WO2019210707A1 - Procédé d'évaluation de la netteté d'images, dispositif et dispositif électronique - Google Patents

Procédé d'évaluation de la netteté d'images, dispositif et dispositif électronique Download PDF

Info

Publication number
WO2019210707A1
WO2019210707A1 PCT/CN2019/071003 CN2019071003W WO2019210707A1 WO 2019210707 A1 WO2019210707 A1 WO 2019210707A1 CN 2019071003 W CN2019071003 W CN 2019071003W WO 2019210707 A1 WO2019210707 A1 WO 2019210707A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
value
initial value
evaluation
filter
Prior art date
Application number
PCT/CN2019/071003
Other languages
English (en)
Chinese (zh)
Inventor
刘刚
张彩红
曾峰
徐鹏
Original Assignee
杭州海康威视数字技术股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 杭州海康威视数字技术股份有限公司 filed Critical 杭州海康威视数字技术股份有限公司
Publication of WO2019210707A1 publication Critical patent/WO2019210707A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/73—Deblurring; Sharpening
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/77—Retouching; Inpainting; Scratch removal
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection

Definitions

  • the present application relates to the field of image quality evaluation technologies, and in particular, to an image sharpness evaluation method, apparatus, and electronic device.
  • the image's push function is a common feature of smart cameras, such as face push and license plate push.
  • face push and license plate push For the push function of smart cameras, it is important to introduce images with high image quality, and image sharpness is one of the indicators for measuring image quality.
  • the change of the edge information after filtering based on the sharp image is compared with the change of the blurred image after the filtering of the edge information.
  • the related technology makes the obtained image by filtering the obtained image.
  • the edge information is changed, and the sharpness of the obtained image is measured based on the amount of change of the edge information of the obtained image, that is, the amount of change of the high frequency information in the image, wherein the amount of change of the edge information may include the gradient amplitude change amount. And/or the amount of change in edge width, etc.
  • the obtained image is filtered to obtain a filtered image, and the obtained image is compared with the filtered image to determine the amount of change of the edge information corresponding to the obtained image; and the amount of change of the edge information is used to evaluate The sharpness of the image obtained.
  • the amount of change of the edge information determined by the edges of different sharpness is different.
  • the edge information corresponding to the image is calculated.
  • the amount of change enables the determination of relatively accurate sharpness.
  • the proportion of edges with different sharpness changes greatly. For example, when the person in the image changes from wearing sunglasses to not wearing sunglasses, the image is measured by the amount of change of the edge information corresponding to the image. Sharpness is not accurate enough.
  • the purpose of the embodiments of the present application is to provide an image sharpness evaluation method, device, and electronic device, so as to achieve an evaluation of the image with high precision.
  • the specific technical solutions are as follows:
  • the embodiment of the present application provides a method for evaluating image sharpness, and the method includes:
  • an intermediate frequency image corresponding to the image where the intermediate frequency image is: an image including intermediate frequency information of the image;
  • a sharpness evaluation result of the image is determined based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value.
  • the step of acquiring an image includes:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the step of filtering the image by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients including:
  • the method before the step of filtering the image by using a low pass filter of the first filter operator corresponding to the first filter coefficient to obtain a filtered image corresponding to the first filter coefficient, the method also includes:
  • the preset correspondence relationship includes a correspondence relationship between a plurality of filter coefficients and a filter operator
  • the different filter coefficients comprise a first filter coefficient and a second filter coefficient, wherein the first filter coefficient is smaller than the second filter coefficient.
  • the step of determining an intermediate frequency image corresponding to the image based on the filtered image corresponding to the obtained different filter coefficients comprises:
  • the step of calculating the difference image of the filtered image corresponding to the first filter coefficient and the filtered image corresponding to the second filter coefficient to determine the first difference image includes:
  • the step of performing horizontal analysis and longitudinal analysis on the intermediate frequency image separately to obtain an initial value of the horizontal sharpness evaluation and an initial value of the vertical sharpness evaluation including:
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value including:
  • the maximum value of the first mean value and the second average value is taken as a result of the sharpness evaluation of the image.
  • the step of calculating an average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation as a first average value includes:
  • An average value of all positive values in the initial value of the longitudinal resolution evaluation is calculated as a second average value.
  • the step of calculating an average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation as a first average value includes:
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value including:
  • determining, as the first quotient, a quotient of the luminance component value corresponding to the value, and the value of the luminance component corresponding to the value is: a pixel corresponding to the value.
  • a resolution evaluation result of the image is determined based on the first quotient and the second quotient.
  • an embodiment of the present application provides an image clarity evaluation apparatus, and the apparatus includes:
  • a first obtaining module configured to acquire an image
  • a filtering module configured to filter the image by using a low-pass filter configured with different filter coefficients to obtain a filtered image corresponding to different filter coefficients
  • a first determining module configured to determine, according to the obtained filtered image corresponding to different filter coefficients, an intermediate frequency image corresponding to the image, where the intermediate frequency image is: an image including intermediate frequency information of the image;
  • An analysis module configured to perform horizontal analysis and longitudinal analysis on the intermediate frequency image respectively, to obtain an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation;
  • a second determining module configured to determine a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value.
  • the first acquiring module is specifically configured to:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the filtering module is specifically configured to
  • the device further includes: a third determining module, configured to filter the image by using a low pass filter based on the first filter operator corresponding to the first filter coefficient to obtain a first Before the filtered image corresponding to the filter coefficient, determining the first filter operator based on the first filter coefficient and a preset correspondence relationship, where the preset correspondence relationship includes a correspondence between a plurality of filter coefficients and a filter operator ;
  • the different filter coefficients comprise a first filter coefficient and a second filter coefficient, wherein the first filter coefficient is smaller than the second filter coefficient.
  • the first determining module includes a first calculating unit and a first determining unit;
  • the first calculating unit is configured to calculate a difference image of the filtered image corresponding to the first filter coefficient and the filtered image corresponding to the second filter coefficient, and determine the first difference image;
  • the first determining unit is configured to determine an intermediate frequency image corresponding to the image based on the first difference image.
  • the first calculating unit is specifically configured to be used
  • the analyzing module is specifically used to
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the second determining module includes a second calculating unit, a third calculating unit, and a second determining unit;
  • the second calculating unit is configured to calculate an average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation as a first average value
  • the third calculating unit is configured to calculate an average value of the second evaluation initial value selected from the vertical sharpness evaluation initial value as a second average value;
  • the second determining unit is configured to determine a maximum value of the first mean value and the second average value as a resolution evaluation result of the image.
  • the second calculating unit is specifically configured to calculate an average value of all the positive values in the initial value of the horizontal sharpness evaluation as a first average value
  • the third calculating unit is specifically configured to calculate an average value of all the positive values in the initial value of the longitudinal sharpness evaluation as a second average value.
  • the second calculating unit is specifically configured to perform all the values in the initial value of the horizontal sharpness evaluation in ascending or descending order to obtain a first sequence
  • the third calculating unit is specifically configured to sort all the values in the initial value of the vertical sharpness evaluation in ascending or descending order to obtain a second sequence
  • the second determining module is specifically configured to:
  • determining, as the first quotient, a quotient of the luminance component value corresponding to the value, and the value of the luminance component corresponding to the value is: a pixel corresponding to the value.
  • a resolution evaluation result of the image is determined based on the first quotient and the second quotient.
  • an embodiment of the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through the communication bus;
  • a memory for storing a computer program
  • the processor when used to execute the computer program stored in the memory, implements the image sharpness evaluation method steps described in any of the above embodiments provided by the embodiments of the present application.
  • the embodiment of the present application provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, the foregoing A method of image sharpness evaluation method described above.
  • the embodiment of the present application provides a computer program product, when it is run on a computer, causing the computer to execute the image sharpness evaluation method step described in any of the above embodiments.
  • an image is acquired; the image is filtered by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients; and the intermediate frequency image of the image is determined based on the filtered image and the image corresponding to the different filter coefficients,
  • the intermediate frequency image is: an image containing the intermediate frequency information of the acquired image; the horizontal analysis and the longitudinal analysis are respectively performed on the intermediate frequency image, and the initial value of the horizontal sharpness evaluation and the initial value of the longitudinal sharpness evaluation are obtained; The initial value of the value and the vertical sharpness evaluation are used to determine the sharpness evaluation result of the image.
  • the intermediate frequency image including the intermediate frequency information of the image is used to determine the image clarity evaluation result, which can overcome the problem that the ratio of the sharp edge changes greatly in the image, and the image sharpness evaluation is not accurate enough.
  • the intermediate frequency information determines the basic structure of the image relative to the high frequency information that determines the edge and detail of the image, and forms the main edge structure of the image, which can be used as the secondary edge of the image, which is less susceptible to noise, and When the proportion of the sharp edge changes greatly, it is less affected by the high-frequency information, so that when the proportion of the sharp edge changes greatly, the image with higher precision can still be achieved. Evaluation.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation are used to determine the image clarity evaluation result, and the image-based information can be more comprehensive. , to determine the image clarity evaluation results, to achieve a more accurate definition of the image.
  • any of the products or methods of the present application necessarily does not necessarily require all of the advantages described above.
  • FIG. 1 is a schematic flow chart of an image sharpness evaluation method provided by an embodiment of the present application
  • FIG. 2 is a schematic flow chart of obtaining an intermediate frequency image provided by an embodiment of the present application.
  • FIG. 3 is a second difference image obtained by the image sharpness evaluation process provided by the embodiment of the present application for the acquired image, and the acquired image is subjected to the image provided by the embodiment of the present application.
  • FIG. 4 is a schematic diagram of a specific implementation process of S105 shown in FIG. 1 provided by the embodiment of the present application;
  • FIG. 5A and FIG. 5B are respectively schematic diagrams of sorting images according to the resolution evaluation result of the image determined based on the flow shown in FIG. 1; FIG.
  • FIG. 6 is a schematic structural diagram of an image sharpness evaluation apparatus that can perform the flow shown in FIG. 1 according to an embodiment of the present application;
  • FIG. 7 is a schematic structural diagram of an electronic device that can perform the process shown in FIG. 1 according to an embodiment of the present application.
  • the embodiment of the present application provides an image sharpness evaluation method, device, and electronic device, so as to implement an evaluation of the image with high precision.
  • an embodiment of the present application provides a method for evaluating image sharpness, which may include the following steps:
  • the image that can be called is the acquired image.
  • the image sharpness evaluation method provided by the embodiment of the present application can be applied to any electronic device that can obtain an image, and the electronic device can be a mobile phone, a computer, a video camera, a camera, or the like.
  • the function software for implementing the image sharpness evaluation method provided by the embodiment of the present application may exist in the form of a special client software, or may exist in the form of a plug-in of a related client software. Yes.
  • the electronic device may directly acquire an image, where the acquired image is an image that includes a region of interest, and the region of interest may be a region that includes a target of interest, and the target of interest may be: Faces, license plates, animals, etc.
  • the subsequent image sharpness evaluation process can be directly executed; when the acquired image is a color image, the acquired image can be first converted into a grayscale image, and then the subsequent image definition is performed. Evaluation process.
  • the subsequent image sharpness evaluation process may also be performed directly for the color image.
  • a subsequent image sharpness evaluation process may be performed for the R color image of the acquired image to obtain an R color. a resolution evaluation result of the image; and performing a subsequent image sharpness evaluation process for the G color image of the acquired image to obtain a sharpness evaluation result of the G color image; and performing a follow-up on the B color image of the acquired image
  • the image sharpness evaluation process obtains the sharpness evaluation result of the B color image; averages the obtained three sharpness evaluation results, and uses the obtained average value as the sharpness evaluation result of the acquired image.
  • the acquired image is a color image of the YUV format
  • a subsequent image sharpness evaluation process may be performed for the Y component image of the acquired image, and the obtained sharpness evaluation result of the Y component image is used as the Get the sharpness evaluation results of the image.
  • the Y component image of the acquired image is an image in which the pixel value of the included pixel is a Y component.
  • the interpretation of the color image of the YUV format is as follows.
  • the acquired image may be one or more.
  • the electronic device may perform the image sharpness evaluation method provided by the embodiment of the present application for each acquired image.
  • a low-pass filter configured with different filter coefficients may be pre-stored in the storage device connected to the electronic device locally or in the electronic device. After acquiring the image, the electronic device separately filters the acquired image by using a low-pass filter configured with different filter coefficients to obtain a filtered image corresponding to different filter coefficients.
  • the above filter coefficient may refer to a filter operator control parameter of the low pass filter.
  • the low pass filter may be a Gaussian filter
  • the filter coefficient is: a Gaussian filter operator control parameter of the Gaussian filter, that is, a detal parameter configured by the Gaussian filter.
  • the Gaussian filter operator is the filter operator mentioned later.
  • the obtained image is filtered by a low-pass filter with different filter coefficients, and different filtered images can be obtained.
  • the information of the images included in different filtered images is different.
  • the filtered image may only contain low frequency information of the acquired image, or the filtered image may include low frequency information and intermediate frequency information of the acquired image, or the filtered image may include low frequency information and intermediate frequency of the acquired image. Information and some high frequency information, this is all right.
  • the information contained in the resulting filtered image can be controlled by setting the filter coefficients of the low pass filter.
  • the filter coefficient of the low-pass filter may have a value range of (0, 1).
  • the filter coefficients of the above low pass filter can be set by the staff based on experience.
  • the low pass filter may be a Gaussian filter.
  • a filter operator corresponding to each filter coefficient is pre-stored in the storage device connected to the electronic device or the electronic device, and the electronic device may acquire the obtained filter based on the filter operator corresponding to each filter coefficient.
  • the image is filtered to obtain a filtered image corresponding to different filter coefficients.
  • one or more of the above low pass filters may be used.
  • the filter coefficient of the low-pass filter is adjustable, and the low-pass filter that adjusts different filter coefficients can be used to separately filter the image to obtain a filtered image corresponding to different filter coefficients; At a time, each filter coefficient may correspond to a low pass filter.
  • the intermediate frequency image is an image including intermediate frequency information of the image.
  • the intermediate frequency information of the image determines the basic structure of the image and forms the main edge structure of the image.
  • the information included in the obtained filtered image may be controlled by setting a filter coefficient of the low pass filter.
  • the electronic device may obtain the filtered image corresponding to the different filter coefficients and the acquired image. The acquired image corresponds to the intermediate frequency image.
  • the obtained filtered image corresponding to the filter coefficient only includes the low frequency information of the acquired image, and is referred to as a first type of filtered image for convenience of description; and another type of filtering coefficient corresponding to filtering
  • the image contains low frequency information and intermediate frequency information of the acquired image, and is referred to as a second type of filtered image for convenience of description.
  • the first type of filtered image may include one or more filtered images
  • the second type of filtered image may include one or more filtered images.
  • the number of filtered images included in the first type of filtered image may be the same as the number of filtered images included in the second type of filtered image.
  • the difference image of the second type of filtered image and the first type of filtered image that is, the second type of filtered image and the first type of filtered image are made pixel by pixel, and the intermediate frequency information can be obtained.
  • the difference image that is, the intermediate frequency image corresponding to the acquired image is obtained.
  • a difference image including high frequency information and intermediate frequency information can be obtained, and further, by calculating the acquired image and the second For the difference image of the filtered image, a difference image containing the high frequency information can be obtained, and then the difference image between the difference image including the high frequency information and the intermediate frequency information and the difference image containing the high frequency information can be calculated.
  • a difference image including the intermediate frequency information can be obtained, that is, an intermediate frequency image corresponding to the acquired image is obtained.
  • the value corresponding to each pixel in the obtained intermediate frequency image is greater than or equal to zero.
  • the difference corresponding to the calculated pixel point when the difference corresponding to the calculated pixel point is a negative number, the difference corresponding to the pixel point needs to be modified to 0.
  • the calculated pixel point corresponds to a difference of 0 or a positive number, and the difference corresponding to the pixel point is retained to obtain an intermediate frequency image corresponding to the acquired image.
  • S104 performing horizontal analysis and longitudinal analysis on the intermediate frequency image respectively, and obtaining an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation;
  • the horizontal analysis and the longitudinal analysis are respectively performed on the intermediate frequency image, that is, the pixels in each line of the intermediate frequency image are respectively analyzed to obtain the initial value of the lateral sharpness evaluation; and each column of the intermediate frequency image is analyzed to obtain the vertical definition. Review the initial value.
  • the process of analyzing the pixels of each row in the intermediate frequency image to obtain the initial value of the horizontal sharpness evaluation may be: for each row of pixels, using the pixel corresponding to the positive value in the row, Calculate the average value corresponding to the row, calculate the average value, and determine the initial value of the horizontal sharpness evaluation according to the order of each row; or, it may be: for each row of pixels, determine the pixel in the row. Corresponding to the pixel with the largest value, the value corresponding to the pixel with the largest value in each row is determined according to the order of each line, and the initial value of the horizontal sharpness evaluation is determined.
  • the process of analyzing each column of pixels in the intermediate frequency image to obtain the initial value of the longitudinal sharpness evaluation may be: for each column of pixels, using the pixel corresponding to the positive value in the column, calculating the average of the column corresponding to the average Value, the calculated average value is determined according to the order of each column, and the initial value of the longitudinal sharpness evaluation is determined; or, it may be: for each column of pixels, determining the pixel with the largest value corresponding to the column of the column. The value corresponding to the pixel corresponding to the largest value in each column is determined according to the order of each column, and the initial value of the longitudinal sharpness evaluation is determined.
  • the size of the intermediate frequency image is W*H
  • the obtained longitudinal resolution is obtained.
  • the p i may identify an average value of the pixel points in the i-th row of the intermediate frequency image, or a value corresponding to the pixel with the largest value in the i-th row of the intermediate frequency image, and i may take 1 to H.
  • the p j may identify an average value of the pixel points in the jth column of the intermediate frequency image, or a value corresponding to the pixel with the largest value in the jth column of the intermediate frequency image, and j may take 1 to W.
  • S105 Determine an image of the sharpness evaluation result based on the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation.
  • the resolution evaluation result is used to characterize the sharpness of the image.
  • the electronic device may determine the sharpness of the acquired image based on the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation. Evaluation results.
  • the resolution evaluation result of the obtained image determined above may be a specific evaluation value, for example, by a score representation, wherein the higher the score, the higher the sharpness of the image; or the clear level; for example, by the degree adverb representation, Including: fuzzy, fuzzy, clear, clear and very clear.
  • the average value corresponding to the initial value of the horizontal sharpness evaluation and the average value corresponding to the initial value of the vertical sharpness evaluation may be respectively calculated, and then the resolution evaluation result of the acquired image may be determined based on the average value.
  • the average value corresponding to the initial value of the horizontal sharpness evaluation is calculated as described above
  • the average value corresponding to the initial value of the horizontal sharpness evaluation may be calculated using only the value of the positive value in the initial value of the horizontal sharpness evaluation.
  • the average value corresponding to the initial value of the longitudinal resolution evaluation is calculated as described above, and the average value corresponding to the initial value of the vertical sharpness evaluation is calculated using only the value of the positive value in the initial value of the longitudinal resolution evaluation.
  • the result of the sharpness evaluation is a specific evaluation value, and the value between the two average values is larger as the sharpness evaluation value of the evaluation image; or, the average value of the above two average values is used as the evaluation image sharpness. Evaluation value.
  • the average value corresponding to the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation may be jointly calculated, and the calculated average value is used as the sharpness evaluation value of the acquired image. Among them, only the above-mentioned horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value are positive values, and the average value corresponding to the two is calculated.
  • the result of the sharpness evaluation is a clear level, wherein each clear level corresponds to a range of values: an average value corresponding to the calculated initial value of the horizontal sharpness evaluation, and an average corresponding to the initial value of the vertical sharpness evaluation.
  • a larger average value is determined, and as a reference value, the reference value is matched with the value range corresponding to each clear level, and the corresponding value range includes the clear level of the reference value as the obtained value.
  • the electronic device may perform smart mapping for the user based on the obtained sharpness evaluation result of the image, so as to introduce a higher definition image to the user.
  • the electronic device may push an image with the evaluation value higher than a preset definition threshold value, in one case, when the resolution evaluation result is a clear level.
  • the electronic device can deliver a clear and very clear image to the user.
  • the intermediate frequency image including the intermediate frequency information of the image is used to determine the image clarity evaluation result, which can overcome the problem that the ratio of the sharp edge changes greatly in the image, and the image sharpness evaluation is not accurate enough.
  • the intermediate frequency information determines the basic structure of the image relative to the high frequency information that determines the edge and detail of the image, and forms the main edge structure of the image, which can be used as the secondary edge of the image, which is less susceptible to noise, and When the proportion of the sharp edge changes greatly, it is less affected by the high-frequency information, so that when the proportion of the sharp edge changes greatly, the image with higher precision can still be achieved. Evaluation.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation are used to determine the image clarity evaluation result, and the image-based information can be more comprehensive. , to determine the image clarity evaluation results, to achieve a more accurate definition of the image.
  • the step of acquiring an image may include:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the color image may be an image of any format, such as an RGB format and a YUV format.
  • the color image when the format of the color image is the RGB format, the color image may be converted into a grayscale image based on a preset conversion formula, where the preset conversion formula may be:
  • I (i, j) identifies the gray value of the pixel at (i, j) in the acquired image
  • R (i, j) identifies the red value of the pixel at (i, j) in the color image
  • G (i, j) identifies the green value of the pixel at (i, j) in the color image
  • B (i, j) identifies the blue value of the pixel at (i, j) in the color image
  • the color image may be first converted from the YUV format to the RGB format, and then the color image is converted into a grayscale image based on the preset conversion formula.
  • the color image is converted into a grayscale image directly based on a conversion relationship between the pixel value and the grayscale value of the pre-stored pixel in the YUV format.
  • YUV also known as YCrCb
  • YCrCb is a color coding method adopted by European television systems, in which "Y” represents brightness (Luminance or Luma), that is, gray scale value; and “U” and “V” “Chrominance or Chroma” is used to describe the color and saturation of an image and is used to specify the color of a pixel.
  • RGB Red Green Blue
  • Chroma defines two aspects of color - hue and saturation, expressed in terms of Cr and Cb, respectively. Among them, Cr reflects the difference between the red part of the RGB input signal and the brightness value of the RGB input signal. Cb reflects the difference between the blue portion of the RGB input signal and the luminance value of the RGB input signal.
  • the image sharpness evaluation process performs the image of the sharpness evaluation; or, it may be: first convert the color image into a grayscale image, and then select a partial image from the grayscale image, for example, an image of the user's region of interest, as the passbook
  • the image sharpness evaluation process provided by the embodiment is applied to perform image evaluation of the sharpness.
  • the foregoing different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the step of filtering the image by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients may include:
  • the different filter coefficients may include a first filter coefficient and a second filter coefficient, wherein the first filter coefficient and the second filter coefficient may be identified by ⁇ 1 and ⁇ 2 , respectively .
  • the low pass filter of different filter coefficients may be a Gaussian filter, and may be a Gaussian filter configured with a first filter coefficient and a Gaussian filter configured with a second filter coefficient, respectively.
  • the first filter operator corresponding to the first filter coefficient and the second filter operator corresponding to the second filter coefficient segment may be operators of N*N, and the embodiment of the present application does not apply to the first filter operator and the second filter operator.
  • the specification is limited, wherein, in order to facilitate determination of the intermediate frequency image, the specification of the first filter operator and the specification of the second filter operator may be the same, and N is a positive integer.
  • the first filter coefficient ⁇ 1 may be 0.5, and the first filter coefficient ⁇ 1 corresponds to a filter operator K 1 of 3 ⁇ 3 , where
  • the second filter coefficient ⁇ 2 may be 0.9, and the second filter coefficient ⁇ 2 corresponds to a filter operator K 2 of 3 ⁇ 3 , where
  • the first filter coefficient ⁇ 1 may take 0.4
  • the second filter coefficient ⁇ 2 may take 0.8
  • the electronic device may perform the acquired image based on the corresponding filter operator when the first filter coefficient ⁇ 1 is 0.4. Filtering to obtain a corresponding filtered image when the first filter coefficient ⁇ 1 is 0.4; and filtering the acquired image based on the corresponding filter operator when the second filter coefficient ⁇ 2 is 0.8, to obtain a second filter coefficient ⁇ 2 is a corresponding filtered image at 0.8, and then a subsequent image sharpness evaluation process is performed based on the obtained two filtered images.
  • the first filter operator and the second filter operator are 3x3 operators.
  • the size of the subsequently obtained intermediate frequency image may be (M-2)*(N-2)
  • the invalid pixel points of the edges in the intermediate frequency image can be removed, so that the resolution evaluation result of the acquired image is more accurate based on the determination of the intermediate frequency image. This is because the 3x3 operator has the possibility that the edge of the edge of the acquired image cannot be filtered.
  • the method before the step of filtering the image by using a low-pass filter of the first filter operator corresponding to the first filter coefficient to obtain a filtered image corresponding to the first filter coefficient, the method further Can include:
  • a second filter operator is determined based on the second filter coefficient and the preset correspondence.
  • the process of determining the first filter operator corresponding to the first filter coefficient is the same as the process of determining the second filter operator corresponding to the second filter coefficient.
  • the process of determining the first filter operator corresponding to the first filter coefficient is taken as an example for description.
  • a preset relationship may be stored in the external storage device of the electronic device or the connected external storage device.
  • the preset correspondence may include a correspondence between multiple filter coefficients and a filter operator, and the preset correspondence may be searched for. And determining a first filter operator corresponding to the first filter coefficient.
  • the first filter coefficient may be compared with the filter coefficient in the preset correspondence relationship, and the filter operator corresponding to the filter coefficient with the same first filter coefficient in the preset correspondence relationship is determined as the first filter. operator.
  • a Gaussian filter function may be stored in the local device or the connected external storage device. After the filter coefficient is input into the Gaussian filter function, the filter operator corresponding to the filter coefficient may be calculated. The first filter coefficient may be input to the Gaussian filter function to obtain a first filter operator corresponding to the first filter coefficient.
  • the Gaussian filter function may be any type of Gaussian filter function, and the specific type of the Gaussian filter function is not limited in the embodiment of the present application.
  • different filter coefficients may include a first filter coefficient and a second filter coefficient, where the first filter coefficient is smaller than the second filter coefficient;
  • the step of determining the intermediate frequency image corresponding to the image based on the obtained filtered image corresponding to the different filter coefficients may include:
  • An intermediate frequency image corresponding to the image is determined based on the first difference image.
  • the step of determining the intermediate frequency image corresponding to the image based on the filtered image corresponding to the different filter coefficients obtained may include:
  • S201 Calculate a difference image of the filtered image corresponding to the first filter coefficient, and determine the second difference image;
  • S203 Calculate a difference image of the third difference image and the second difference image, and determine the first difference image
  • S204 Determine an intermediate frequency image corresponding to the image based on the first difference image.
  • the S201 to S203 are an alternative implementation manner of determining the first difference image.
  • the acquired image is filtered by the low pass filter of the first filter coefficient and the low pass filter of the second filter coefficient, respectively, and the filtered image including the low frequency information and the intermediate frequency information may be respectively obtained, and A filtered image containing low frequency information.
  • the filtered image including the low frequency information and the intermediate frequency information may be directly used to perform a difference with the filtered image including the low frequency information to obtain an intermediate frequency image including the intermediate frequency information, that is, the filtered image corresponding to the first filter coefficient is calculated.
  • the difference image of the filtered image corresponding to the second filter coefficient is determined as the first difference image, and the intermediate frequency image is determined based on the first difference image.
  • the process of determining the intermediate frequency image based on the first difference image may be: comparing, for each pixel in the first difference image, a value of the pixel in the first difference image and a size of 0, when the first If the value of the pixel in the difference image is greater than or equal to 0, the value of the pixel in the first difference image is retained, and when the value of the pixel in the first difference image is less than 0, the first The value of the pixel in the difference image is replaced by zero.
  • the FCS i,j identifies the value of the pixel at the (i,j) position in the intermediate frequency image
  • the I i,j identifies the gray of the pixel at the position of (i,j) in the acquired image.
  • the value K 1 identifies a filter operator corresponding to the first filter coefficient
  • the K 2 identifies a filter operator corresponding to the second filter coefficient.
  • the above max(I i,j *K 1 -I i,j *K 2 ,0) identifies the maximum value between I i,j *K 1 -I i,j *K 2 and 0.
  • the acquired image may be directly used to perform a difference with the filtered image including the low frequency information and the intermediate frequency information to obtain an image including the high frequency information, that is, the second difference image is obtained, and the second difference image is obtained.
  • the acquired image is compared with the filtered image containing the low frequency information to obtain an image containing the high frequency information and the intermediate frequency information, that is, the third difference image is obtained; and further, the image containing the high frequency information and the intermediate frequency information and the image containing the high frequency information
  • the image is made to be different to obtain an image containing the intermediate frequency information, that is, the first difference image is obtained, and then the intermediate frequency image corresponding to the image to be evaluated is obtained based on the first difference image. As shown in FIG.
  • the second difference image obtained by the image sharpness evaluation process provided by the embodiment of the present application is obtained by the embodiment of the present application.
  • a third difference image obtained by the image sharpness evaluation process and an example image of the intermediate frequency image obtained based on the flow of obtaining the intermediate frequency image shown in FIG. wherein, the "original picture” shown in FIG. 3 represents the acquired image, the "high frequency picture” shown in FIG. 3 represents the second difference image, and the "medium high frequency picture” shown in FIG. 3 represents the above.
  • the third difference image, the "intermediate frequency map" shown in Fig. 3, characterizes the above intermediate frequency image.
  • FCS i,j max((I i,j -I i,j *K 2 )-(I i,j -I i,j *K 1 ),0)...(2)
  • the FCS i,j identifies the value of the pixel at the (i,j) position in the intermediate frequency image
  • the I i,j identifies the gray of the pixel at the position of (i,j) in the acquired image.
  • the value K 1 identifies a filter operator corresponding to the first filter coefficient
  • the K 2 identifies a filter operator corresponding to the second filter coefficient.
  • the above max((I i,j -I i,j *K 2 )-(I i,j -I i,j *K 1 ),0) identifies (I i,j -I i,j *K 2
  • the intermediate frequency image and the longitudinal analysis may be separately performed to obtain the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation.
  • the foregoing steps of performing horizontal analysis and longitudinal analysis on the intermediate frequency image respectively to obtain an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation may include:
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the image may be determined based on the average value corresponding to the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation.
  • the result of the clarity evaluation may include:
  • the maximum value of the first mean value and the second average value is determined as the sharpness evaluation result of the image.
  • the resolution evaluation result of the acquired image may be calculated based only on the value of the positive number, that is, the first evaluation initial value is a positive value, and the second evaluation initial value is A positive number.
  • the calculating, as the first average value, the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation may include: calculating an average value of all positive values in the initial value of the horizontal sharpness evaluation As the first average;
  • the calculating, as a second average value, the average value of the second evaluation initial value selected from the initial value of the vertical sharpness evaluation may include: calculating an average value of all positive values in the initial value of the vertical sharpness evaluation , as the second average.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation may be separately selected to select a part of the higher reliability.
  • the foregoing steps of determining the resolution evaluation result of the image based on the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation may include:
  • S401 Sort all the values in the initial value of the horizontal sharpness evaluation in ascending or descending order to obtain a first sequence
  • S402 Filter all the values in the preset range from the first sequence; calculate an average value of all the filtered values as the first average value;
  • S403 Sort all the values in the initial value of the vertical sharpness evaluation in ascending or descending order to obtain a second sequence
  • S404 Filter all the values in the preset range from the second sequence; calculate an average value of all the filtered values as the second average value;
  • S405 Determine a maximum value of the first average value and the second average value as a result of the sharpness evaluation of the image.
  • S401 and S402 are an optional implementation manner of the step of calculating, as the first average value, an average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation; the S403 and the S404 are the foregoing Calculating an average of the second evaluation initial value selected from the initial value of the longitudinal definition evaluation as an alternative implementation of the step of the second average.
  • the preset range may be set according to a sorting order of the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation.
  • the preset range corresponding to the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation in the ascending order is preset corresponding to the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation.
  • the range can be the same or different.
  • the preset range may be set manually by the staff according to requirements, or may be a default setting of the electronic device, wherein the default setting may represent: the setting of the electronic device when leaving the factory.
  • the above preset range may be [20%, 80%], or [0, 95%], or [5%, 100%], and the like.
  • the first average value may be identified as RowClarity
  • the second average value may be identified as ColClarity
  • the foregoing process of determining the first average value and the process of determining the second average value may be performed in parallel or sequentially, which is all possible.
  • the step of determining an image of a sharpness evaluation result based on the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation may include:
  • a quotient of the value of the luminance component corresponding to the value is calculated as the first quotient, wherein the value of the luminance component corresponding to the value is: corresponding to the pixel corresponding to the value Luminance component value;
  • a quotient of the luminance component value corresponding to the value is calculated as a second quotient, wherein the value of the luminance component corresponding to the value is: corresponding to the pixel corresponding to the value Luminance component value;
  • the image clarity evaluation result is determined.
  • the luminance component value of the pixel value of the pixel may be determined.
  • each value of the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation may be divided by the brightness component value corresponding to the value, The effect of the illumination factor, the brightness, is removed from this value. Further, based on the obtained quotient, the resolution evaluation result of the acquired image is determined to obtain a more accurate definition evaluation result.
  • the brightness component value corresponding to the above pixel point can be represented by the Y component value of the pixel value of the pixel point in the YUV format, and the range of the Y component corresponding to the pixel point is 0-255.
  • the minimum value of the Y component may be set to 1, that is, the electronic device may set the Y component value corresponding to each pixel point in the acquired image to 1, and set the Y component value smaller than 1 to 1.
  • the original value of the Y component value of not less than 1 is reserved.
  • the process of determining the resolution evaluation result of the image based on the first quotient and the second quotient may be: respectively calculating an average value of the first quotient and an average value of the second quotient, and further, The larger of the above two average values is selected as the result of the sharpness evaluation of the image.
  • the average of the above two average values is taken as the result of the sharpness evaluation of the image.
  • the value of the two average values may be used as a reference value, and the reference value is matched with a value range corresponding to each clear level stored in advance, and the corresponding value range includes the above reference.
  • the clear level of the value is used as the result of the image clarity evaluation.
  • the image sharpness evaluation method provided by the embodiment of the present application is used to perform the resolution evaluation on the acquired series of images, and the resolution evaluation result corresponding to each image is respectively A schematic diagram of sorting each image, that is, a schematic diagram of sorting images based on the sharpness evaluation result of the image determined based on the flow shown in FIG. Among them, the lower the sorting order, the higher the sharpness of the image.
  • the embodiment of the present application provides an image sharpness evaluation device, and the device includes:
  • a first obtaining module 610 configured to acquire an image
  • the filtering module 620 is configured to filter the image by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients;
  • a first determining module 630 configured to determine, according to the obtained filtered image corresponding to the different filter coefficients, an intermediate frequency image corresponding to the image, where the intermediate frequency image is: an image including intermediate frequency information of the image;
  • the analysis module 640 is configured to perform horizontal analysis and longitudinal analysis on the intermediate frequency image respectively to obtain an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation;
  • the second determining module 650 is configured to determine a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value.
  • the resolution evaluation result is used to characterize the sharpness of the image.
  • the intermediate frequency image including the intermediate frequency information of the image is used to determine the image clarity evaluation result, which can overcome the problem that the ratio of the sharp edge changes greatly in the image, and the image sharpness evaluation is not accurate enough.
  • the intermediate frequency information determines the basic structure of the image relative to the high frequency information that determines the edge and detail of the image, and forms the main edge structure of the image, which can be used as the secondary edge of the image, which is less susceptible to noise, and When the proportion of the sharp edge changes greatly, it is less affected by the high-frequency information, so that when the proportion of the sharp edge changes greatly, the image with higher precision can still be achieved. Evaluation.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation are used to determine the image clarity evaluation result, and the image-based information can be more comprehensive. , to determine the image clarity evaluation results, to achieve a more accurate definition of the image.
  • the first acquiring module 610 is specifically configured to:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the filtering module 620 is specifically configured to be used
  • the apparatus further includes: a third determining module, configured to filter the image in the low pass filter based on the first filter operator corresponding to the first filter coefficient, Before obtaining the filtered image corresponding to the first filter coefficient, determining the first filter operator based on the first filter coefficient and a preset correspondence, wherein the preset correspondence includes multiple filter coefficients and a filter operator Correspondence relationship;
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, wherein the first filter coefficient is smaller than the second filter coefficient.
  • the first determining module 630 includes a first calculating unit and a first determining unit
  • the first calculating unit is configured to calculate a difference image of the filtered image corresponding to the first filter coefficient and the filtered image corresponding to the second filter coefficient, and determine the first difference image;
  • the first determining unit is configured to determine an intermediate frequency image corresponding to the image based on the first difference image.
  • the first calculating unit is specifically configured to be used
  • the analyzing module 640 is specifically configured to be used
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the second determining module 650 includes a second calculating unit, a third calculating unit, and a second determining unit;
  • the second calculating unit is configured to calculate an average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation as a first average value
  • the third calculating unit is configured to calculate an average value of the second evaluation initial value selected from the vertical sharpness evaluation initial value as a second average value;
  • the second determining unit is configured to determine a maximum value of the first mean value and the second average value as a resolution evaluation result of the image.
  • the second calculating unit is specifically configured to calculate an average value of all the positive values in the initial value of the horizontal sharpness evaluation as a first average value
  • the third calculating unit is specifically configured to calculate an average value of all the positive values in the initial value of the longitudinal sharpness evaluation as a second average value.
  • the second calculating unit is specifically configured to perform an ascending or descending sorting of all values in the initial value of the horizontal sharpness evaluation to obtain a first sequence
  • the third calculating unit is specifically configured to sort all the values in the initial value of the vertical sharpness evaluation in ascending or descending order to obtain a second sequence
  • the second determining module 650 is specifically configured to:
  • determining, as the first quotient, a quotient of the luminance component value corresponding to the value, and the value of the luminance component corresponding to the value is: a pixel corresponding to the value.
  • a resolution evaluation result of the image is determined based on the first quotient and the second quotient.
  • the embodiment of the present application further provides an electronic device, as shown in FIG. 7, including a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720 The memory 730 completes communication with each other through the communication bus 740.
  • a memory 730 configured to store a computer program
  • the processor 710 when used to execute the computer program stored in the memory 730, implements the image sharpness evaluation method according to any one of the embodiments of the present application, and may include the following steps:
  • an intermediate frequency image corresponding to the image where the intermediate frequency image is: an image including intermediate frequency information of the image;
  • a sharpness evaluation result of the image is determined based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value.
  • the resolution evaluation result is used to characterize the sharpness of the image.
  • the intermediate frequency image including the intermediate frequency information of the image is used to determine the image clarity evaluation result, which can overcome the problem that the ratio of the sharp edge changes greatly in the image, and the image sharpness evaluation is not accurate enough.
  • the intermediate frequency information determines the basic structure of the image relative to the high frequency information that determines the edge and detail of the image, and forms the main edge structure of the image, which can be used as the secondary edge of the image, which is less susceptible to noise, and When the proportion of the sharp edge changes greatly, it is less affected by the high-frequency information, so that when the proportion of the sharp edge changes greatly, the image with higher precision can still be achieved. Evaluation.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation are used to determine the image clarity evaluation result, and the image-based information can be more comprehensive. , to determine the image clarity evaluation results, to achieve a more accurate definition of the image.
  • the acquiring an image includes:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the step of filtering the image by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients including:
  • the method before the step of filtering the image by using a low pass filter of the first filter operator corresponding to the first filter coefficient to obtain a filtered image corresponding to the first filter coefficient, the method further includes:
  • the preset correspondence relationship includes a correspondence relationship between a plurality of filter coefficients and a filter operator
  • the different filter coefficients comprise a first filter coefficient and a second filter coefficient, wherein the first filter coefficient is smaller than the second filter coefficient.
  • the step of determining an intermediate frequency image corresponding to the image based on the filtered image corresponding to the obtained different filter coefficients comprises:
  • the step of calculating the difference image of the filtered image corresponding to the first filter coefficient and the filtered image corresponding to the second filter coefficient to determine the first difference image includes:
  • the step of performing horizontal analysis and longitudinal analysis on the intermediate frequency image separately to obtain an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation including:
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value includes:
  • the maximum value of the first mean value and the second average value is taken as a result of the sharpness evaluation of the image.
  • the calculating, by using the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation, as the first average value includes:
  • An average value of all positive values in the initial value of the longitudinal resolution evaluation is calculated as a second average value.
  • the calculating, by using the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation, as the first average value includes:
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value includes:
  • determining, as the first quotient, a quotient of the luminance component value corresponding to the value, and the value of the luminance component corresponding to the value is: a pixel corresponding to the value.
  • a resolution evaluation result of the image is determined based on the first quotient and the second quotient.
  • the communication bus mentioned in the above electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus.
  • PCI Peripheral Component Interconnect
  • EISA Extended Industry Standard Architecture
  • the communication bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.
  • the communication interface is used for communication between the above electronic device and other devices.
  • the memory may include a random access memory (RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.
  • RAM random access memory
  • NVM non-volatile memory
  • the memory may also be at least one storage device located away from the aforementioned processor.
  • the above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; or may be a digital signal processing (DSP), dedicated integration.
  • CPU central processing unit
  • NP network processor
  • DSP digital signal processing
  • ASIC Application Specific Integrated Circuit
  • FPGA Field-Programmable Gate Array
  • the embodiment of the present application provides a computer readable storage medium, where the computer readable storage medium stores a computer program, and when the computer program is executed by the processor, the embodiment of the present application is provided.
  • Any of the image clarity evaluation methods described may include the steps of:
  • an intermediate frequency image corresponding to the image where the intermediate frequency image is: an image including intermediate frequency information of the image;
  • a sharpness evaluation result of the image is determined based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value.
  • the resolution evaluation result is used to characterize the sharpness of the image.
  • the intermediate frequency image including the intermediate frequency information of the image is used to determine the image clarity evaluation result, which can overcome the problem that the ratio of the sharp edge changes greatly in the image, and the image sharpness evaluation is not accurate enough.
  • the intermediate frequency information determines the basic structure of the image relative to the high frequency information that determines the edge and detail of the image, and forms the main edge structure of the image, which can be used as the secondary edge of the image, which is less susceptible to noise, and When the proportion of the sharp edge changes greatly, it is less affected by the high-frequency information, so that when the proportion of the sharp edge changes greatly, the image with higher precision can still be achieved. Evaluation.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation are used to determine the image clarity evaluation result, and the image-based information can be more comprehensive. , to determine the image clarity evaluation results, to achieve a more accurate definition of the image.
  • the acquiring an image includes:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the step of filtering the image by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients including:
  • the method before the step of filtering the image by using a low pass filter of the first filter operator corresponding to the first filter coefficient to obtain a filtered image corresponding to the first filter coefficient, the method further includes:
  • the preset correspondence relationship includes a correspondence relationship between a plurality of filter coefficients and a filter operator
  • the different filter coefficients comprise a first filter coefficient and a second filter coefficient, wherein the first filter coefficient is smaller than the second filter coefficient.
  • the step of determining an intermediate frequency image corresponding to the image based on the filtered image corresponding to the obtained different filter coefficients comprises:
  • the step of calculating the difference image of the filtered image corresponding to the first filter coefficient and the filtered image corresponding to the second filter coefficient to determine the first difference image includes:
  • the step of performing horizontal analysis and longitudinal analysis on the intermediate frequency image separately to obtain an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation including:
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value includes:
  • the maximum value of the first mean value and the second average value is taken as a result of the sharpness evaluation of the image.
  • the calculating, by using the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation, as the first average value includes:
  • An average value of all positive values in the initial value of the longitudinal resolution evaluation is calculated as a second average value.
  • the calculating, by using the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation, as the first average value includes:
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value includes:
  • determining, as the first quotient, a quotient of the luminance component value corresponding to the value, and the value of the luminance component corresponding to the value is: a pixel corresponding to the value.
  • a resolution evaluation result of the image is determined based on the first quotient and the second quotient.
  • the embodiment of the present application provides a computer program product, which when executed on a computer, causes the computer to perform any of the image sharpness evaluation methods provided by the embodiments of the present application, which may include step:
  • an intermediate frequency image corresponding to the image where the intermediate frequency image is: an image including intermediate frequency information of the image;
  • a sharpness evaluation result of the image is determined based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value.
  • the resolution evaluation result is used to characterize the sharpness of the image.
  • the intermediate frequency image including the intermediate frequency information of the image is used to determine the image clarity evaluation result, which can overcome the problem that the ratio of the sharp edge changes greatly in the image, and the image sharpness evaluation is not accurate enough.
  • the intermediate frequency information determines the basic structure of the image relative to the high frequency information that determines the edge and detail of the image, and forms the main edge structure of the image, which can be used as the secondary edge of the image, which is less susceptible to noise, and When the proportion of the sharp edge changes greatly, it is less affected by the high-frequency information, so that when the proportion of the sharp edge changes greatly, the image with higher precision can still be achieved. Evaluation.
  • the initial value of the horizontal sharpness evaluation and the initial value of the vertical sharpness evaluation are used to determine the image clarity evaluation result, and the image-based information can be more comprehensive. , to determine the image clarity evaluation results, to achieve a more accurate definition of the image.
  • the acquiring an image includes:
  • the color image is converted into a grayscale image, and a local grayscale image is selected as the image.
  • the different filter coefficients include a first filter coefficient and a second filter coefficient, where the first filter coefficient corresponds to a first filter operator, and the second filter coefficient corresponds to a second filter operator;
  • the step of filtering the image by using a low-pass filter with different filter coefficients to obtain a filtered image corresponding to different filter coefficients including:
  • the method before the step of filtering the image by using a low pass filter of the first filter operator corresponding to the first filter coefficient to obtain a filtered image corresponding to the first filter coefficient, the method further includes:
  • the preset correspondence relationship includes a correspondence relationship between a plurality of filter coefficients and a filter operator
  • the different filter coefficients comprise a first filter coefficient and a second filter coefficient, wherein the first filter coefficient is smaller than the second filter coefficient.
  • the step of determining an intermediate frequency image corresponding to the image based on the filtered image corresponding to the obtained different filter coefficients comprises:
  • the step of calculating the difference image of the filtered image corresponding to the first filter coefficient and the filtered image corresponding to the second filter coefficient to determine the first difference image includes:
  • the step of performing horizontal analysis and longitudinal analysis on the intermediate frequency image respectively to obtain an initial value of the horizontal sharpness evaluation and an initial value of the longitudinal sharpness evaluation including:
  • the values corresponding to all the determined second type of pixel points are determined as initial values of the longitudinal sharpness evaluation.
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value includes:
  • the maximum value of the first mean value and the second average value is taken as a result of the sharpness evaluation of the image.
  • the calculating, by using the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation, as the first average value includes:
  • An average value of all positive values in the initial value of the longitudinal resolution evaluation is calculated as a second average value.
  • the calculating, as the first average value, the average value of the first evaluation initial value selected from the initial value of the horizontal sharpness evaluation includes:
  • the step of determining a resolution evaluation result of the image based on the horizontal sharpness evaluation initial value and the vertical sharpness evaluation initial value includes:
  • determining, as the first quotient, a quotient of the luminance component value corresponding to the value, and the value of the luminance component corresponding to the value is: a pixel corresponding to the value.
  • a resolution evaluation result of the image is determined based on the first quotient and the second quotient.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

L'invention concerne un dispositif électronique, un dispositif et un procédé d'évaluation de la netteté d'images, le procédé consistant à : obtenir une image (S101) ; filtrer l'image à l'aide de filtres passe-bas présentant différents coefficients de filtre pour obtenir des images filtrées correspondant à différents coefficients de filtre (S102) ; déterminer une image de fréquence intermédiaire sur la base des images filtrées correspondant aux différents coefficients de filtre et à l'image (S103), l'image de fréquence intermédiaire étant : une image contenant des informations de fréquence intermédiaire de l'image ; effectuer respectivement une analyse horizontale et une analyse verticale sur l'image de fréquence intermédiaire, pour obtenir une valeur initiale d'évaluation de netteté horizontale et une valeur initiale d'évaluation de netteté verticale (S104) ; et déterminer un résultat d'évaluation de netteté de l'image sur la base de la valeur initiale d'évaluation de netteté horizontale et de la valeur initiale d'évaluation de netteté verticale (S105). Ainsi, une évaluation de netteté d'image plus précise est réalisée.
PCT/CN2019/071003 2018-05-02 2019-01-09 Procédé d'évaluation de la netteté d'images, dispositif et dispositif électronique WO2019210707A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810411295.XA CN110458789B (zh) 2018-05-02 2018-05-02 一种图像清晰度评测方法、装置及电子设备
CN201810411295.X 2018-05-02

Publications (1)

Publication Number Publication Date
WO2019210707A1 true WO2019210707A1 (fr) 2019-11-07

Family

ID=68386234

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/071003 WO2019210707A1 (fr) 2018-05-02 2019-01-09 Procédé d'évaluation de la netteté d'images, dispositif et dispositif électronique

Country Status (2)

Country Link
CN (1) CN110458789B (fr)
WO (1) WO2019210707A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116506719A (zh) * 2023-06-21 2023-07-28 深圳华强电子网集团股份有限公司 一种基于光电二极管cmos图像传感器的传输管理方法
CN117252783A (zh) * 2023-11-06 2023-12-19 上海为旌科技有限公司 一种失焦模糊图像的清晰度计算方法、装置和设备

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111179259B (zh) * 2019-12-31 2023-09-26 北京灵犀微光科技有限公司 光学清晰度测试方法和装置
CN113542796B (zh) * 2020-04-22 2023-08-08 腾讯科技（深圳）有限公司 视频评测方法、装置、计算机设备和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102881003A (zh) * 2012-08-30 2013-01-16 暨南大学 一种消除ccd天文图像中宇宙射线的方法
CN103034979A (zh) * 2012-11-30 2013-04-10 声泰特（成都）科技有限公司 一种超声图像清晰度提升方法
US20130089256A1 (en) * 2010-06-30 2013-04-11 Nec Corporation Color image processing method, color image processing device, and color image processing program
CN106127775A (zh) * 2016-06-28 2016-11-16 乐视控股（北京）有限公司 图像清晰度评价方法及装置
CN106934804A (zh) * 2017-03-13 2017-07-07 重庆贝奥新视野医疗设备有限公司 图像清晰度检测方法及装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9552506B1 (en) * 2004-12-23 2017-01-24 Cognex Technology And Investment Llc Method and apparatus for industrial identification mark verification
US20150363912A1 (en) * 2014-06-12 2015-12-17 Samsung Electronics Co., Ltd. Rgbw demosaic method by combining rgb chrominance with w luminance
CN106548468B (zh) * 2016-10-13 2019-08-23 广州酷狗计算机科技有限公司 图像清晰度的判别方法及装置
CN107507173B (zh) * 2017-08-15 2021-07-27 上海交通大学 一种全切片图像的无参考清晰度评估方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130089256A1 (en) * 2010-06-30 2013-04-11 Nec Corporation Color image processing method, color image processing device, and color image processing program
CN102881003A (zh) * 2012-08-30 2013-01-16 暨南大学 一种消除ccd天文图像中宇宙射线的方法
CN103034979A (zh) * 2012-11-30 2013-04-10 声泰特（成都）科技有限公司 一种超声图像清晰度提升方法
CN106127775A (zh) * 2016-06-28 2016-11-16 乐视控股（北京）有限公司 图像清晰度评价方法及装置
CN106934804A (zh) * 2017-03-13 2017-07-07 重庆贝奥新视野医疗设备有限公司 图像清晰度检测方法及装置

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116506719A (zh) * 2023-06-21 2023-07-28 深圳华强电子网集团股份有限公司 一种基于光电二极管cmos图像传感器的传输管理方法
CN116506719B (zh) * 2023-06-21 2023-09-15 深圳华强电子网集团股份有限公司 一种基于光电二极管cmos图像传感器的传输管理方法
CN117252783A (zh) * 2023-11-06 2023-12-19 上海为旌科技有限公司 一种失焦模糊图像的清晰度计算方法、装置和设备
CN117252783B (zh) * 2023-11-06 2024-05-17 上海为旌科技有限公司 一种失焦模糊图像的清晰度计算方法、装置和设备

Also Published As

Publication number Publication date
CN110458789B (zh) 2022-04-05
CN110458789A (zh) 2019-11-15

Similar Documents

Publication Publication Date Title
WO2019210707A1 (fr) Procédé d'évaluation de la netteté d'images, dispositif et dispositif électronique
CN107403421B (zh) 一种图像去雾方法、存储介质及终端设备
CN107507173B (zh) 一种全切片图像的无参考清晰度评估方法及系统
US7986830B2 (en) Radiometric calibration from a single image
CN109389135B (zh) 一种图像筛选方法及装置
CN107451969A (zh) 图像处理方法、装置、移动终端及计算机可读存储介质
CN109871845B (zh) 证件图像提取方法及终端设备
CN107590447A (zh) 一种文字标题识别方法及装置
CN110969170B (zh) 一种图像主题色提取方法、装置及电子设备
CN107563976A (zh) 美颜参数获取方法、装置、可读存储介质和计算机设备
CN111131688B (zh) 一种图像处理方法、装置及移动终端
CN110111347B (zh) 图像标志提取方法、装置及存储介质
CN107454285A (zh) 图像噪声估计方法及装置与图像提取装置
CN111368717A (zh) 视线确定方法、装置、电子设备和计算机可读存储介质
CN109903265B (zh) 一种图像变化区域侦测阀值设定方法、系统及其电子装置
WO2019029573A1 (fr) Procédé de floutage d'image, support d'informations lisible par ordinateur et dispositif informatique
JP2004310475A (ja) 画像処理装置、画像処理を行う携帯電話、および画像処理プログラム
CN112150368A (zh) 图像处理方法、装置、电子设备及计算机可读存储介质
CN113132695A (zh) 镜头阴影校正方法、装置及电子设备
CN114584700A (zh) 对焦标记方法、标记装置以及电子设备
CN114998122A (zh) 一种低照度图像增强方法
CN111738272A (zh) 一种目标特征提取方法、装置及电子设备
CN113766203B (zh) 一种图像白平衡处理方法
CN112634166A (zh) 一种图像处理方法、装置、电子设备及存储介质
CN107424134A (zh) 图像处理方法、装置、计算机可读存储介质和计算机设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19796676

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19796676

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 19796676

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11.05.2021)

122 Ep: pct application non-entry in european phase

Ref document number: 19796676

Country of ref document: EP

Kind code of ref document: A1