therefore, ordinary user but has no way of learning after intelligent terminal photographs photo whether the photo come out captured by it meets card The execution standard of part photo.

Summary of the invention

The purpose of the present invention is to solve the defects of the above-mentioned prior art, provide a kind of card of detectable shooting quality Camera, the license camera carry out quality testing item by item to every dimensions by the information of certificate photograph after acquisition shooting, Ensure to meet by the certificate photograph of detection the requirement of certificate photograph relevant criterion.

Another object of the present invention is to provide a kind of shooting quality detection methods.

The purpose of the present invention can be reached by adopting the following technical scheme that:

A kind of license camera of detectable shooting quality, the license camera include:

Data obtaining module obtains file attribute information, portrait feature letter for the certificate photograph according to shooting imaging Breath, background information, luminance information, chrominance information and sharpness information；

Quality detection module, for being believed according to the file attribute information of acquisition, portrait characteristic information, background information, brightness Breath, chrominance information and sharpness information carry out quality testing item by item to every dimensions；Wherein, every dimensions are root According to the examination criteria of government's public security industry standard formulation；

Grading module, for the default score value according to every dimensions, to the quality measurements of every dimensions It scores, obtains the score value of every dimensions；

Quality overall score computing module calculates certificate for the score value and scoring weight according to every dimensions The quality overall score of photo；

Suggestion module is shot, for the quality overall score according to certificate photograph, provides the shooting suggestion of profession.

Further, the data obtaining module includes:

File attribute information acquiring unit, for obtaining the file attribute information of certificate photograph；Wherein, the file attribute Information includes shooting time, colored digit, file suffixes, file size and image pixel；

Portrait characteristic acquisition unit, for obtaining the portrait characteristic information of certificate photograph；Wherein, the portrait feature Information includes eyes, nose, ear, mouth, facial contour, trunk profile, portrait profile, the crown and chin characteristic；

Background information acquiring unit, for when background is not replaced, certificate photograph to be converted to Lab color space image, To Lab color space image, each pixel of foreground area is read, the mixed Gauss model of prospect coloration is established；To background area Each pixel in domain obtains the histogram of background coloration, calculates the histogram of background coloration and the mixed Gaussian mould of prospect coloration The overlapping possibility of type takes standard mean-square value P to overlapping possibility；Or for when background is substituted for normal background, to certificate photograph Portrait profile is detected using edge detecting technology, obtains the background information other than certificate photograph portrait profile；

Luminance information acquiring unit carries out data point to brightness histogram for obtaining the brightness histogram of certificate photograph Analysis calculates brightness exception COEFFICIENT K₁With average luminance shift value DA；

Chrominance information acquiring unit counts each pixel and exists for certificate photograph to be converted to Lab color space image The a axis of Lab color space image and coloration the average value Da and Db of b axis calculate colour cast coefficient according to coloration average value Da and Db K₂；

Sharpness information acquiring unit calculates sharpness factor DR for certificate photograph to be converted to gray level image.

Further, the quality detection module includes:

Shooting time detection unit judges whether for obtaining shooting time from file attribute information preset close Shooting in time phase, if so, photochrome is judged whether it is, if so, meeting the requirements；

Colored digit detection unit is used to obtain colored digit from file attribute information, judge whether not less than 24, If so, meeting the requirements；

It compresses Quality Detection unit and judges whether file is compression for obtaining file suffixes from file attribute information File judges whether it is JPEG compression when file is compressed file, if so, judging to compress whether quality meets corresponding certificate According to requirement；

File size detection unit judges whether file size is big for obtaining file size from file attribute information In the preset value of corresponding certificate photo, if so, meeting the requirements；

Image pixel detection unit judges whether image pixel accords with for obtaining image pixel from file attribute information Close the requirement of corresponding certificate photo；

Background detection unit, for obtaining standard mean-square value P, judgment criteria from background information when background is not replaced Whether mean-square value P is greater than default mean-square value, if it is not, then meeting the requirements；Or for when background is substituted for normal background, judgement to be obtained Whether the background information taken meets the requirement of corresponding certificate photo；

Bounding box features detection unit detects the whether well-regulated side of certificate photograph surrounding for using edge detecting technology Frame feature meets the requirements if nothing；

Brightness detection unit, for obtaining brightness exception COEFFICIENT K from luminance information₁With average luminance shift value DA, if Brightness exception COEFFICIENT K₁Less than 1, then brightness is normal, meets the requirements；If brightness exception COEFFICIENT K₁More than or equal to 1, and average brightness Deviant DA is greater than 0, then brightness is excessively bright；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is less than 0, Then too dark brightness；

Colorimetric detection units, for obtaining colour cast COEFFICIENT K from chrominance information₂, coloration average value Da and Db, if colour cast system Number K₂Less than 1, then coloration is normal, meets the requirements；If colour cast COEFFICIENT K₂More than or equal to 1, when coloration average value Da is greater than 0, table Show partially red, when coloration average value Da is less than 0, indicates partially green, when coloration average value Db is greater than 0, indicate partially yellow, it is flat in coloration When mean value Db is less than 0, indicate partially blue；

Clarity detection unit, for obtaining sharpness factor DR from sharpness information, if sharpness factor DR is greater than Default sharpness factor value, then explanation is clear, meets the requirements；

Face detection unit, for obtaining eyes, nose, ear, mouth, the crown and lower Bart from portrait characteristic information Data are levied, the eyes that will acquire, nose, ear, mouth, the crown and chin characteristic are compared with preset standard feature It is right, whether complete face are judged, if so, meeting the requirements；

Facial contour detection unit detects face wheel for obtaining facial contour feature data from portrait characteristic information Diversity factor between wide characteristic and preset facial contour feature is distorted, if diversity factor is greater than preset value if poor Different degree is less than or equal to preset value, then meets the requirements；Wherein, the facial contour feature data and preset facial contour feature Between diversity factor pass through the ratio calculation of non-overlap area and overlapping area；

Portrait posture detecting unit, for obtaining eyes, nose, ear, mouth, face wheel from portrait characteristic information Exterior feature, trunk profile, portrait profile characteristic, judge whether portrait posture is rectified, if so, meeting the requirements；

Head portrait size and location detection unit, for obtaining eyes, ear, the crown and lower Bart from portrait characteristic information Data are levied, eyes, ear, the crown and being associated property of chin characteristic are calculated, the wide pixel in head, head height picture are obtained Element, eyes range pixel, eyes position are gained fame and fortune away from up/down Edge Distance pixel and the crown away from certificate photograph top edge picture These data are compared with the parameter of corresponding certificate photo, judge whether to meet the requirements by element.

Further, the portrait posture detecting unit includes:

Face state-detection subelement, for sentencing according to the eyes of acquisition, nose, mouth and facial contour feature data Whether disconnected face is rectified, if so, meeting the requirements；

Shoulder state-detection subelement judges whether shoulder flushes for the trunk contour feature data according to acquisition, if It is then to meet the requirements；

Eyes open width detection subelement, for the eye feature data according to acquisition, calculate and open portion among eyes The width divided meets the requirements if width is greater than preset value；

Mouth closure opens detection sub-unit, for being sentenced according to the mouth characteristic of acquisition using edge detecting technology Among lower lip whether it is a curve on disconnected mouth, if so, illustrating that mouth is closed, meets the requirements, if it is not, then illustrating mouth It opens；

Detection sub-unit without a hat on, for the portrait profile characteristic according to acquisition, detect portrait profile characteristic with Diversity factor between preset portrait profile feature, if diversity factor is less than preset value, and head color then illustrates without obvious segmentation It is without a hat on, it meets the requirements；

Redeye detection subelement calculates the colour cast COEFFICIENT K of eyes for the eye feature data according to acquisition₃And coloration Average value Da1, if colour cast COEFFICIENT K₃Less than 1, then eyes coloration is normal, meets the requirements；If colour cast COEFFICIENT K₃More than or equal to 1, and When coloration average value Da1 is greater than 0, then there is blood-shot eye illness；

Adornment coloured silk detection sub-unit, for detecting the color of face according to the facial contour of acquisition, eyes and mouth characteristic Color deviation, the color error ratio value of eyes, the color error ratio value at mouth position, if the color error ratio value of one of them is greater than Preset deviation, then explanation has heavy make-up；If the color error ratio value of three is respectively less than or is equal to preset deviation, illustrate nothing Heavy make-up meets the requirements；

Face detection subelement, for obtaining face mask characteristic, detection face mask characteristic with it is preset Diversity factor between face mask feature, if diversity factor is less than preset value, hair does not cover face, meets the requirements.

Further, the face state-detection subelement judges whether face is rectified, and specifically includes:

According to eye feature data, two lines are generated, the angle of two lines and certificate photograph horizontal direction is calculated, sentences Whether the angle that breaks exceeds preset range, if it is not, then explanation faces camera；

According to eyes, nose and mouth characteristic, the center point coordinate of two lines is obtained, the center for obtaining nose is sat Mark, obtains the centre coordinate of mouth, does straight line fitting to three points and obtain straight line LC, calculate straight line LC and certificate photograph is vertical The angle in direction, judges whether the angle exceeds preset range, if it is not, then illustrating that head is vertical；

According to facial contour feature data, the center line FLC of human face region vertical direction is obtained, it is vertical to calculate human face region The distance DC of the center line FLC and straight line LC in direction, calculate the ratio of distance DC and human face region width, judge that the ratio is It is no to exceed preset range, if it is not, then illustrating face horizontal center；

When facing camera, head vertical and face horizontal center, judges that face is rectified, meet the requirements.

Further, the shoulder state-detection subelement judges whether shoulder flushes, and specifically includes:

To trunk profile coordinate array, it is fitted to a polygon, the center of gravity for calculating polygon obtains the coordinate of center of gravity (X1, Y1) calculates the distance Dweight of X1 and certificate photograph vertical center of gravity line, calculates distance Dweight and certificate photograph width The ratio of WF, judges whether the ratio exceeds preset range；

One minimum rectangle is obtained using minimum circumscribed rectangle algorithm to trunk profile coordinate array, calculates rectangular vertical The angle of center of gravity line and certificate photograph vertical center of gravity line, judges whether the angle exceeds preset range；

If the ratio of distance Dweight and certificate photograph width WF, rectangular vertical center of gravity line and certificate photograph vertical center of gravity The angle of line is all without departing from preset range, then shoulder flushes, and meets the requirements.

Another object of the present invention can be reached by adopting the following technical scheme that:

A kind of shooting quality detection method is applied in license camera, which comprises

The license camera obtains file attribute information, portrait characteristic information, background according to the certificate photograph of shooting imaging Information, luminance information, chrominance information and sharpness information；

The license camera is according to the file attribute information of acquisition, portrait characteristic information, background information, luminance information, color Information and sharpness information are spent, quality testing item by item is carried out to every dimensions；Wherein, every dimensions are according to government The examination criteria of public security industry standard formulation；

The license camera according to the default score values of every dimensions, to the quality measurements of every dimensions into Row scoring, obtains the score value of every dimensions；

The license camera calculates the quality of certificate photograph according to the score value and scoring weight of every dimensions Overall score；

The license camera provides the shooting suggestion of profession according to the quality overall score of certificate photograph.

Further, the license camera obtains file attribute information, portrait feature according to the certificate photograph of shooting imaging Information, background information, luminance information, chrominance information and sharpness information, specifically include:

Obtain the file attribute information of certificate photograph；Wherein, the file attribute information includes shooting time, color bit Number, file suffixes, file size and image pixel；

Obtain the portrait characteristic information of certificate photograph；Wherein, the portrait characteristic information includes eyes, nose, ear, mouth Bar, facial contour, trunk profile, portrait profile, the crown and chin characteristic；

When background is not replaced, certificate photograph is converted into Lab color space image, to Lab color space image, is read Each pixel of foreground area, establishes the mixed Gauss model of prospect coloration；To each pixel of background area, background colour is obtained The histogram of degree calculates the overlapping possibility of the histogram of background coloration and the mixed Gauss model of prospect coloration, to overlapping possibility Take standard mean-square value P；Or when background is substituted for normal background, portrait wheel is detected using edge detecting technology to certificate photograph Exterior feature obtains the background information other than certificate photograph portrait profile；

The brightness histogram for obtaining certificate photograph carries out data analysis to brightness histogram, calculates brightness exception COEFFICIENT K₁ With average luminance shift value DA；

Certificate photograph is converted into Lab color space image, counts each pixel in a axis and b of Lab color space image Coloration the average value Da and Db of axis calculate colour cast COEFFICIENT K according to coloration average value Da and Db₂；

Certificate photograph is converted into gray level image, calculates sharpness factor DR.

Further, the license camera is according to the file attribute information of acquisition, portrait characteristic information, background information, bright Information, chrominance information and sharpness information are spent, quality testing item by item is carried out to every dimensions, is specifically included:

Shooting time is obtained from file attribute information, judges whether to shoot in preset recent times, if so, sentencing Whether disconnected is photochrome, if so, meeting the requirements；

Colored digit is obtained from file attribute information, is judged whether not less than 24, if so, meeting the requirements；

File suffixes is obtained from file attribute information, judges whether file is compressed file, when file is compressed file When, JPEG compression is judged whether it is, if so, judging to compress the requirement whether quality meets corresponding certificate photo；

File size is obtained from file attribute information, judges whether file size is greater than the preset value of corresponding certificate photo, If so, meeting the requirements；

Image pixel is obtained from file attribute information, judges whether image pixel meets the requirement of corresponding certificate photo；

When background is not replaced, standard mean-square value P is obtained from background information, it is pre- whether judgment criteria mean-square value P is greater than If mean-square value, if it is not, then meeting the requirements；Or for when background is substituted for normal background, judge obtain background information whether Meet the requirement of corresponding certificate photo；

Using edge detecting technology, detects the whether well-regulated bounding box features of certificate photograph surrounding and conformed to if nothing It asks；

Brightness exception COEFFICIENT K is obtained from luminance information₁With average luminance shift value DA, if brightness exception COEFFICIENT K₁It is less than 1, then brightness is normal, meets the requirements；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is greater than 0, then Brightness is excessively bright；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is less than 0, then too dark brightness；

Colour cast COEFFICIENT K is obtained from chrominance information₂, coloration average value Da and Db, if colour cast COEFFICIENT K₂Less than 1, then coloration is being just Often, it meets the requirements；If colour cast COEFFICIENT K₂More than or equal to 1, when coloration average value Da is greater than 0, indicate partially red, it is average in coloration It when value Da is less than 0, indicates partially green, when coloration average value Db is greater than 0, indicates partially yellow, when coloration average value Db is less than 0, table Show partially blue；

Sharpness factor DR is obtained from sharpness information, if sharpness factor DR is greater than default sharpness factor value, It is clear to illustrate, meets the requirements；

Eyes, nose, ear, mouth, the crown and chin characteristic, the eye that will acquire are obtained from portrait characteristic information Eyeball, nose, ear, mouth, the crown and chin characteristic, are compared with preset standard feature, judge whether face are neat Entirely, if so, meeting the requirements；

Facial contour feature data, detection facial contour feature data and preset face are obtained from portrait characteristic information Diversity factor between contour feature is distorted if diversity factor is greater than preset value, if diversity factor is less than or equal to preset value, Then meet the requirements；Wherein, the diversity factor between the facial contour feature data and preset facial contour feature passes through non-heavy The ratio calculation of folded area and overlapping area；

It is special that eyes, nose, ear, mouth, facial contour, trunk profile, portrait profile are obtained from portrait characteristic information Data are levied, judge whether portrait posture is rectified, if so, meeting the requirements；

Eyes, ear, the crown and chin characteristic are obtained from portrait characteristic information, to eyes, ear, the crown and under Bar being associated property of characteristic calculates, and obtains the wide pixel in head, the high pixel in head, eyes range pixel, eyes position It gains fame and fortune away from up/down Edge Distance pixel and the crown away from certificate photograph top edge pixel, by these data and corresponding certificate photo Parameter is compared, and judges whether to meet the requirements.

Further, described to judge whether portrait posture is rectified, it specifically includes:

According to the eyes of acquisition, nose, mouth and facial contour feature data, judge whether face is rectified, if so, symbol It closes and requires；

According to the trunk contour feature data of acquisition, judge whether shoulder flushes, if so, meeting the requirements；

According to the eye feature data of acquisition, the width that part is opened among eyes is calculated, if width is greater than preset value, It meets the requirements；

According to the mouth characteristic of acquisition, using edge detecting technology, judge on mouth among lower lip whether to be one Curve meets the requirements if so, illustrating that mouth is closed, if it is not, then illustrating that mouth opens；

According to the portrait profile characteristic of acquisition, detect portrait profile characteristic and preset portrait profile feature it Between diversity factor, if diversity factor is less than preset value, and head color without obvious segmentation, then explanation is without a hat on, meets the requirements；

According to the eye feature data of acquisition, the colour cast COEFFICIENT K of eyes is calculated₃With coloration average value Da1, if colour cast coefficient K₃Less than 1, then eyes coloration is normal, meets the requirements；If colour cast COEFFICIENT K₁More than or equal to 1, and coloration average value Da1 is greater than 0 When, then there is blood-shot eye illness；

According to the facial contour of acquisition, eyes and mouth characteristic, the color error ratio value of face, eyes are detected Color error ratio value, the color error ratio value at mouth position illustrate if the color error ratio value of one of them is greater than preset deviation There is heavy make-up；If the color error ratio value of three is respectively less than or is equal to preset deviation, illustrates no heavy make-up, meet the requirements；

Face mask characteristic is obtained, the difference between face mask characteristic and preset face mask feature is detected Different degree, if diversity factor is less than preset value, hair does not cover face, meets the requirements.

Further, described to judge whether face is rectified, it specifically includes:

Further, described to judge whether shoulder flushes, it specifically includes:

The present invention have compared with the existing technology it is following the utility model has the advantages that

1, license camera of the invention and method are according to the every dimensions of government's public security industry standard formulation, and establish The standard database of different type certificate photo, by the information of certificate photograph after acquisition shooting, (such as file attribute information, portrait are special Reference breath, background information, luminance information, chrominance information, sharpness information), quality testing item by item is carried out to every dimensions, Ensure to meet by the certificate photograph of detection the requirement of certificate photograph relevant criterion.

2, whether the dimensions that license camera of the invention and method are formulated share 14, be respectively: being recent colour Whether whether photo be not less than 24 RGB true color, be JPEG compression technology and compress whether quality meets the requirements, file holds Amount whether meet the requirements, whether the wide pixel of image image height meets the requirements, whether background color meets the requirements, photo surrounding whether Whether Rimless, image are clear, whether face are visible, whether facial contour is without obvious distortion, the whether uniform sufficient, image of brightness Whether no color differnece, whether posture is rectified, whether head portrait size and location meet the requirements, it is seen that realize to certificate photograph after shooting Complete detection, so that the quality testing of certificate photograph is reached objective and accurate requirement.

3, license camera of the invention and method can detect the shooting quality of the certificate photograph after shooting imaging automatically, pass through Marking and shooting proposed mechanism, allow the certificate photograph for being unsatisfactory for requiring to be truncated at source, save subsequent artefacts' one by onechecking Time, improve detection efficiency.

Detailed description of the invention

Fig. 1 is the license camera function module map of the embodiment of the present invention 1.

Fig. 2 is the portrait posture detecting unit structure chart of the embodiment of the present invention 1.

Fig. 3 is the shooting quality detection method flow chart of the embodiment of the present invention 2.

Specific embodiment

Present invention will now be described in further detail with reference to the embodiments and the accompanying drawings, but embodiments of the present invention are unlimited In this.

Embodiment 1:

License camera is mounted in the APP (Application, application program) on intelligent terminal, it can have Android Version and iOS version, are responsible for shooting a width and meet the license original image that certificate making uses, and license original image refers to be passed through by user Obtained by license camera captured in real-time, the useful region of image, and color, brightness, background, personage biology spy are cut according to license standard The image datas of the image informations without any processing such as sign.

The license camera of the present embodiment is directed to shoot the certificate photograph of imaging, can carry out quality to certificate photograph Detection, mainly according to the every dimensions of government's public security industry standard formulation, and establishes the criterion numeral of different type certificate photo According to library, the certificate photograph data information that will acquire is compared with the data of database, and one by onechecking assesses whether to meet the requirements, And mass overall score is calculated according to testing result and is then built when preset value (such as 60 points) are not achieved in the score value of quality overall score View is re-shoot.

Therefore, the license camera of the present embodiment includes that data obtaining module, quality detection module, grading module, quality are total Score computing module and shooting suggestion module, as shown in Figure 1；The concrete function of modules is as follows:

The data obtaining module obtains file attribute information, portrait feature for the certificate photograph according to shooting imaging Information, background information, luminance information, chrominance information and sharpness information；The data obtaining module includes:

Portrait characteristic acquisition unit, for obtaining the portrait characteristic information of certificate photograph；Wherein, the portrait feature Information includes eyes, nose, ear, mouth, facial contour, trunk profile, portrait profile, the crown and chin characteristic；People As the extraction of characteristic information can be as follows:

1) certificate photograph is converted into gray level image；

2) upper part of the body characteristic region detection is carried out to gray level image, when detecting upper part of the body characteristic region, It is upper part of the body region rectangle data structure by the area pixel information preservation, and upper part of the body ROI region, note is arranged to gray level image For ROIB；

3) detection of facial contour feature data area is carried out to ROIB, when detecting facial contour feature data area, It is face contour area rectangle data structure by the area pixel information preservation, and the area facial contour ROI is arranged to gray level image Domain is denoted as ROIF；

4) eyes characteristic region detection is carried out to the specific region of ROIF, when detecting eyes characteristic region When, it is eyes region rectangle data structure by the area pixel information preservation, and obtain the coordinate of eyes, most with the Y-axis of eyes Low spot is to take eyes region below at the top of new ROI region to facial contour feature data area, be set as nose target ROI region is denoted as ROIN；

5) ears characteristic region detection is carried out to ROIF, when detecting ears characteristic region, by the region Pixel Information saves as ears region rectangle data structure；

6) nose characteristic region detection is carried out to ROIN, when detecting nose characteristic region, by the region Pixel Information saves as nasal area rectangle data structure, using the bottom edge of nasal area rectangle data structure as the area Xin ROI The top margin in domain takes nose following region to facial contour feature data area, is set as mouth target ROI region, is denoted as ROIM；

7) mouth characteristic region detection is carried out to ROIM, when detecting mouth characteristic region, by the region Pixel Information saves as mouth region rectangle data structure；

8) to the region of ROIB, the region for removing ROIF is cut out, trunk contour feature data area ROIMB is obtained, by the region Pixel Information saves as trunk contour area rectangle data structure；

9) edge detection is carried out to ROIF, obtains the line segment at the sign mutation edge in contouring head regional scope, then Convex closure contour detecting in head is carried out to the head edge line segment result, the coordinate for the key point that head convex closure contour detecting is obtained Save as contouring head array；

10) edge detection is carried out to ROIMB, obtains the line segment at the sign mutation edge within the scope of trunk contour area, so Trunk convex closure contour detecting is carried out to the trunk edge line segment result afterwards, the seat for the key point that trunk convex closure contour detecting is obtained Mark saves as trunk profile array；

11) contouring head array and trunk number of contours group are merged, forms portrait profile array.

By the extraction of above-mentioned portrait feature to get arrive eyes, nose, ear, mouth, facial contour, trunk profile, people As contour feature data, according to the available crown characteristic of eye position, this part has in detail in other patent document It introduces, details are not described herein, and the characteristic of chin is also to be obtained according to same principle.

Luminance information acquiring unit carries out data point to brightness histogram for obtaining the brightness histogram of certificate photograph Analysis calculates brightness exception COEFFICIENT K₁With average luminance shift value DA, specifically include:

1) brightness value (Y-component, Y (n)) that each pixel is read from the image array of certificate photograph, counts each brightness 0~255 number of pixels, 128 be median, obtains the brightness histogram of certificate photograph；

2) data analysis is carried out to brightness histogram, calculates brightness integrated value:

Wherein, constant 128 is intensity deviation value；Col is the column of the image array of certificate photograph, and cols is maximum column； Row is the row of the image array of certificate photograph, and rows is maximum row；Y (row, col) is certain in the image array of certificate photograph The brightness value of a bit；SumDA is intensity deviation summation of each pixel with respect to 128；

3) the intensity deviation summation acquired is averaging sum of all pixels, obtains average luminance shift value, is expressed as DA:

DA=SumDA/ (cols*rows)

The absolute value for taking average intensity deviation value DA, remembers D₁:

D₁=abs (DA)

4) brightness histogram is an array HIST [i], under be designated as brightness value, range is 0 to 255, and cell value is that this is bright The number of pixels of angle value, obtains:

Wherein, Mda is the intensity deviation value summation of each brightness degree of brightness histogram；

5) it takes absolute value, and averages to Mda:

M₁=abs (Mda)/(cols*rows)

6) according to M₁、D₁, acquire brightness exception coefficient:

K₁=D₁/M₁

Wherein, K₁For brightness exception coefficient.

Chrominance information acquiring unit counts each pixel and exists for certificate photograph to be converted to Lab color space image The a axis of Lab color space image and coloration the average value Da and Db of b axis calculate colour cast coefficient according to coloration average value Da and Db K₂, it specifically includes:

1) certificate photograph is converted into Lab color space image, each pixel is counted in Lab color space figure using following formula The a axis of picture and the coloration average value Da (red green colour cast estimated value) of b axis and Db (yellow blue colour cast estimated value):

Wherein, Vec3b (row, col) is the structural body read on the Lab color space of a pixel, its first is Brightness value, second are a axis values of Lab, and third position is the b axis value of Lab；Col is Lab color space image matrix column, Cols is maximum column；Row is the row of Lab color space image, and rows is maximum row；Constant 128 is intensity deviation value；

2) the variance yields D of a axis and b axis average value is calculated₂:

3) brightness histogram HIST_A and HIST_B of the image on a axis and b axis are calculated, two groups of brightness histograms have respectively 256 statistical values；

4) the average value Ma and Mb of a axis and b axis color difference are asked:

5) colour cast total value M is sought₂:

6) according to variance yields D₂With colour cast total value M₂, acquire colour cast coefficient:

K₂=D₂/M₂

Wherein, K₂For colour cast coefficient.

Sharpness information acquiring unit calculates sharpness factor DR, specifically for certificate photograph to be converted to gray level image Include:

1) certificate photograph is converted into gray level image, sharpness meter is based on implementing to calculate on the basis of gray scale at last, clearly Degree can also be expressed as sharpness of vision, be feeling of the people to the sharp keen degree of the image seen, by a large amount of experiment and research point Analysis, it is recognized that concept be that sharpness of vision is made of two factors of resolution ratio and object edge profile contrast.Sharpness of vision is transition Steepness (slope), equal to output brightness variation divided by position variation；

Calculated to simplify, using the excessive slope of brightness for calculating horizontally adjacent two pixels and vertically adjacent to four The average value of the brightness interpolating of pixel, the measurement index as evaluation clarity；

2) for a gray level image, it is assumed that pixel is following matrix:

Consecutive points are A, B, C and D；

3) it using adjacent 4 points of luminance differences both horizontally and vertically as two right-angle sides of right angled triangle, calculates Adjacent 4 points brightness change curvature both horizontally and vertically:

And the luminance difference of two o'clock adjacent with A:

Δ 1=abs (B-A)+abs (C-A)

4) the luminance difference average value of each point and its consecutive points in grayscale image matrix is calculated using following formula:

Wherein, θ (row, col) be adjacent 4 points brightness change curvature both horizontally and vertically, Δ (row, It col is) luminance difference of two o'clock adjacent with certain point, col is grayscale image matrix column, and cols is maximum column；Row is grayscale image The row of matrix, rows are maximum row, and DR is sharpness factor.

The quality detection module, for according to the file attribute information of acquisition, portrait characteristic information, background information, bright Information, chrominance information and sharpness information are spent, quality testing item by item is carried out to every dimensions；Wherein, every dimensions For according to the examination criteria of government's public security industry standard formulation, including 14 sports, the detection orderings of every dimensions can be by Whether user's self-setting is respectively: 1) being recent photochrome；2) whether it is not less than 24 RGB true color；3) whether it is JPEG compression technology, and compress whether quality meets the requirements；4) whether file size meets the requirements；5) the wide pixel of image image height Whether meet the requirements；6) whether background color meets the requirements；7) photo surrounding whether Rimless；8) whether accurate, that is, scheme if focusing It seem no clear；9) whether face is clear, level is abundant, i.e., whether face are visible；10) whether facial contour is without obvious distortion； 11) whether brightness uniformly sufficient, face whether shadow-free, speck；12) whether image color is balanced, and whether skin is presented really Tone, i.e., whether no color differnece；13) whether posture is rectified；14) whether head portrait size and location meet the requirements；Therefore, which examines Surveying module includes:

Shooting time detection unit judges whether for obtaining shooting time from file attribute information preset close Shooting in time phase (being usually set to 6 months, section sets are 3 months), if so, judge whether it is photochrome, if so, Then explanation is recent photochrome, is met the requirements；

Colored digit detection unit judges whether for obtaining colored digit from file attribute information not less than 24 (being greater than or equal to 24), if so, meeting the requirements；

It compresses Quality Detection unit and judges whether file is compression for obtaining file suffixes from file attribute information File judges whether it is JPEG compression when file is compressed file, if so, judging to compress whether quality meets corresponding certificate According to requirement, such as:

[identity card/residence permit/social security card] compresses quality factor >=85；

[driver's license] compresses quality factor >=85；

[entry and exit card] maximum compression multiple is no more than 25 times；

If meeting above-mentioned standard, prove that compression quality meets the requirements.

File size detection unit judges whether file size is big for obtaining file size from file attribute information In the preset value of corresponding certificate photo, if so, meeting the requirements；The requirement of different certificate photos is as follows:

[identity card/residence permit/social security card] photo files capacity > 30K；

[driver's license] photo files capacity > 30K；

[entry and exit card] photo files capacity > 80K.

Image pixel detection unit judges whether image pixel accords with for obtaining image pixel from file attribute information Close the requirement of corresponding certificate photo；It the following is the requirement of different certificate photos:

[identity card/residence permit/social security card] image image height is wide > and 441 (height) × 358 (width) pixels (meet accreditation Dpi350 or more meets stamp with the size > 26mm*32mm)；

[driver's license] image image height is wide > and 378 (height) × 260 (width) pixels (meet accreditation dpi300 or more, meet printing Size > 32mm*22mm)；

[entry and exit card] image image height is wide>640 pixels (height) × 480 pixel (width) pixel (472~640 pixels of satisfaction< Height>× 354~480 pixels<wide>, meet the ratio of width to height 3:4, meet accreditation dpi300 or more, meet stamp with the size>33mm* 48mm)。

Background detection unit, for obtaining standard mean-square value P, judgment criteria from background information when background is not replaced Whether mean-square value P is greater than default mean-square value (being usually set to 50), if it is not, then foreground and background color overlapping degree is lower, symbol It closes and requires；Or for when background is substituted for normal background, judging whether the background information obtained meets wanting for corresponding certificate photo It asks, such as:

[identity card/residence permit/social security card] white background

[driver's license] white background

[entry and exit card] Guangdong blue background, other regional white backgrounds, white background tone value and intensity value are 0, Brightness value is 240；Light blue tone value is 135, and intensity value 240, brightness value is not less than 167；

Colorimetric detection units, for obtaining colour cast COEFFICIENT K from chrominance information₂, coloration average value Da and Db, if colour cast system Number K₂Less than 1, then coloration is normal, illustrates that true tone is presented in image color balance, skin, meets the requirements；If colour cast COEFFICIENT K₂Greatly In or be equal to 1, coloration average value Da be greater than 0 when, indicate it is partially red, when coloration average value Da is less than 0, indicate it is partially green, in color It when spending average value Db greater than 0, indicates partially yellow, when coloration average value Db is less than 0, indicates partially blue；

Clarity detection unit, for obtaining sharpness factor DR from sharpness information, if sharpness factor DR is greater than Default sharpness factor value (being usually set to 14), then explanation is clear, and focusing is accurate, meets the requirements；

Wherein, a variety of different certificate photos have different standards, as follows:

Wide 207 ± 14 pixel in [identity card/residence permit/social security card] head；Eyes position apart from photograph lower edge away from From ≮ 207 pixels；It gains fame and fortune away from 7~21 pixel of photograph upper edge on the crown

Wide 165~189 pixel in [driver's license] head (meeting stamp with the size 14mm~16mm)；Height of head is 224~260 (meet stamp with the size 19mm~22mm) between pixel；It gains fame and fortune away from 10~20 pixel of photograph upper edge on the crown

189~283 pixel (meeting stamp with the size 16mm~22mm, dpi300 or more) of [entry and exit card] head width；Head Portion's height (meets stamp with the size 30mm~34mm, dpi300 or more) between 354~402 pixels；Interpupillary distance is 0.231 times Wide~0.333 times of wide pixel；The high pixel in 0.301 times away from photo upper edge of adult eye position height~0.500 times is (full Sufficient stamp with the size 14mm~22mm, dpi300 or more)；0.301 times away from photo upper edge of children's eyes position height~0.600 High pixel (meeting stamp with the size 14mm~26mm, dpi300 or more) again；The crown gain fame and fortune 0.025 times away from photograph upper edge height~ 0.074 times of high pixel.

As shown in Fig. 2, above-mentioned portrait posture detecting unit, specifically includes:

Face state-detection subelement, for sentencing according to the eyes of acquisition, nose, mouth and facial contour feature data Whether disconnected face is rectified, if so, meeting the requirements；Judge whether face is rectified, specifically include:

Shoulder state-detection subelement judges whether shoulder flushes for the trunk contour feature data according to acquisition, if It is then to meet the requirements；Judge whether shoulder flushes, specifically include:

Eyes open width detection subelement, for the eye feature data according to acquisition, calculate and open portion among eyes The width divided meets the requirements if width is greater than preset value.

Mouth closure opens detection sub-unit, for being sentenced according to the mouth characteristic of acquisition using edge detecting technology Among lower lip whether it is a curve on disconnected mouth, if so, illustrating that mouth is closed, meets the requirements, if it is not, then illustrating mouth It opens.

Detection sub-unit without a hat on, for the portrait profile characteristic according to acquisition, detect portrait profile characteristic with Diversity factor between preset portrait profile feature, if diversity factor is less than preset value, and head color then illustrates without obvious segmentation It is without a hat on, it meets the requirements.

Redeye detection subelement calculates the colour cast COEFFICIENT K of eyes for the eye feature data according to acquisition₃And coloration Average value Da1 is (referring to above-mentioned colour cast COEFFICIENT K₂With the calculating process of coloration average value Da), if colour cast COEFFICIENT K₃Less than 1, then eyes Coloration is normal, meets the requirements；If colour cast COEFFICIENT K₃More than or equal to 1, and coloration average value Da1 be greater than 0 when, then exist blood-shot eye illness.

Adornment coloured silk detection sub-unit, for detecting the color of face according to the facial contour of acquisition, eyes and mouth characteristic Color deviation, the color error ratio value of eyes, the color error ratio value at mouth position, if the color error ratio value of one of them is greater than Preset deviation, then explanation has heavy make-up；If the color error ratio value of three is respectively less than or is equal to preset deviation, illustrate nothing Heavy make-up meets the requirements；The meter of the color error ratio value of face, the color error ratio value of eyes, the color error ratio value at mouth position It calculates, can refer to above-mentioned colour cast COEFFICIENT K₂Calculating process.

Institute's scoring module, the quality testing for the default score value according to every dimensions, to every dimensions As a result it scores, obtains the score value of every dimensions；The default score value of every dimensions of the present embodiment is as follows:

It 1) whether is recent photochrome, individual event is 5 points when meeting, individual event score=(- 40) point when not meeting；

2) whether it is not less than 24 RGB true color, individual event is 5 points when meeting, individual event score=(- 40) point when not meeting；

3) whether it is JPEG compression technology, and compresses whether quality meets the requirements, individual event is 5 points, when not meeting when meeting Individual event score=(- 40) point；

4) whether file size meets the requirements, image file size >=type of credential minimum, and individual event is 5 points, image text Individual event score=(- 40) point when part size < type of credential minimum；

5) whether the wide pixel of image image height meets the requirements, image image height >=type of credential minimum pixel, wide >=certificate Type minimum pixel, individual event are 10 points, and image image height < type of credential minimum pixel is single when wide < type of credential minimum pixel Item score=(- 40) point；

6) whether background color meets the requirements, and individual event is 5 points when meeting, individual event score=(- 40) point when not meeting；

7) photo surrounding whether Rimless, individual event is 5 points when meeting, individual event score=(- 40) point when not meeting；

8) whether accurate, i.e., whether image is clear if focusing, individual event is 5 points when meeting, individual event score=(- 40) when not meeting Point；

9) whether face is clear, level is abundant, and individual event is 5 points when meeting, individual event score=(- 40) point when not meeting；

10) whether facial contour is without obvious distortion, and individual event is 5 points when meeting, individual event score=(- 40) point when not meeting；

11) whether brightness uniformly sufficient, face whether shadow-free, speck, individual event is 5 points when meeting, individual event when not meeting Score=(- 40) point；

12) whether image color is balanced, and whether skin is presented true tone, and individual event is 5 points when meeting, individual event when not meeting Score=(- 40) point；

13) whether posture is rectified, and all detections meet to obtain 15 points, individual event score=(- 40) point when not meeting；

14) whether head portrait size and location meet the requirements, and all detections obtain 20 points when meeting, and individual event score when not meeting= (- 40) divide；

The quality overall score computing module is calculated for the score value and scoring weight according to every dimensions The quality overall score of certificate photograph；When general comment score value is lower than 0 timesharing, overall score all carrys out recording and displaying by 0 point.

The shooting suggestion module provides the shooting suggestion of profession, example for the quality overall score according to certificate photograph Such as:

1) 90-100 points: advanced photographer: your shooting effect is very good, can enter in next step！

2) 80--89 points: intermediate photographer: your shooting effect is fine, can enter in next step！

3) 60-79 points: primary photographer: your shooting effect is general, then claps one and have a try.

4) 0-59 points: photography green hand: your shooting effect is not good enough, it is proposed that re-shoots.

Embodiment 2:

As shown in figure 3, present embodiments providing a kind of shooting quality detection method, this method is directed to shooting imaging, But the shooting quality detection for not replacing the certificate photograph of background, is applied in license camera, comprising the following steps:

S1, the certificate photograph being imaged according to shooting, obtain file attribute information, portrait characteristic information, background information, brightness Information, chrominance information and sharpness information, specifically include:

S101, the file attribute information for obtaining certificate photograph；Wherein, the file attribute information includes shooting time, coloured silk Color bits number, file suffixes, file size and image pixel；

S102, the portrait characteristic information for obtaining certificate photograph；Wherein, the portrait characteristic information includes eyes, nose, ear Piece, mouth, facial contour, trunk profile, portrait profile, the crown and chin characteristic；

S103, certificate photograph is converted to Lab color space image, to Lab color space image, reads foreground area Each pixel establishes the mixed Gauss model of prospect coloration；To each pixel of background area, the histogram of background coloration is obtained Figure calculates the overlapping possibility of the histogram of background coloration and the mixed Gauss model of prospect coloration, takes standard equal overlapping possibility Side value P；

S104, the brightness histogram for obtaining certificate photograph carry out data analysis to brightness histogram, calculate brightness and are extremely Number K₁With average luminance shift value DA；

S105, certificate photograph is converted to Lab color space image, counts each pixel in a of Lab color space image Coloration the average value Da and Db of axis and b axis calculate colour cast COEFFICIENT K according to coloration average value Da and Db₂；

S106, certificate photograph is converted to gray level image, calculates sharpness factor DR；

S2, according to the file attribute information of acquisition, portrait characteristic information, background information, luminance information, chrominance information and clear Clear degree information carries out quality testing item by item to every dimensions, specific as follows:

S201, shooting time is obtained from file attribute information, judge whether to shoot in preset recent times, if It is then to judge whether it is photochrome, if so, meeting the requirements；It if not being shot in preset recent times, or is not color It is when color photo, then undesirable；

S202, colored digit is obtained from file attribute information, judge whether not less than 24, if so, meet the requirements, If it is not, then undesirable；

S203, file suffixes is obtained from file attribute information, judge whether file is compressed file, when file is compression When file, JPEG compression is judged whether it is, if so, judging to compress the requirement whether quality meets corresponding certificate photo；

S204, file size is obtained from file attribute information, judge whether file size is greater than the pre- of corresponding certificate photo If value, if so, meeting the requirements, if it is not, then undesirable；

S205, image pixel is obtained from file attribute information, judge whether image pixel meets wanting for corresponding certificate photo It asks；

S206, standard mean-square value P is obtained from background information, whether judgment criteria mean-square value P is greater than default mean-square value, if It is no, then it meets the requirements；If so, undesirable；

S207, using edge detecting technology, detect the whether well-regulated bounding box features of certificate photograph surrounding and accorded with if nothing It closes and requires；If so, then undesirable；

S208, brightness exception COEFFICIENT K is obtained from luminance information₁With average luminance shift value DA, if brightness exception COEFFICIENT K₁ Less than 1, then brightness is normal, meets the requirements；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is greater than 0, then brightness is excessively bright；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is less than 0, then too dark brightness； It is undesirable when brightness is excessive lightness or darkness；

S209, colour cast COEFFICIENT K is obtained from chrominance information₂, coloration average value Da and Db, if colour cast COEFFICIENT K₂Less than 1, then Coloration is normal, meets the requirements；If colour cast COEFFICIENT K₂More than or equal to 1, when coloration average value Da is greater than 0, indicate partially red, in color It when spending average value Da less than 0, indicates partially green, when coloration average value Db is greater than 0, indicates partially yellow, in coloration average value Db less than 0 When, indicate partially blue；It is undesirable when partially red, partially green, partially yellow or partially blue；

S210, sharpness factor DR is obtained from sharpness information, if sharpness factor DR is greater than default sharpness factor Value, then explanation is clear, meets the requirements；If sharpness factor DR is less than or equal to default sharpness factor value, undesirable；

S211, eyes, nose, ear, mouth, the crown and chin characteristic are obtained from portrait characteristic information, will obtain Eyes, nose, ear, mouth, the crown and the chin characteristic taken, is compared with preset standard feature, judges face It is whether complete, if so, meeting the requirements, if it is not, then undesirable；

S212, facial contour feature data are obtained from portrait characteristic information, detect facial contour feature data and preset Facial contour feature between diversity factor, if diversity factor be greater than preset value, be distorted, it is undesirable；If diversity factor Less than or equal to preset value, then meet the requirements；

S213, eyes, nose, ear, mouth, facial contour, trunk profile, portrait wheel are obtained from portrait characteristic information Wide characteristic, judges whether portrait posture is rectified, if so, meeting the requirements, if it is not, then undesirable；

S214, eyes, ear, the crown and chin characteristic are obtained from portrait characteristic information, to eyes, ear, head Top and being associated property of chin characteristic calculate, and obtain the wide pixel in head, the high pixel in head, eyes range pixel, eyes institute It gains fame and fortune away from up/down Edge Distance pixel and the crown away from certificate photograph top edge pixel in position, by these data and corresponding card The parameter that part shines is compared, and judges whether to meet the requirements；

S3, the default score value according to every dimensions, to the matter of dimensions every in above-mentioned steps S201~S214 Amount testing result scores, and obtains the score value of every dimensions；

S4, score value and scoring weight according to every dimensions, calculate the quality overall score of certificate photograph；

S5, the quality overall score according to certificate photograph provide the shooting suggestion of profession.

Embodiment 3:

The shooting quality detection method of the present embodiment is directed to shooting imaging, and is substituted for the certificate photograph of normal background Shooting quality detection, and be in place of the difference of embodiment 2:

S103, portrait profile detected using edge detecting technology to certificate photograph, obtain certificate photograph portrait profile with Outer background information；

S206, judge whether the background information obtained meets the requirement of corresponding certificate photo.

In conclusion license camera and method of the invention are tieed up according to each item rating of government's public security industry standard formulation Degree, and the standard database of different type certificate photo is established, by the information of certificate photograph after acquisition shooting, each item rating is tieed up Degree carries out quality testing item by item, it is ensured that can meet the requirement of certificate photograph relevant criterion by the certificate photograph of detection.

The above, only the invention patent preferred embodiment, but the scope of protection of the patent of the present invention is not limited to This, anyone skilled in the art is in the range disclosed in the invention patent, according to the present invention the skill of patent Art scheme and its inventive concept are subject to equivalent substitution or change, belong to the scope of protection of the patent of the present invention.

Claims

1. a kind of license camera of detectable shooting quality, it is characterised in that: the license camera includes:

Data obtaining module obtains file attribute information, portrait characteristic information, back for the certificate photograph according to shooting imaging Scape information, luminance information, chrominance information and sharpness information；

The data obtaining module includes:

File attribute information acquiring unit, for obtaining the file attribute information of certificate photograph；Wherein, the file attribute information Including shooting time, colored digit, file suffixes, file size and image pixel；

Portrait characteristic acquisition unit, for obtaining the portrait characteristic information of certificate photograph；Wherein, the portrait characteristic information Including eyes, nose, ear, mouth, facial contour, trunk profile, portrait profile, the crown and chin characteristic；

Background information acquiring unit, for certificate photograph being converted to Lab color space image, to Lab when background is not replaced Color space image reads each pixel of foreground area, establishes the mixed Gauss model of prospect coloration；To the every of background area A pixel obtains the histogram of background coloration, calculates the weight of the histogram of background coloration and the mixed Gauss model of prospect coloration Folded probability, takes standard mean-square value P to overlapping possibility；Or for using side to certificate photograph when background is substituted for normal background Edge detection technique detects portrait profile, obtains the background information other than certificate photograph portrait profile；

Luminance information acquiring unit carries out data analysis, meter to brightness histogram for obtaining the brightness histogram of certificate photograph Calculate brightness exception COEFFICIENT K₁With average luminance shift value DA；

Chrominance information acquiring unit counts each pixel in Lab color for certificate photograph to be converted to Lab color space image The a axis of color space image and coloration the average value Da and Db of b axis calculate colour cast COEFFICIENT K according to coloration average value Da and Db₂；

Sharpness information acquiring unit calculates sharpness factor DR for certificate photograph to be converted to gray level image；

Quality detection module, for according to the file attribute information of acquisition, portrait characteristic information, background information, luminance information, color Information and sharpness information are spent, quality testing item by item is carried out to every dimensions；Wherein, every dimensions are according to government The examination criteria of public security industry standard formulation, comprising: whether be recent photochrome, whether be not less than 24 RGB true color, be It is no to be JPEG compression technology and compress whether quality meets the requirements, whether file size meets the requirements, the wide pixel of image image height Whether meet the requirements, whether background color meets the requirements, whether Rimless, image clear, whether face may be used for photo surrounding See, facial contour whether without obvious distortion, brightness whether uniformly sufficient, image whether no color differnece, posture whether rectify, head portrait it is big Whether small and position meets the requirements；

Grading module carries out the quality measurements of every dimensions for the default score value according to every dimensions Scoring, obtains the score value of every dimensions；

Quality overall score computing module calculates certificate photograph for the score value and scoring weight according to every dimensions Quality overall score；

2. a kind of license camera of detectable shooting quality according to claim 1, it is characterised in that: the quality testing Module includes:

Shooting time detection unit judges whether when preset recent for obtaining shooting time from file attribute information Interior shooting, if so, photochrome is judged whether it is, if so, meeting the requirements；

Colored digit detection unit is used to obtain colored digit from file attribute information, judge whether not less than 24, if It is then to meet the requirements；

Quality Detection unit is compressed, for obtaining file suffixes from file attribute information, judges whether file is compressed file, When file is compressed file, JPEG compression is judged whether it is, if so, judging to compress whether quality meets wanting for corresponding certificate photo It asks；

File size detection unit judges whether file size is greater than phase for obtaining file size from file attribute information The preset value of certificate photo is answered, if so, meeting the requirements；

Image pixel detection unit judges whether image pixel meets phase for obtaining image pixel from file attribute information Answer the requirement of certificate photo；

Background detection unit, for obtaining standard mean-square value P from background information when background is not replaced, judgment criteria is square Whether value P is greater than default mean-square value, if it is not, then meeting the requirements；Or for when background is substituted for normal background, judging acquisition Whether background information meets the requirement of corresponding certificate photo；

Bounding box features detection unit, for using edge detecting technology, whether well-regulated detection certificate photograph surrounding frame be special Sign, if nothing, meets the requirements；

Brightness detection unit, for obtaining brightness exception COEFFICIENT K from luminance information₁With average luminance shift value DA, if brightness is different Constant coefficient K₁Less than 1, then brightness is normal, meets the requirements；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is greater than 0, then brightness is excessively bright；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is less than 0, then brightness It crosses dark；

Colorimetric detection units, for obtaining colour cast COEFFICIENT K from chrominance information₂, coloration average value Da and Db, if colour cast COEFFICIENT K₂ Less than 1, then coloration is normal, meets the requirements；If colour cast COEFFICIENT K₂More than or equal to 1, when coloration average value Da is greater than 0, indicate It is partially red, it when coloration average value Da is less than 0, indicates partially green, when coloration average value Db is greater than 0, indicates partially yellow, it is average in coloration When value Db is less than 0, indicate partially blue；

Clarity detection unit is preset for obtaining sharpness factor DR from sharpness information if sharpness factor DR is greater than Sharpness factor value, then explanation is clear, meets the requirements；

Face detection unit, for obtaining eyes, nose, ear, mouth, the crown and chin characteristic from portrait characteristic information According to eyes, nose, ear, mouth, the crown and the chin characteristic that will acquire are compared with preset standard feature, sentence Whether disconnected face are complete, if so, meeting the requirements；

Facial contour detection unit, for obtaining facial contour feature data from portrait characteristic information, detection facial contour is special The diversity factor between data and preset facial contour feature is levied, if diversity factor is greater than preset value, is distorted, if diversity factor Less than or equal to preset value, then meet the requirements；Wherein, between the facial contour feature data and preset facial contour feature Diversity factor pass through the ratio calculation of non-overlap area and overlapping area；

Portrait posture detecting unit, for obtaining eyes, nose, ear, mouth, facial contour, body from portrait characteristic information Dry profile, portrait profile characteristic, judge whether portrait posture is rectified, if so, meeting the requirements；

Head portrait size and location detection unit, for obtaining eyes, ear, the crown and chin characteristic from portrait characteristic information According to calculating eyes, ear, the crown and being associated property of chin characteristic, obtain the wide pixel in head, the high pixel in head, double Eye range pixel, eyes position are gained fame and fortune away from up/down Edge Distance pixel and the crown away from certificate photograph top edge pixel, will These data are compared with the parameter of corresponding certificate photo, judge whether to meet the requirements.

3. a kind of license camera of detectable shooting quality according to claim 2, it is characterised in that: the portrait posture Detection unit includes:

Face state-detection subelement, for judging people according to the eyes of acquisition, nose, mouth and facial contour feature data Whether face is rectified, if so, meeting the requirements；

Shoulder state-detection subelement judges whether shoulder flushes for the trunk contour feature data according to acquisition, if so, Then meet the requirements；

Eyes open width detection subelement, for the eye feature data according to acquisition, calculate and open part among eyes Width meets the requirements if width is greater than preset value；

Mouth closure opens detection sub-unit, for judging mouth using edge detecting technology according to the mouth characteristic of acquisition Whether it is a curve among bar upper lower lip, if so, illustrating that mouth is closed, meets the requirements, if it is not, then illustrating mouth It opens；

Detection sub-unit without a hat on detects portrait profile characteristic and presets for the portrait profile characteristic according to acquisition Portrait profile feature between diversity factor, if diversity factor is less than preset value, and head color without obvious segmentation, then explanation is exempted from Hat, meets the requirements；

Redeye detection subelement calculates the colour cast COEFFICIENT K of eyes for the eye feature data according to acquisition₃With coloration average value Da1, if colour cast COEFFICIENT K₃Less than 1, then eyes coloration is normal, meets the requirements；If colour cast COEFFICIENT K₃More than or equal to 1, and coloration is flat When mean value Da1 is greater than 0, then there is blood-shot eye illness；

Adornment coloured silk detection sub-unit, for according to the facial contour of acquisition, eyes and mouth characteristic, the color for detecting face to be inclined Difference, the color error ratio value of eyes, the color error ratio value at mouth position are preset if the color error ratio value of one of them is greater than Deviation, then explanation have heavy make-up；If the color error ratio value of three is respectively less than or is equal to preset deviation, illustrate without dense Adornment meets the requirements；

Face detection subelement, for obtaining face mask characteristic, detection face mask characteristic and preset face Diversity factor between contour feature, if diversity factor is less than preset value, hair does not cover face, meets the requirements.

4. a kind of license camera of detectable shooting quality according to claim 3, it is characterised in that: the face state Detection sub-unit judges whether face is rectified, and specifically includes:

According to eye feature data, two lines are generated, calculate the angle of two lines and certificate photograph horizontal direction, judgement should Whether angle exceeds preset range, if it is not, then explanation faces camera；

According to eyes, nose and mouth characteristic, the center point coordinate of two lines is obtained, obtains the centre coordinate of nose, Three points are done straight line fitting and obtain straight line LC, calculate straight line LC and certificate photograph Vertical Square by the centre coordinate for obtaining mouth To angle, judge whether the angle exceeds preset range, if it is not, then illustrating that head is vertical；

According to facial contour feature data, the center line FLC of human face region vertical direction is obtained, calculates human face region vertical direction Center line FLC and straight line LC distance DC, calculate the ratio of distance DC and human face region width, judge whether the ratio surpasses Preset range out, if it is not, then illustrating face horizontal center；

5. a kind of license camera of detectable shooting quality according to claim 3, it is characterised in that: the shoulder state Detection sub-unit judges whether shoulder flushes, and specifically includes:

To trunk profile coordinate array, be fitted to a polygon, calculate polygon center of gravity obtain the coordinate of center of gravity (X1, Y1), the distance Dweight of X1 and certificate photograph vertical center of gravity line are calculated, calculates distance Dweight's and certificate photograph width WF Ratio, judges whether the ratio exceeds preset range；

One minimum rectangle is obtained using minimum circumscribed rectangle algorithm to trunk profile coordinate array, calculates rectangular vertical center of gravity The angle of line and certificate photograph vertical center of gravity line, judges whether the angle exceeds preset range；

If the ratio of distance Dweight and certificate photograph width WF, rectangular vertical center of gravity line and certificate photograph vertical center of gravity line Angle is all without departing from preset range, then shoulder flushes, and meets the requirements.

6. a kind of shooting quality detection method is applied in license camera, it is characterised in that: the described method includes:

The license camera obtains file attribute information, portrait characteristic information, background letter according to the certificate photograph of shooting imaging Breath, luminance information, chrominance information and sharpness information, specifically include:

Obtain the file attribute information of certificate photograph；Wherein, the file attribute information includes shooting time, colored digit, text Part suffix, file size and image pixel；

Obtain the portrait characteristic information of certificate photograph；Wherein, the portrait characteristic information include eyes, nose, ear, mouth, Facial contour, trunk profile, portrait profile, the crown and chin characteristic；

When background is not replaced, certificate photograph is converted into Lab color space image, to Lab color space image, reads prospect Each pixel in region, establishes the mixed Gauss model of prospect coloration；To each pixel of background area, background coloration is obtained Histogram calculates the overlapping possibility of the histogram of background coloration and the mixed Gauss model of prospect coloration, takes mark to overlapping possibility Quasi- mean-square value P；Or when background is substituted for normal background, portrait profile is detected using edge detecting technology to certificate photograph, Obtain the background information other than certificate photograph portrait profile；

The brightness histogram for obtaining certificate photograph carries out data analysis to brightness histogram, calculates brightness exception COEFFICIENT K₁With it is average Intensity deviation value DA；

Certificate photograph is converted into Lab color space image, counts each pixel in a axis and b axis of Lab color space image Coloration average value Da and Db calculate colour cast COEFFICIENT K according to coloration average value Da and Db₂；

Certificate photograph is converted into gray level image, calculates sharpness factor DR；

The license camera is believed according to the file attribute information of acquisition, portrait characteristic information, background information, luminance information, coloration Breath and sharpness information carry out quality testing item by item to every dimensions；Wherein, every dimensions are according to government's public security The examination criteria of industry standard formulation, comprising: whether be recent photochrome, whether not less than 24 RGB true color, whether be JPEG compression technology and compression whether quality meets the requirements, whether file size meets the requirements, the wide pixel of image image height whether Meet the requirements, whether background color meets the requirements, photo surrounding whether Rimless, whether image clear, whether face visible, people Face profile whether without obvious distortion, brightness whether uniformly sufficient, image whether no color differnece, posture whether rectify, head portrait size and Whether position meets the requirements；

The license camera comments the quality measurements of every dimensions according to the default score value of every dimensions Point, obtain the score value of every dimensions；

The license camera calculates the quality general comment of certificate photograph according to the score value and scoring weight of every dimensions Point；

7. a kind of shooting quality detection method according to claim 6, it is characterised in that: the license camera is according to acquisition File attribute information, portrait characteristic information, background information, luminance information, chrominance information and sharpness information, to each item rating Dimension carries out quality testing item by item, specifically includes:

Shooting time is obtained from file attribute information, judges whether to shoot in preset recent times, if so, judgement is No is photochrome, if so, meeting the requirements；

File suffixes is obtained from file attribute information, judges whether file is compressed file, when file is compressed file, is sentenced Whether disconnected is JPEG compression, if so, judging to compress the requirement whether quality meets corresponding certificate photo；

File size is obtained from file attribute information, judges whether file size is greater than the preset value of corresponding certificate photo, if so, Then meet the requirements；

When background is not replaced, standard mean-square value P is obtained from background information, whether judgment criteria mean-square value P is greater than default equal Side's value, if it is not, then meeting the requirements；Or for when background is substituted for normal background, judging whether the background information obtained meets The requirement of corresponding certificate photo；

Using edge detecting technology, detects the whether well-regulated bounding box features of certificate photograph surrounding and meet the requirements if nothing；

Brightness exception COEFFICIENT K is obtained from luminance information₁With average luminance shift value DA, if brightness exception COEFFICIENT K₁Less than 1, then Brightness is normal, meets the requirements；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is greater than 0, then brightness It crosses bright；If brightness exception COEFFICIENT K₁More than or equal to 1, and average luminance shift value DA is less than 0, then too dark brightness；

Colour cast COEFFICIENT K is obtained from chrominance information₂, coloration average value Da and Db, if colour cast COEFFICIENT K₂Less than 1, then coloration is normal, It meets the requirements；If colour cast COEFFICIENT K₂More than or equal to 1, when coloration average value Da is greater than 0, indicate partially red, in coloration average value It when Da is less than 0, indicates partially green, when coloration average value Db is greater than 0, indicates partially yellow, when coloration average value Db is less than 0, indicate It is partially blue；

Sharpness factor DR is obtained from sharpness information, if sharpness factor DR is greater than default sharpness factor value, is illustrated Clearly, it meets the requirements；

Acquisition eyes, nose, ear, mouth, the crown and chin characteristic from portrait characteristic information, the eyes that will acquire, Nose, ear, mouth, the crown and chin characteristic, are compared with preset standard feature, judge whether face are complete, If so, meeting the requirements；

Facial contour feature data, detection facial contour feature data and preset facial contour are obtained from portrait characteristic information Diversity factor between feature is distorted if diversity factor is greater than preset value, if diversity factor is less than or equal to preset value, is accorded with It closes and requires；Wherein, the diversity factor between the facial contour feature data and preset facial contour feature passes through non-overlap face Long-pending and overlapping area ratio calculation；

Eyes, nose, ear, mouth, facial contour, trunk profile, portrait profile characteristic are obtained from portrait characteristic information According to judging whether portrait posture is rectified, if so, meeting the requirements；

Eyes, ear, the crown and chin characteristic are obtained from portrait characteristic information, to eyes, ear, the crown and lower Bart Levy being associated property of data calculate, obtain the wide pixel in head, the high pixel in head, eyes range pixel, eyes position away from it is upper/ Lower edge range pixel and the crown are gained fame and fortune away from certificate photograph top edge pixel, by these data and the parameter of corresponding certificate photo into Row compares, and judges whether to meet the requirements.

8. a kind of shooting quality detection method according to claim 7, it is characterised in that: described whether to judge portrait posture Rectify, specifically include:

According to the eyes of acquisition, nose, mouth and facial contour feature data, judge whether face is rectified, if so, conforming to It asks；

According to the eye feature data of acquisition, the width for opening part among eyes is calculated, if width is greater than preset value, is met It is required that；

According to the mouth characteristic of acquisition, using edge detecting technology, judge on mouth among lower lip whether to be a song Line meets the requirements if so, illustrating that mouth is closed, if it is not, then illustrating that mouth opens；

According to the portrait profile characteristic of acquisition, detect between portrait profile characteristic and preset portrait profile feature Diversity factor, if diversity factor is less than preset value, and head color without obvious segmentation, then explanation is without a hat on, meets the requirements；

According to the eye feature data of acquisition, the colour cast COEFFICIENT K of eyes is calculated₃With coloration average value Da1, if colour cast COEFFICIENT K₃It is small In 1, then eyes coloration is normal, meets the requirements；If colour cast COEFFICIENT K₁More than or equal to 1, and coloration average value Da1 be greater than 0 when, then There are blood-shot eye illness；

According to the facial contour of acquisition, eyes and mouth characteristic, the color error ratio value of face, the color of eyes are detected Deviation, the color error ratio value at mouth position illustrate to have dense if the color error ratio value of one of them is greater than preset deviation Adornment；If the color error ratio value of three is respectively less than or is equal to preset deviation, illustrates no heavy make-up, meet the requirements；

Face mask characteristic is obtained, the difference between face mask characteristic and preset face mask feature is detected Degree, if diversity factor is less than preset value, hair does not cover face, meets the requirements.

9. a kind of shooting quality detection method according to claim 8, it is characterised in that: described to judge whether face is held Just, it specifically includes:

10. a kind of shooting quality detection method according to claim 8, it is characterised in that: described to judge whether shoulder is neat It is flat, it specifically includes: