CN103905737B - Backlighting detecting and device - Google Patents
Backlighting detecting and device Download PDFInfo
- Publication number
- CN103905737B CN103905737B CN201210572564.3A CN201210572564A CN103905737B CN 103905737 B CN103905737 B CN 103905737B CN 201210572564 A CN201210572564 A CN 201210572564A CN 103905737 B CN103905737 B CN 103905737B
- Authority
- CN
- China
- Prior art keywords
- backlight
- gray scale
- environment
- gray
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
There is provided a kind of backlighting detecting and device.The backlighting detecting includes：Gather an image；Extract the grey level histogram of described image；And determine whether current environment is backlight environment according to the grey level histogram.Backlight detection is simply carried out based on acquired image in the case of without CMOS backlight detection sensors, so as to advantageously reduce the cost of backlight detection, and and then the cost of the electronic equipment using the backlighting detecting and device is reduced.
Description
Technical field
The present invention relates to backlight environment measuring, and relate more specifically to a kind of backlighting detecting and device.
Background technology
At present, the application of digital camera is more and more extensive.In addition to the digital camera product of specialty, set in numerous electronics
The standby upper digital camera for being integrated with different accuracy.For example, digital camera head is integrated with mostly in notebook computer,
In personal digital assistant（PDA）Be integrated with digital camera mostly in smart mobile phone, digital phase is also integrated with tablet personal computer
Machine.
When carrying out image taking using digital camera, inevitably run into backlight scene and shoot and non-backlight scene
The problem of shooting., it is necessary to adjust the acquisition parameters of digital camera, to suppress, subject is excessively dark and background is excessively bright under backlight scene.
The acquisition parameters generally comprise F-number, time for exposure etc..
CMOS backlight detection sensors are frequently included in existing digital camera, and biography is detected using the CMOS backlight
Sensor carrys out the shooting state of automatic detection backlight., can be with adjust automatically acquisition parameters when digital camera detects backlight environment
It is excessively bright to suppress subject dark and background excessively.
As digital camera is applied in various electronic equipments more and more, in order to reduce electronic equipment as much as possible
Manufacturing cost, the cost for how reducing digital camera is also increasingly taken seriously.Therefore, the present invention desirable to provide one kind in nothing
The method for needing to carry out backlight environment estimation in the case of CMOS backlight detection sensors, so that in digital camera or even various electronics
Even if also not resulting in high-quality shooting image including CMOS backlight detection sensor in equipment.
The content of the invention
In view of above mentioned problem, the present invention has been made.The present invention is intended to provide a kind of backlighting detecting and device, its
Backlight detection is directly carried out according to the image currently gathered in the case of without CMOS backlight detection sensors.As a result,
May be according to the backlight testing result adjust automatically acquisition parameters.
According to an aspect of the present invention there is provided a kind of backlighting detecting, including：Gather an image；Extract described image
Grey level histogram；And determine whether current environment is backlight environment according to the grey level histogram.
Preferably, in the backlighting detecting, determine whether current environment is backlight according to the grey level histogram
Environment includes：Calculate the variance of the grey level histogram of described image；And determine whether current environment is inverse according to the variance
Luminous environment.
Preferably, in the backlighting detecting, determine whether current environment is backlight according to the grey level histogram
Environment includes：Determine whether current environment is backlight ring according to the local peaking of gray probability distribution in the grey level histogram
Border, wherein, the gray scale at the place of local peaking being distributed in the gray probability is less than the first gray scale or during more than the second gray scale, really
Current environment is determined for backlight environment；And the gray scale at the local peaking that the gray probability is distributed is more than the 3rd gray scale and small
When four gray scales, it is non-backlight environment to determine current environment, wherein, first gray scale is less than or equal to the 3rd gray scale, described
3rd gray scale is less than the 4th gray scale, and the 4th gray scale is less than or equal to second gray scale.
Preferably, in the backlighting detecting, the image is coloured image, extracts the grey level histogram bag of the image
Include：Convert the image to gray level image；And extract the grey level histogram of the gray level image.
Preferably, in the backlighting detecting, determine whether current environment is backlight environment bag according to the variance
Include：Backlight degree is determined according to the variance；And when the backlight degree is more than predetermined threshold, it is backlight ring to determine current environment
Border.
Preferably, in the backlighting detecting, determine whether current environment is backlight environment bag according to the variance
Include：Backlight degree is determined according to the variance；And the probability that current environment is backlight environment is determined according to the backlight degree.
Preferably, in the backlighting detecting, according to the variance determine current environment whether be backlight environment also
Including：When identified probability is more than predetermined probability threshold value, it is backlight environment to determine current environment.
Preferably, in the backlighting detecting, methods described also includes：By the probability with passing through another method meter
The backlight ambient probability combination of calculation, to determine whether current environment is backlight environment.
According to a further aspect of the invention there is provided a kind of backlight detection means, including：Image acquisition component, for obtaining
Image, the image is the image shot under the present circumstances；Histogram extracting parts, the intensity histogram for extracting described image
Figure；And informating part, for determining whether the current environment is backlight environment according to the grey level histogram.
Preferably, the informating part includes：Variance calculating unit, the side of the grey level histogram for calculating described image
Difference；And environment determines part, for determining whether current environment is backlight environment according to the variance.
Preferably, the informating part determines to work as according to the local peaking of gray probability distribution in the grey level histogram
Whether preceding environment is backlight environment, wherein, the gray scale at the place of local peaking being distributed in the gray probability less than the first gray scale or
When person is more than the second gray scale, the informating part determines that current environment is backlight environment；And be distributed in the gray probability
When gray scale at local peaking is more than the 3rd gray scale and is less than four gray scales, the informating part determines that current environment is non-backlight
Environment, wherein, first gray scale is less than or equal to the 3rd gray scale, and the 3rd gray scale is less than the 4th gray scale, the 4th gray scale
Less than or equal to second gray scale.
Preferably, the backlight detection means also includes：Image converting member, for when described image is coloured image
The coloured image is converted into gray level image, wherein, the gray scale that the histogram extracting parts extracts the gray level image is straight
Fang Tu.
Preferably, the environment determines that part includes：Backlight degree calculating unit, for determining backlight according to the variance
Degree, and backlight determine part, for determining that current environment is backlight environment when the backlight degree is more than predetermined threshold.
Preferably, the environment determines that part includes：Backlight degree calculating unit, for determining backlight according to the variance
Degree；Backlight ambient probability calculating unit, for determining probability of the current environment for backlight environment according to the backlight degree；And it is inverse
Light determines part, for determining that current environment is backlight environment when identified probability is more than predetermined probability threshold value.
Preferably, the environment determines that part includes：Backlight degree calculating unit, for determining backlight according to the variance
Degree；Backlight ambient probability calculating unit, for determining probability of the current environment for backlight environment according to the backlight degree；And it is inverse
Light determines part, for identified probability to be combined with the backlight ambient probability calculated by another method, current to determine
Whether environment is backlight environment.
Preferably, in the backlighting detecting and device, the grey level histogram includes multiple gray scales and each
Pixel quantity corresponding to gray scale, or the grey level histogram are normalized grey level histogram, the normalized gray scale
Histogram includes the probability of occurrence corresponding to multiple gray scales and each gray scale.
, can be without CMOS backlight detection sensors using backlighting detecting and device according to embodiments of the present invention
In the case of backlight detection is simply carried out based on acquired image so that advantageously reduce backlight detection cost, and
And then reduce the cost of the electronic equipment using the backlighting detecting and device.
Brief description of the drawings
According to embodiments of the present invention by reference to accompanying drawing description, the various features and advantage of the embodiment of the present invention will be brighter
It is aobvious, and be also easier to be understood, in the accompanying drawings：
The flow chart of backlighting detecting according to embodiments of the present invention Fig. 1；
Fig. 2 shows the specific of the first sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart；
Fig. 3 shows the specific of second of sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart；
Fig. 4 shows the specific of the third sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart；
Fig. 5 shows the specific of the 4th kind of sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart；
Fig. 6 shows the schematic block diagram of backlight detection means according to embodiments of the present invention；
Fig. 7 A show the grey level histogram distribution under exemplary backlight environment；
Fig. 7 B show the grey level histogram distribution under exemplary non-backlight environment；And
Fig. 8 shows the curve map of exemplary backlight degree and backlight ambient probability.
Embodiment
Backlighting detecting and electronic equipment according to embodiments of the present invention described below with reference to the accompanying drawings.
In Fig. 7 A and Fig. 7 B, grey level histogram distribution and the non-backlight ring under exemplary backlight environment are respectively illustrated
Grey level histogram distribution under border, wherein transverse axis represents gray scale.
Under a kind of sample situation, grey level histogram can be embodied between each gray scale and pixel count corresponding to each gray scale
Relation histogram, in the case, grey level histogram include multiple gray scales（Such as 128,256,512 gray scales
Deng）And the pixel quantity corresponding to each gray scale.In the case, Fig. 7 A and Fig. 7 B longitudinal axis represent pixel quantity.
Under another sample situation, grey level histogram can be normalized grey level histogram, the normalized ash
The relation between probability of occurrence corresponding to degree histogram each gray scale of embodiment and each gray scale, in the case, grey level histogram bag
Include multiple gray scales（Such as 128,256,512 gray scales etc.）And the probability of occurrence corresponding to each gray scale, it is described go out
Existing probability can be the ratio between pixel count and total pixel number corresponding to each gray scale.In the case, Fig. 7 A and Fig. 7 B longitudinal axis
Represent the probability of occurrence of each gray scale.
As can be seen that the grey level histogram distribution under backlight environment and the gray scale under non-backlight environment from Fig. 7 A and Fig. 7 B
Histogram distribution is entirely different.As shown in Figure 7 A, the characteristics of grey level histogram under backlight environment is distributed is：It is distributed in incandescent
It is many with pixel in very dark gray scale, and it is relatively fewer to be distributed in the pixel in middle gray.In contrast, as shown in Figure 7 B, it is non-
Grey level histogram distribution under backlight environment is characterised by：Pixel is concentrated mainly in middle gray, and be distributed in incandescent and
Pixel in very dark gray scale is considerably less.As can be seen here, the grey level histogram distribution under backlight environment and the ash under non-backlight environment
There is significant difference in degree histogram distribution.
Although only giving grey level histogram distribution and the non-backlight under exemplary backlight environment in Fig. 7 A and Fig. 7 B
Grey level histogram distribution under environment, and for different photographed scenes, the grey level histogram distribution under backlight environment may
It is not quite similar, and for different photographed scenes, the grey level histogram distribution under non-backlight environment is also not quite similar, but nothing
By being that grey level histogram distribution under backlight environment is also grey level histogram distribution under non-backlight environment, good system is respectively provided with
Count characteristic.That is, the characteristics of grey level histogram under backlight environment is distributed is：It is distributed in the pixel on incandescent and very dark gray scale
It is many, and it is relatively fewer to be distributed in the pixel in middle gray；Grey level histogram distribution under non-backlight environment is characterised by：Picture
Element is concentrated mainly in middle gray, and the pixel being distributed on incandescent and very dark gray scale is considerably less.
In view of above-mentioned characteristic, a kind of method based on image to carry out backlight detection in itself is inventors herein proposed.
First, backlighting detecting 100 according to embodiments of the present invention will be described with reference to Figure 1.According to present invention implementation
The backlighting detecting 100 of example starts in step S101.
In step S110, image is gathered.Gathering image can include gathering image with exposal model, or with image pickup mode
Image is gathered, or image is gathered with preview mode.In addition, collection image can include storage image or non-storage image.Example
Such as, collection image can be included in focused when taking pictures during gather image and store or not storage image, can include
Image is gathered under preview mode and is stored or not storage image, can be included under image pickup mode and be gathered image and store
Image.Gather image in addition, collection image can be included under default backlight detection pattern and store or not storage image.
In step S120, the grey level histogram of acquired image is extracted.Acquired image can be gray level image or
Coloured image.In the case where acquired image is coloured image, acquired image first can be converted into gray level image.
Then, then based on gray level image extract grey level histogram.It is people in the art how coloured image to be converted into gray level image
Member is known, therefore omits its specific descriptions herein.
How to travel through each pixel of acquired image with extract grey level histogram be also well known to a person skilled in the art, because
This also omits its specific descriptions herein.
In step S130, determine whether current environment is backlight environment according to the grey level histogram.Specifically, according to institute
The feature of grey level histogram is stated to determine whether current environment is backlight environment.
As it was previously stated, the grey level histogram under backlight environment is the characteristics of distribution：It is distributed on incandescent and very dark gray scale
Pixel it is many, and it is relatively fewer to be distributed in the pixel in middle gray；The feature of grey level histogram distribution under non-backlight environment
It is：Pixel is concentrated mainly in middle gray, and the pixel being distributed on incandescent and very dark gray scale is considerably less.
In addition, in terms of statistical property, be distributed for the grey level histogram under backlight environment, the distribution of its grey level histogram
Gray variance σ2It is very big；And be distributed for the grey level histogram under non-backlight environment, the gray variance σ of its grey level histogram distribution2
Very little.
Then, in step S199, backlighting detecting 100 according to embodiments of the present invention terminates.
Assuming that acquired image I normalized grey level histogram is X=(x1,x2,…,x256)T, wherein xiRepresent i-th
The probability that individual gray scale occurs in acquired image I.
Next, based on the normalized grey level histogram, backlight according to embodiments of the present invention will be illustrated and detected
Step S130 specific implementation in method 100.
The tool of the first sample implementation of step S130 in Fig. 1 according to embodiments of the present invention is shown in Fig. 2
Body flow chart.
In the first sample implementation, the local peaking being distributed according to gray probability in the grey level histogram is come really
Whether determine current environment is backlight environment.
In step S2310, the local peaking that gray probability is distributed in the grey level histogram is determined.Moreover, correspondingly true
Gray scale corresponding to the fixed local peaking.With reference to Fig. 7 A, identified local peaking can be corresponding to such as gray value 3
Gray probability and the gray probability for example corresponding to gray value 250.
Then, in step S2320, determine whether the gray scale at the local peaking of the gray probability distribution is less than the first ash
Spend or more than the second gray scale.
Gray scale at the local peaking for determining the gray probability distribution in step S2320 is less than the first gray scale or big
When the second gray scale, in step S2330, it is backlight environment to determine current environment.For example, still with Fig. 7 A, first gray scale
Can be, for example, gray value 30, second gray scale can be such as gray value 180, identified local peaking（Gray value 3）
Less than first gray scale, and identified local peaking（Gray value 250）More than second gray scale, therefore, the image
Current environment be confirmed as backlight environment.
When the gray scale at the place of local peaking that gray probability distribution is determined in step S2320 be not less than the first gray scale and
During no more than the second gray scale, the backlighting detecting proceeds to step S2340.
In step S2340, determine the gray probability distribution place of local peaking gray scale whether more than the 3rd gray scale and
Less than the 4th gray scale.
Gray scale at the local peaking for determining the gray probability distribution in step S2340 is more than the 3rd gray scale and is less than
During four gray scales, in step S2350, it is non-backlight environment to determine current environment, for example, with reference to Fig. 7 B, the 3rd gray scale can
To be, for example, gray value 80, the 4th gray scale can be such as gray value 180, identified local peaking（Gray value 100）
More than the 3rd gray scale and less than the 4th gray scale, therefore, the current environment of the image is confirmed as non-backlight environment.
Gray scale at the local peaking for determining gray probability distribution in step S2340 is not more than the 3rd gray scale and not
During less than four gray scales, the backlighting detecting can be back to step S110 and resurvey image, or use other backlight
Detection method carries out backlight detection.
First gray scale is less than or equal to the 3rd gray scale, and the 3rd gray scale is less than the 4th gray scale, and the 4th gray scale is small
In equal to second gray scale.
For first gray scale less than the gray scale at the 3rd gray scale and identified local peaking between described
Situation between first gray scale and the 3rd gray scale, can resurvey image and re-start backlight detection, Huo Zheke
To carry out backlight detection using other methods.
Similarly, the gray scale at second gray scale and identified local peaking is less than for the 4th gray scale to be situated between
Situation between the 4th gray scale and second gray scale, can resurvey image and re-start backlight detection,
Or backlight detection can be carried out using other methods.
Fig. 3 shows the specific of second of sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart.
Whether in second of sample implementation, it is backlight ring that current environment is determined according to the variance of grey level histogram
Border.
For the normalized grey level histogram X=(x1,x2..., x256)T, the population mean μ and variance of intensity profile
σ2It can be represented as：
Variance threshold values can be set, when the variance of the grey level histogram extracted is more than the variance threshold values, it is determined that currently
Environment is backlight environment, otherwise current environment is non-backlight environment.The variance threshold values can according to experiment, rule of thumb, root
Set according to personal like etc..
In step S3310, the variance of extracted grey level histogram is calculated as described above.
Then, in step S3320, determine whether current environment is backlight environment according to the variance calculated.
Fig. 4 shows the specific of the third sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart.
In the third sample implementation, backlight degree can be calculated based on variance, and can define backlight degree threshold value,
Determine whether current environment is backlight environment based on the relation between the backlight degree and the backlight degree threshold value calculated.
In step S4310, the variance of extracted grey level histogram is calculated as described above.
Then, in step S4320, backlight degree is determined according to the variance calculated.Backlight degree can represent present image
Backlight degree.Backlight degree can be the function of the variance or the variance., can be by backlight as an example
Spend dBLIt is defined as：
dBL=f(X)=σ
In step S4330, judge whether calculated backlight degree is more than predetermined backlight degree threshold value.
When judging that calculated backlight degree is more than predetermined backlight degree threshold value in step S4330, determined in step S4340
Current environment is backlight environment.
It is true in step S4350 when judging that calculated backlight degree is not more than predetermined backlight degree threshold value in step S4330
Current environment is determined for non-backlight environment.
In order to reduce False Rate, the 4th kind of sample implementation is further provided.
Fig. 5 shows the specific of the 4th kind of sample implementation of the step S130 in Fig. 1 according to embodiments of the present invention
Flow chart.
In the 4th kind of sample implementation, backlight degree can be calculated based on variance, and it is true based on the backlight degree calculated
Determine backlight ambient probability, then determine whether current environment is backlight environment based on backlight ambient probability again.
Consider to build following sample (dBL, Y), wherein Y only takes positive and negative 1 (± 1), and positive 1 (+1) represents backlight, minus 1 (- 1) table
Show non-backlight, dBLFor backlight degree defined above.
Returned by gathering multiple samples, and binary Logistic being using these samples, training obtains Sigmoid forms
Probability function：
The Sigmoid form probability functions are represented when backlight degree is dBLWhen, current environment is the probability of backlight environment.By
This, can be by backlight degree dBLBe converted to the probability that acquired image I is under backlight environment.It is then possible to pass through the probability
Come decision diagram as I whether backlight.
Figure 8 illustrates the transformation curve, wherein transverse axis represents backlight degree dBL, the longitudinal axis represents backlight ambient probability.When
When backlight degree is more than some threshold value T, P (Y=1) is more than 0.5, and when backlight degree is less than T, P (Y=1) is less than 0.5.
In step S5310, the variance of extracted grey level histogram is calculated as described above.
In step S5320, backlight degree is determined according to the variance calculated as described above.
Then, in step S5330, according to the backlight degree determine current environment whether be backlight environment probability, i.e. root
Backlight ambient probability is determined according to the backlight degree.
Can be such as formula（3）It is shown or as illustrated in fig. 8, backlight ambient probability is determined according to backlight degree.
In step S5340, determine whether current environment is backlight environment according to identified backlight ambient probability.
In one example, backlight ambient probability threshold value is set（For example, 0.7）, it is big in identified backlight ambient probability
When the backlight ambient probability threshold value, it is backlight environment to determine current environment.Conversely, determining that current environment is non-backlight environment.
In another example, in step S5340, can combine from another backlighting detecting calculated it is inverse
Luminous environment probability or other parameters, and the backlight ambient probability calculated based on backlighting detecting according to embodiments of the present invention
Final backlight detection is carried out with the testing result from another backlighting detecting, so as to improve accuracy of detection.
Next, backlight detection means 600 according to embodiments of the present invention will be described with reference to Fig. 6.
Figure 6 illustrates the schematic block diagram of backlight detection means 600 according to embodiments of the present invention.
As shown in fig. 6, the backlight detection means 600 include image acquisition component 610, histogram extracting parts 620 and
Informating part 630.
The image that image acquisition component 610 is shot under the present circumstances from shooting component retrieval.
Histogram extracting parts 620 extracts the grey level histogram of described image.
Informating part 630 determines whether the current environment is backlight environment according to the grey level histogram.
In the first sample implementation, the informating part 630 is according to gray probability in the grey level histogram point
The local peaking of cloth determines whether current environment is backlight environment.
Specifically, the gray scale at the local peaking that the gray probability is distributed is less than the first gray scale or more than the second ash
When spending, the informating part 630 determines that current environment is backlight environment；Ash at the local peaking that the gray probability is distributed
When degree is more than the 3rd gray scale and is less than four gray scales, the informating part 630 determines that current environment is non-backlight environment, wherein,
First gray scale is less than or equal to the 3rd gray scale, and the 3rd gray scale is less than the 4th gray scale, and the 4th gray scale is less than or equal to institute
State the second gray scale.
In second of sample implementation, the informating part 630 can include variance calculating unit and environment is determined
Part.
Variance calculating unit calculates the variance of the grey level histogram of described image as described above.
Environment determines that part determines whether current environment is backlight environment according to the variance.
Specifically, when the variance of the grey level histogram extracted is more than pre-determined variance threshold value, the environment determines part
It is backlight environment to determine current environment.
In the third sample implementation, the informating part 630 can include variance calculating unit and environment is determined
Part, and the environment determines that part can further include backlight degree calculating unit and backlight determines part.
Backlight degree calculating unit determines backlight degree according to the variance as described above.It is more than predetermined threshold in the backlight degree
During value, the backlight determines that part determines that current environment is backlight environment.
In the 4th kind of sample implementation, the informating part 630 can include variance calculating unit and environment is determined
Part, and the environment determine part further can include backlight degree calculating unit, backlight ambient probability calculating unit and
Backlight determines part.
Backlight degree calculating unit determines backlight degree according to the variance as described above.
Backlight ambient probability calculating unit determines that current environment is backlight environment according to the backlight degree as described above
Probability.
When identified probability is more than predetermined probability threshold value, the backlight determines that part determines that current environment is backlight ring
Border.
Alternatively, the backlight determines that part can also be by identified probability and the backlight ring calculated by another method
Border probabilistic combination, to determine whether current environment is backlight environment.
As it was previously stated, the image acquired in described image obtaining widget 610 can be coloured image or gray level image.
The backlight detection means can also include image converting member, for the coloured image to be converted into gray-scale map
Picture.Moreover, the histogram extracting parts extracts the grey level histogram of the gray level image.
The backlight detection means can be applied in the various electronic equipments with digital camera, and can be with list
Only processor is realized to realize, or by the corresponding program of central processing unit operation of corresponding electronic equipment.
Although backlighting detecting according to embodiments of the present invention is illustrated by taking normalized grey level histogram as an example above
And device, it should be understood, however, that backlighting detecting according to embodiments of the present invention and device can equally be applied in a similar manner
Embody the grey level histogram of the relation between each gray scale and pixel count corresponding to each gray scale.
Backlighting detecting and device according to embodiments of the present invention can letter based on the image shot under the present circumstances
Backlight detection singly is carried out, the need for eliminating to special CMOS backlight detection sensor, is manufactured into so as to advantageously reduce
This.
Backlighting detecting and device according to embodiments of the present invention is described by reference to accompanying drawing above.Art technology
Personnel in the case of the spirit without departing from the present invention it will be appreciated that the invention is not restricted to embodiment described above, can make
Various modifications, the modification should be also included within the scope of the present invention.The scope of the present invention should by appended claims and its
Equivalent is limited.
Claims (17)
1. a kind of backlighting detecting, including：
Gather an image；
Extract the grey level histogram of described image；And
Determine whether current environment is backlight environment according to the grey level histogram
Wherein, determine whether current environment is that backlight environment includes according to the grey level histogram：According to the grey level histogram
The local peaking of middle gray probability distribution determines whether current environment is backlight environment, wherein, in gray probability distribution
Local peaking at gray scale be less than the first gray scale or during more than the second gray scale, it is backlight environment to determine current environment；And
When gray scale at the local peaking that the gray probability is distributed is more than the 3rd gray scale and is less than four gray scales, current environment is determined
For non-backlight environment, wherein, first gray scale is less than or equal to the 3rd gray scale, and the 3rd gray scale is less than the 4th gray scale, described
4th gray scale is less than or equal to second gray scale,
Wherein, the gray scale at the local peaking that the gray probability is distributed between first gray scale and the 3rd gray scale it
Between when, or the gray probability be distributed local peaking at gray scale between the 4th gray scale and second gray scale
When, resurvey image and re-start backlight detection.
2. backlighting detecting as claimed in claim 1, wherein, according to the grey level histogram determine current environment whether be
Backlight environment includes：
Calculate the variance of the grey level histogram of described image；And
Determine whether current environment is backlight environment according to the variance.
3. backlighting detecting as claimed in claim 1, wherein, the image is coloured image, and the gray scale for extracting the image is straight
Square figure includes：
Convert the image to gray level image；And
Extract the grey level histogram of the gray level image.
4. backlighting detecting as claimed in claim 1, wherein, the grey level histogram includes multiple gray scales and each ash
The corresponding pixel quantity of degree, or
The grey level histogram be normalized grey level histogram, the normalized grey level histogram include multiple gray scales and
Probability of occurrence corresponding to each gray scale.
5. backlighting detecting as claimed in claim 2, wherein, determine whether current environment is backlight ring according to the variance
Border includes：
Backlight degree is determined according to the variance；And
When the backlight degree is more than predetermined threshold, it is backlight environment to determine current environment.
6. backlighting detecting as claimed in claim 5, wherein,
The backlight degree is the variance, or is the function of the variance.
7. backlighting detecting as claimed in claim 2, wherein, determine whether current environment is backlight ring according to the variance
Border includes：
Backlight degree is determined according to the variance；And
The probability that current environment is backlight environment is determined according to the backlight degree.
8. backlighting detecting as claimed in claim 7, wherein, determine whether current environment is backlight ring according to the variance
Border also includes：
When identified probability is more than predetermined probability threshold value, it is backlight environment to determine current environment.
9. backlighting detecting as claimed in claim 7, wherein, methods described also includes：
The probability is combined with the backlight ambient probability calculated by another method, to determine whether current environment is backlight ring
Border.
10. a kind of backlight detection means, including：
Image acquisition component, for obtaining image, the image is the image shot under the present circumstances；
Histogram extracting parts, the grey level histogram for extracting described image；And
Informating part, for determining whether the current environment is backlight environment according to the grey level histogram
Wherein, the informating part determines current environment according to the local peaking of gray probability distribution in the grey level histogram
Whether it is backlight environment, wherein, the gray scale at the local peaking that the gray probability is distributed is less than the first gray scale or is more than
During the second gray scale, the informating part determines that current environment is backlight environment；And the local peaks being distributed in the gray probability
When gray scale at value is more than the 3rd gray scale and is less than four gray scales, the informating part determines that current environment is non-backlight environment,
Wherein, first gray scale be less than or equal to the 3rd gray scale, the 3rd gray scale be less than the 4th gray scale, the 4th gray scale be less than etc.
In second gray scale,
Wherein, the gray scale at the local peaking that the gray probability is distributed between first gray scale and the 3rd gray scale it
Between when, or the gray probability be distributed local peaking at gray scale between the 4th gray scale and second gray scale
When, described image obtaining widget resurveys image, and the informating part re-starts backlight detection.
11. backlight detection means as claimed in claim 10, wherein, the informating part includes：
Variance calculating unit, the variance of the grey level histogram for calculating described image；And
Environment determines part, for determining whether current environment is backlight environment according to the variance.
12. backlight detection means as claimed in claim 10, wherein, described image is coloured image,
The backlight detection means also includes：Image converting member, for the coloured image to be converted into gray level image,
Wherein, the histogram extracting parts extracts the grey level histogram of the gray level image.
13. backlight detection means as claimed in claim 10, wherein, the grey level histogram includes multiple gray scales and each
Pixel quantity corresponding to gray scale, or
The grey level histogram be normalized grey level histogram, the normalized grey level histogram include multiple gray scales and
Probability of occurrence corresponding to each gray scale.
14. backlight detection means as claimed in claim 11, wherein, the environment determines that part includes：
Backlight degree calculating unit, for determining backlight degree according to the variance,
Backlight determines part, for determining that current environment is backlight environment when the backlight degree is more than predetermined threshold.
15. backlight detection means as claimed in claim 11, wherein, the environment determines that part includes：
Backlight degree calculating unit, for determining backlight degree according to the variance；And
Backlight ambient probability calculating unit, for determining probability of the current environment for backlight environment according to the backlight degree.
16. backlight detection means as claimed in claim 15, wherein, the environment determines that part also includes：
Backlight determines part, for determining that current environment is backlight environment when identified probability is more than predetermined probability threshold value.
17. backlight detection means as claimed in claim 15, wherein, the environment determines that part also includes：
Backlight determines part, for identified probability to be combined with the backlight ambient probability calculated by another method, with true
Whether determine current environment is backlight environment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210572564.3A CN103905737B (en) | 2012-12-25 | 2012-12-25 | Backlighting detecting and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210572564.3A CN103905737B (en) | 2012-12-25 | 2012-12-25 | Backlighting detecting and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103905737A CN103905737A (en) | 2014-07-02 |
| CN103905737B true CN103905737B (en) | 2017-09-29 |
Family
ID=50996867
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210572564.3A Active CN103905737B (en) | 2012-12-25 | 2012-12-25 | Backlighting detecting and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103905737B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105812622A (en) * | 2014-12-30 | 2016-07-27 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| CN105611140A (en) * | 2015-07-31 | 2016-05-25 | 宇龙计算机通信科技(深圳)有限公司 | Photographing control method, photographing control device and terminal |
| CN105049743B (en) * | 2015-08-21 | 2019-03-22 | 宇龙计算机通信科技(深圳)有限公司 | Backlighting detecting, backlight detection system, photographing device and terminal |
| CN105657287A (en) * | 2015-08-24 | 2016-06-08 | 宇龙计算机通信科技(深圳)有限公司 | Backlighting scene detection method and device and imaging device |
| CN105227843A (en) * | 2015-10-20 | 2016-01-06 | 宇龙计算机通信科技(深圳)有限公司 | The filming control method of terminal, the imaging control device of terminal and terminal |
| CN105872399B (en) * | 2016-04-19 | 2019-10-08 | 奇酷互联网络科技（深圳）有限公司 | Backlighting detecting and backlight detection system |
| CN107451969B (en) * | 2017-07-27 | 2020-01-10 | Oppo广东移动通信有限公司 | Image processing method, image processing device, mobile terminal and computer readable storage medium |
| CN107395988A (en) * | 2017-08-31 | 2017-11-24 | 华勤通讯技术有限公司 | The control method and system of the camera of mobile terminal |
| CN107888840B (en) * | 2017-10-30 | 2019-10-15 | Oppo广东移动通信有限公司 | High-dynamic-range image acquisition method and device |
| CN107682611B (en) * | 2017-11-03 | 2020-12-18 | Oppo广东移动通信有限公司 | Focusing method and device, computer readable storage medium and electronic equipment |
| CN107862290B (en) * | 2017-11-10 | 2021-09-24 | 智车优行科技（北京）有限公司 | Lane line detection method and system |
| CN108337448B (en) * | 2018-04-12 | 2020-09-25 | Oppo广东移动通信有限公司 | High dynamic range image acquisition method and device, terminal equipment and storage medium |
| CN110708469B (en) * | 2018-07-10 | 2021-03-19 | 北京地平线机器人技术研发有限公司 | Method and device for adapting exposure parameters and corresponding camera exposure system |
| CN112312001B (en) * | 2019-07-30 | 2022-08-23 | 阿波罗智能技术(北京)有限公司 | Image detection method, device, equipment and computer storage medium |
| CN110868533B (en) * | 2019-10-15 | 2021-06-18 | 宇龙计算机通信科技(深圳)有限公司 | HDR mode determination method, device, storage medium and terminal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101582991A (en) * | 2008-05-13 | 2009-11-18 | 深圳华为通信技术有限公司 | Method and device for processing image |
| CN102196182A (en) * | 2010-03-09 | 2011-09-21 | 株式会社理光 | Backlight detection equipment and method |
| CN102542552A (en) * | 2010-12-21 | 2012-07-04 | 北京汉王智通科技有限公司 | Frontlighting and backlighting judgment of video images and detection method of shooting time |
-
2012
- 2012-12-25 CN CN201210572564.3A patent/CN103905737B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101582991A (en) * | 2008-05-13 | 2009-11-18 | 深圳华为通信技术有限公司 | Method and device for processing image |
| CN102196182A (en) * | 2010-03-09 | 2011-09-21 | 株式会社理光 | Backlight detection equipment and method |
| CN102542552A (en) * | 2010-12-21 | 2012-07-04 | 北京汉王智通科技有限公司 | Frontlighting and backlighting judgment of video images and detection method of shooting time |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103905737A (en) | 2014-07-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103905737B (en) | Backlighting detecting and device | |
| CN105933589B (en) | A kind of image processing method and terminal | |
| CN104767933B (en) | A method of having the portable digital equipment and screening photo of camera function | |
| CN101834986B (en) | Imaging apparatus, mobile body detecting method, mobile body detecting circuit and program | |
| CN107316035A (en) | Object identifying method and device based on deep learning neutral net | |
| CN110572636B (en) | Camera contamination detection method and device, storage medium and electronic equipment | |
| CN105027553B (en) | Image processing apparatus, image processing method and the storage medium for storing image processing program | |
| CN110930329B (en) | Star image processing method and device | |
| CN105227855B (en) | A kind of image processing method and terminal | |
| CN103905727A (en) | Object area tracking apparatus, control method, and program of the same | |
| CN110072078A (en) | Monitor camera, the control method of monitor camera and storage medium | |
| CN110728705B (en) | Image processing method, image processing device, storage medium and electronic equipment | |
| JP2014106794A (en) | Face authentication device, authentication method and program thereof, and information apparatus | |
| CN103093458A (en) | Detecting method and detecting device for key frame | |
| WO2022127814A1 (en) | Method and apparatus for detecting salient object in image, and device and storage medium | |
| CN106254807B (en) | Electronic device and method for extracting still image | |
| WO2023005827A1 (en) | Exposure compensation method and apparatus, and electronic device | |
| CN109961403B (en) | Photo adjusting method and device, storage medium and electronic equipment | |
| CN106569946A (en) | Mobile terminal performance testing method and system | |
| CN106954022B (en) | Image processing method, device and terminal | |
| CN109948630A (en) | Recognition methods, device, system and the storage medium of target sheet image | |
| CN101860680B (en) | Image processing device and image processing method thereof | |
| CN111247787A (en) | Photographing method and photographing terminal | |
| CN111753796A (en) | Method and device for identifying key points in image, electronic equipment and storage medium | |
| CN113992904B (en) | Information processing method, device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |