The content of the invention
In view of the foregoing, it is necessary to provide a kind of jump portrait system and method, it can detecting real-time electronics dress
The face location of personage in the image of shooting is put, when the average height of all face locations is more than threshold values set in advance, from
The dynamic shutter that starts is shot, and obtains high-quality jump photo.
One kind jump portrait system, the system includes:Photo acquisition module, for obtaining what is installed on electronic installation
First scene image of image unit intake;Face recognition module, for detecting face area in the first scene image for obtaining
Domain;Human face analysis module, for obtaining the center point coordinate of all human face regions for detecting, and calculates the vertical of all central points
The average value of axial coordinate;Control module, for when the ordinate of orthogonal axes average value that this is calculated is more than pre-set threshold value, starting shooting
The shutter of unit is shot, and obtains the second scene image.
One kind jump portrait method, the method includes:Photo obtaining step, obtains the shooting installed on electronic installation
First scene image of unit intake;Recognition of face step, human face region is detected in the first scene image for obtaining;Face point
Analysis step, obtains the center point coordinate of all human face regions for detecting, and calculates the average of the ordinate of orthogonal axes of all central points
Value;Rate-determining steps, when the ordinate of orthogonal axes average value that this is calculated is more than pre-set threshold value, the shutter for starting image unit is clapped
Take the photograph, obtain the second scene image.
Compared to prior art, described jump portrait system and method, it can the shooting of detecting real-time electronic installation
Image in personage face location, when all face locations average height be more than threshold values set in advance when, automatic
Shutter is shot, and obtains high-quality jump photo, it is to avoid between photographer and the person of being taken because of tacit deficiency caused by
Personage's jump has the problem of difference of height in shooting picture.
Brief description of the drawings
Fig. 1 is present invention jump portrait systematic difference environment schematic.
Fig. 2 is the functional block diagram of present invention jump portrait system.
Fig. 3 is the flow chart of present invention jump portrait method preferred embodiment.
Fig. 4 is to set the schematic diagram for shooting datum line.
Fig. 5 is the schematic diagram that human face region is detected in the first scene image.
Fig. 6 is the schematic diagram of the Y-axis coordinate average value of the central point for calculating all human face regions.
Fig. 7 and Fig. 8 are that the Y-axis coordinate average value that will be calculated in Fig. 6 is carried out with the Y-axis coordinate of default shooting datum line
The schematic diagram of comparison.
Fig. 9 to Figure 13 is a mode of operation schematic diagram for application example of the invention.Main element symbol description
Electronic installation |
2 |
Image unit |
20 |
Display screen |
22 |
Memory |
23 |
Jump portrait system |
24 |
Processor |
25 |
Datum line setup module |
240 |
Photo acquisition module |
241 |
Face recognition module |
242 |
Human face analysis module |
243 |
Control module |
244 |
Specific embodiment
As shown in figure 1, being present invention jump portrait systematic difference environment schematic.In the present embodiment, the jump
Jump portrait system 24 is run in an electronic installation 2.The electronic installation 2 also includes passing through holding wire and data wire phase
Image unit 20 even, display screen 22, memory 23 and processor 25.In the present embodiment, the electronic installation 2 can be
Smart mobile phone, digital camera and PDA(Personal Digital Assistant, personal digital assistant)Deng.
The jump portrait system 24 is used for people in the image that the image unit 20 of detecting real-time electronic installation 2 shoots
The face location of thing, when the average height of all face locations is more than threshold values set in advance, automatic image unit 20
Shutter shot, obtain high-quality jump photo, below detailed process describe.
In the present embodiment, the jump portrait system 24 can be divided into one or more modules, described one
Individual or multiple modules are stored in the memory 23 and are configured to by one or more processors(The present embodiment is one
Processor 25)Perform, to complete the present invention.For example, refering to shown in Fig. 2, the jump portrait system 24 is divided into base
Directrix setup module 240, photo acquisition module 241, face recognition module 242, human face analysis module 243 and control module 244.
Module alleged by the present invention be complete a specific function program segment, than program more suitable for description software in processor 25
Implementation procedure, the function on each module will be specifically described in the flowchart of fig. 3.
Before Fig. 3 flow charts are introduced, user sets a bat by datum line setup module 240 on display screen 22
Take the photograph datum line(Refering to shown in Fig. 9).With continued reference to shown in Fig. 4, square region represents the scope of current scene image, current scene
The transverse axis of image represents that the longitudinal axis is represented with Y-axis with X-axis, after user sets the shooting datum line on display screen 22, benchmark
Line setup module 240 will obtain the Y-axis coordinate Yc of the shooting datum line, as the shooting datum line in current scene image
Y-axis is highly stored in memory 23.
As shown in figure 3, being the flow chart of present invention jump portrait method preferred embodiment.
Step S10, photo acquisition module 241 is every Preset Time(Such as 0.1 second), that is installed on acquisition electronic installation 2 takes the photograph
As the first scene image of the intake of unit 20(Refering to shown in Figure 10).
Step S11, face recognition module 242 detects human face region in the first scene image for obtaining(Refering to Figure 11 institutes
Show).In the present embodiment, the face recognition module 242 can be theoretical using the colour of skin(Such as YCbCr color space models)From
All of human face region is identified in one scene image, it is also possible to utilize face feature sample comparison technology(FaceTemplate
Matching)Or other facial recognition techniques identify all human face regions from the first scene image.Wherein, the face area
Domain can be a minimum rectangle for surrounding face.
For example, refering to shown in Fig. 5, face recognition module 242 identified from the first scene image three human face region A,
B, C, wherein, the center point coordinate of human face region A is(X0, Y0), the center point coordinate of human face region B is(X1, Y1), face area
The center point coordinate of domain C is(X2, Y2).Wherein, the Y-axis coordinate of a face regional center point represents the human face region first
Y-axis height value in scene image.
Step S12, face recognition module 242 judges whether to detect human face region in the first scene image.If
Human face region is detected in first scene image, then performs step S13.If not detecting face in the first scene image
Region, then return to step S10.
Step S13, human face analysis module 243 obtains the center point coordinate of each human face region, and calculates all central points
Y-axis coordinate average value Yavg(Refering to shown in Figure 11).
By taking the coordinate of the human face region central point obtained in Fig. 5 as an example, human face analysis module 243 obtains human face region A's
The Y-axis coordinate Y0 of central point, the Y-axis coordinate Y1 of the central point of human face region B, the Y-axis coordinate Y2 of the central point of human face region C,
Then Yavg=(Y0+Y1+Y2)/3(Refering to shown in Fig. 6).
Step S14, the Y-axis coordinate average value Y that human face analysis module 243 calculates thisavgWith the shooting base for pre-setting
The Y-axis coordinate Y of directrixcCompare, judge YavgWhether Y is more thanc.If YavgLess than or equal to Yc(Refering to Fig. 7 and Figure 12 institutes
Show), then judge also have personage not takeoff in the first scene image, flow return to step S10;If YavgMore than Yc(Refering to figure
Shown in 8 and Figure 13), then judge that all persons in the first scene image have all takeoff, perform step S15.
Step S15, the shutter that control module 244 starts image unit 20 is shot, and obtains the second scene image.Its
In, all persons in second scene image have all takeoff, and reach the best time for catching and shooting jump photo.
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention and it is unrestricted, although reference
Preferred embodiment has been described in detail to the present invention, it will be understood by those within the art that, can be to of the invention
Technical scheme is modified or equivalent, without deviating from the spirit and scope of technical solution of the present invention.