CN108492289B - Digital image quality evaluation system - Google Patents

Digital image quality evaluation system Download PDF

Info

Publication number
CN108492289B
CN108492289B CN201810222242.3A CN201810222242A CN108492289B CN 108492289 B CN108492289 B CN 108492289B CN 201810222242 A CN201810222242 A CN 201810222242A CN 108492289 B CN108492289 B CN 108492289B
Authority
CN
China
Prior art keywords
module
identification module
pixel
pixel points
gradient
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
Application number
CN201810222242.3A
Other languages
Chinese (zh)
Other versions
CN108492289A (en
Inventor
严天宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Baoyi Pictures Co ltd
Original Assignee
Shanghai Baoyi Pictures Co ltd
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 Shanghai Baoyi Pictures Co ltd filed Critical Shanghai Baoyi Pictures Co ltd
Priority to CN201810222242.3A priority Critical patent/CN108492289B/en
Publication of CN108492289A publication Critical patent/CN108492289A/en
Application granted granted Critical
Publication of CN108492289B publication Critical patent/CN108492289B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • 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
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G06T7/41—Analysis of texture based on statistical description of texture
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

The invention relates to a digital image quality evaluation system, which comprises a reading module, a fault identification module, an aperture identification module, an analysis module and a feedback module, wherein the reading module, the fault identification module, the aperture identification module, the analysis module and the feedback module can respectively preprocess a color image collected by a digital camera, identify color faults and color blocks possibly appearing in the image through an intelligent algorithm, effectively evaluate the definition and the color richness of the image, and can make data fully play the value and improve the use efficiency. The evaluation system avoids the image inspection through human vision, reduces man-machine interaction operation, has higher automation degree and more objective and efficient evaluation process.

Description

Digital image quality evaluation system
Technical Field
The invention relates to the field of image processing, in particular to a digital image quality evaluation system.
Background
With the development of printing technology, photography technology, modern communication technology, computer technology and the like and the popularization of digital transmission technology, the communication and communication between people and the transmission and replacement of information have entered the era of 'visualization' and 'imaging', the application range of images has obviously surpassed language characters, and the images become the main media for people to acquire information and communication ideas. And one of the keywords closely related to "visualization" and "imaging" is "digital image". The multi-element characteristics of digital images create a rich image world, so that people are in the propagation environment and the life style of image culture.
However, some factors that reduce the quality of the digital image inevitably occur in the process of collecting, transmitting, storing and processing the digital data, so that the application potential of the digital image is greatly limited. Most notably, digital images are affected by various factors during the acquisition process, resulting in noise generation, resulting in degradation of the quality of the acquired images and image blurring. Therefore, it is important to evaluate the quality of digital images in order to make the data sufficiently use the value and improve the use efficiency.
Disclosure of Invention
Accordingly, the present invention provides a digital image quality evaluation system that solves or partially solves the above-mentioned problems.
In order to achieve the effect of the technical scheme, the technical scheme of the invention is as follows: a digital image quality evaluation system comprises a reading module, a fault identification module, an aperture identification module, an analysis module and a feedback module;
the reading module and the feedback module are sequentially connected with the fault identification module, the diaphragm identification module and the analysis module;
the input quantity of the digital image quality evaluation system is a color image shot by a digital camera, and is firstly input into a reading module for preprocessing, wherein the preprocessing process comprises the following steps: each pixel point on the color image has a unique coordinate (x, y), and the color value of the pixel point is read and converted into a gray value to obtain a gray image; secondly, dividing gradient domains from the gray level image, and numbering each gradient domain, wherein the dividing method comprises the following steps: if the pixel points exist at the upper, lower, left and right positions of the pixel point, the pixel point is defined as a central pixel point, a gradient domain comprises the central pixel point and four adjacent pixel points thereof, the value (| l-c | + | r-c | + | u-c | + | d-c |)/4c is calculated in the gradient domain as a gradient sum, c is the gray value of the central pixel point, u, d, l and r are the gray values of the upper, lower, left and right adjacent pixel points of the central pixel point respectively, and the value ranges of c, l, r, u and d are all [0, 255 ]; finally, 5 cross sections are defined, the serial numbers are respectively 1-5, pixel points with gradient sum ranges of [0.5, 0.6 ], [0.6, 0.7 ], [0.7, 0.8 ], [0.8, 0.9 ], [0.9, + ∞ ] are sequentially reserved on the cross sections, and the cross sections are transmitted to a fault identification module and a diaphragm identification module;
the fault identification module is used for counting the number of color faults of the color image, and the counting process is as follows: the cross sections respectively correspond to a fault coefficient, and the fault coefficients of the cross sections with the numbers of 1-5 sequentially take values of 0.5, 0.6, 0.7, 0.8 and 0.9; firstly, a linear fitting is carried out on all pixel points on each cross section by utilizing a least square method in sequence to obtain a linear function, and then a formula I is used for calculating the fault rate:
the formula I is as follows:
Figure BDA0001600310670000021
wherein, delta is the fault rate, is the percentage between 0 and 100 percent and has no unit; gamma is a fault coefficient; i is the serial number of the pixel points on the cross section, and the serial numbers are sequentially numbered from 1 in the row direction, and the value is a positive integer; m is the number of pixel points on the cross section; a is a first order coefficient of a linear function, and b is a constant term of the linear function; x is the abscissa of the pixel point, and y is the ordinate of the pixel point; x is the number ofiIs the abscissa, y, of a pixel point with a sequence number i on the cross sectioniIs the ordinate of the pixel point with serial number i on the cross section;
finally, counting the number of cross sections with the fault rate larger than the fault coefficient and transmitting the number to a feedback module;
the aperture identification module is used for judging the number of color blocks of the color image, and the judgment process is as follows: the cross sections respectively correspond to an aperture coefficient, and the aperture coefficients of the cross sections with the numbers of 1-5 sequentially take values of 0.1, 0.3, 0.5, 0.7 and 0.9; firstly, defining a bottom surface of a color block, wherein the defining method comprises the following steps: comparing the number of pixel points on the cross section with lambda2N, λ is aperture coefficient, N is total number of pixels of color image, and the number of pixels on the cross section is not less than λ2The cross section of the N is defined as the bottom surface of the color block, if the bottom surface of the color block does not exist, the judgment process of the aperture identification module is finished, and if the bottom surface of the color block exists, the judgment is continued; then, color blocks are identified on the bottom surface of each color block, and the identification process is as follows: selecting on the bottom surface of the color block
Figure BDA0001600310670000031
The method comprises the steps that a pixel point with the smallest value is used as a base point, a pixel chain is established by taking the base point as a starting point, the pixel chain is a set of pixel points and is used for storing coordinates of the pixel points, the pixel point which is closest to the last pixel point on the pixel chain is selected on the bottom surface of a color block and is added to the pixel chain, the distance between the two pixel points is calculated by adopting an Euclidean distance formula, when the number of the pixel points on the pixel chain reaches 2 times of the total number of the pixel points on the bottom surface of the color block, the addition of the pixel points is stopped, the pixel points on the pixel chain are traversed, the unrepeated pixel points are counted as effective pixel points, and if the number of the effective pixel points exceeds 80% of the total number of the pixel points on the bottom surface of the color block, the color block exists on the bottom surface of the color block; finally, counting the number of the bottom surfaces of the color blocks and transmitting the number to a feedback module;
the feedback module judges the analysis result after receiving the analysis results of the fault identification module and the diaphragm identification module, if the analysis result and the analysis result are both 0, the feedback module sends a starting instruction to the reading module, and the reading module sends the preprocessing information of the color image to the analysis module after receiving the starting instruction; if the two are not 0, the analysis process of the digital image quality evaluation system is finished, and the feedback module only receives the analysis results of the fault identification module and the aperture identification module;
after the analysis module receives the preprocessing information of the color image, the definition and the gray scale information entropy of the color image can be calculated and transmitted to the feedback module as an analysis result, and the gray scale information entropy is calculated by a formula II:
the formula II is as follows:
Figure BDA0001600310670000032
wherein H is a gray scale information entropy; t is the number of the gradient domain, and the value is a positive integer; r is the number of gradient domains and takes the value of a positive integer; j is a gray value with a value range of [0, 255]](ii) a S is a gradient sum; i is the probability of pixel appearance, and the value range is [0, 1 ]]；StIs the sum of the gradients of the gradient field numbered t; i (t, j) is the probability of the occurrence of the pixel point with the gray value j in the gradient domain with the number of t, and the value range is [0, 1]；
Calculating the definition by using a formula three:
the formula III is as follows:
Figure BDA0001600310670000033
wherein T is sharpness; z and h are adjustment coefficients which are adjusted by workers according to requirements; t is the number of the gradient domain; r is the number of gradient domains; u is the gradient of the central pixel point in the row direction, v is the gradient of the central pixel point in the column direction, utIs the gradient, v, of the central pixel point of the gradient field with the number t in the row directiontIs the gradient in the column direction of the central pixel point of the gradient field with the number t, u, v, ut、vtAre all rational numbers; s is a gradient sum, StIs the sum of the gradients of the gradient field numbered t; d is the number of lines of pixel points of the color image, F is the number of columns of pixel points of the color image, and D, F are positive integers;
the feedback module is provided with a display screen for displaying the received analysis result; if the analysis results of the fault identification module and the aperture identification module are not both 0, the feedback module only receives the analysis results of the fault identification module and the aperture identification module; and if the analysis results of the fault identification module and the diaphragm identification module are both 0, the feedback module receives the analysis results of the fault identification module, the diaphragm identification module and the analysis module.
The beneficial results of the invention are as follows: the invention provides a digital image quality evaluation system, which is provided with a reading module, a fault identification module, an aperture identification module, an analysis module and a feedback module, wherein the reading module, the fault identification module, the aperture identification module, the analysis module and the feedback module can respectively preprocess a color image acquired by a digital camera, identify color faults and color blocks possibly appearing in the image through an intelligent algorithm, effectively evaluate the definition and the richness of colors of the image, and can make data fully play the value and improve the use efficiency. The evaluation system avoids the image inspection through human vision, reduces man-machine interaction operation, has higher automation degree and more objective and efficient evaluation process.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention more apparent, the present invention is described in detail below with reference to the embodiments. It should be noted that the specific embodiments described herein are only for illustrating the present invention and are not to be construed as limiting the present invention, and products that can achieve the same functions are included in the scope of the present invention. The specific method comprises the following steps:
example 1: this embodiment specifically introduces a commonly used image quality evaluation method, which is as follows:
(1) reference type evaluation index
Mean square error: the blurring degree of the image is determined by comparing the mean square value of the pixel difference values of the reference image and the target image. Generally, the smaller the mean square value, the better the quality of the image.
Peak signal-to-noise ratio: it is commonly used as a measure of the quality of signal reconstruction in the fields of image compression, transmission, reconstruction, etc.
Structural similarity: the method is used for measuring the similarity between two images, and the calculated value is closer to 1 when the similarity between the target image and the original image is higher.
Gradient ringing: according to the image quality evaluation method based on the gradient, the similarity degree of two images is higher when the calculated value is larger.
(2) Non-reference type evaluation index
Definition: the degree of definition of lines and boundaries of each part on the image is an objective reference result for subjective evaluation of image quality by human eyes, but is greatly influenced by noise.
Information entropy: from the perspective of information theory, the method is a way of measuring the richness of image information, and the result shows how much information is carried by the image. Generally, the larger the information entropy of a video is, the more abundant the information content of the video is, and the better the quality of the video is.
Gray level: the brightness difference of pixel points in the remote sensing image is represented, and the more the gray level is, the clearer and more vivid the image hierarchy is. The gray distribution can also be used as a representation of the image quality to a certain extent, the gray range of the image is 0-100, which means that the image is dark and the image quality is relatively poor, and the gray range of the image is 150-250, which means that the image is bright and the image quality is not ideal. Under the condition of underexposure or overexposure of the image, the range of the gray scale of the image is very small, and then a blurred image without gray scale hierarchy can be obtained. The gray level histogram can clearly represent the statistical relationship between each gray level of an image and the frequency corresponding to each gray level pixel, and can reflect the overall situation of the image, such as the average brightness, the contrast, the distribution of each gray level, namely the occurrence frequency and the gray level range, and the like. The gray level histogram can better reflect the gray level distribution condition of an image, a digital image should use all gray levels, if the utilization degree of the gray levels is too low, the quantization interval is increased, if the brightness of the image exceeds the processing range of the digital quantizer, the gray levels beyond the range are set as the minimum value or the maximum value, more pixels are gathered at two ends of the histogram, great difference is caused between the two ends of the histogram and real image information, and effective information of the image cannot be reflected. In the practical application process, the input influence type needs to be analyzed first, and the linear stretching preprocessing is performed on the image data of the non-byte type. The shape of the gray level histogram of the stretched image is basically the same as the original effect, but the image is smoother, and generally, the gray level distribution evaluation can be performed if the remote sensing image processed by the linear stretching function does not change the original gray level distribution characteristics. The smaller the root mean square value of the grey level, the more uniform the influence on the grey distribution.
Amount of video information: the remote sensing image information is a specific representation form of the information. The information is carried along with the energy and the change of the material and the energy, has the characteristic of knowledge, and human beings carry out various cognitive activities by means of the acquired information. According to this, the remote sensing image information is various quantities with specific meaning and utilization value obtained from the image by the operator, and different forms and different degrees of knowledge of the image target attribute.
Example 2: this embodiment specifically exemplifies the structure of the digital image quality evaluation system, as follows:
the digital image quality evaluation system comprises: the device comprises a reading module, a fault identification module, an aperture identification module, an analysis module and a feedback module;
the reading module and the feedback module are sequentially connected with the fault identification module, the diaphragm identification module and the analysis module;
the input quantity of the digital image quality evaluation system is a color image shot by a digital camera, and is firstly input into a reading module for preprocessing, wherein the preprocessing process comprises the following steps: each pixel point on the color image has a unique coordinate (x, y), and the color value of the pixel point is read and converted into a gray value to obtain a gray image; secondly, dividing gradient domains from the gray level image, and numbering each gradient domain, wherein the dividing method comprises the following steps: if the pixel points exist at the upper, lower, left and right positions of the pixel point, the pixel point is defined as a central pixel point, a gradient domain comprises the central pixel point and four adjacent pixel points thereof, the value (| l-c | + | r-c | + | u-c | + | d-c |)/4c is calculated in the gradient domain as a gradient sum, c is the gray value of the central pixel point, u, d, l and r are the gray values of the upper, lower, left and right adjacent pixel points of the central pixel point respectively, and the value ranges of c, l, r, u and d are all [0, 255 ]; finally, 5 cross sections are defined, the serial numbers are respectively 1-5, pixel points with gradient sum ranges of [0.5, 0.6 ], [0.6, 0.7 ], [0.7, 0.8 ], [0.8, 0.9 ], [0.9, + ∞ ] are sequentially reserved on the cross sections, and the cross sections are transmitted to a fault identification module and a diaphragm identification module;
the fault identification module is used for counting the number of color faults of the color image, and the counting process is as follows: the cross sections respectively correspond to a fault coefficient, and the fault coefficients of the cross sections with the numbers of 1-5 sequentially take values of 0.5, 0.6, 0.7, 0.8 and 0.9; firstly, a linear fitting is carried out on all pixel points on each cross section by utilizing a least square method in sequence to obtain a linear function, and then a formula I is used for calculating the fault rate:
the formula I is as follows:
Figure BDA0001600310670000061
wherein, delta is the fault rate, is the percentage between 0 and 100 percent and has no unit; gamma is a fault coefficient; i is the serial number of the pixel points on the cross section, and the serial numbers are sequentially numbered from 1 in the row direction, and the value is a positive integer; m is the number of pixel points on the cross section; a is a first order coefficient of a linear function, and b is a constant term of the linear function; x is the abscissa of the pixel point, and y is the ordinate of the pixel point; x is the number ofiIs the abscissa, y, of a pixel point with a sequence number i on the cross sectioniIs the ordinate of the pixel point with serial number i on the cross section;
finally, counting the number of cross sections with the fault rate larger than the fault coefficient and transmitting the number to a feedback module;
the aperture identification module is used for judging the number of color blocks of the color image, and the judgment process is as follows: the cross sections respectively correspond to an aperture coefficient, and the aperture coefficients of the cross sections with the numbers of 1-5 sequentially take values of 0.1, 0.3, 0.5, 0.7 and 0.9; first, define the color block baseThe definition method is as follows: comparing the number of pixel points on the cross section with lambda2N, λ is aperture coefficient, N is total number of pixels of color image, and the number of pixels on the cross section is not less than λ2The cross section of the N is defined as the bottom surface of the color block, if the bottom surface of the color block does not exist, the judgment process of the aperture identification module is finished, and if the bottom surface of the color block exists, the judgment is continued; then, color blocks are identified on the bottom surface of each color block, and the identification process is as follows: selecting on the bottom surface of the color block
Figure BDA0001600310670000071
The method comprises the steps that a pixel point with the smallest value is used as a base point, a pixel chain is established by taking the base point as a starting point, the pixel chain is a set of pixel points and is used for storing coordinates of the pixel points, the pixel point which is closest to the last pixel point on the pixel chain is selected on the bottom surface of a color block and is added to the pixel chain, the distance between the two pixel points is calculated by adopting an Euclidean distance formula, when the number of the pixel points on the pixel chain reaches 2 times of the total number of the pixel points on the bottom surface of the color block, the addition of the pixel points is stopped, the pixel points on the pixel chain are traversed, the unrepeated pixel points are counted as effective pixel points, and if the number of the effective pixel points exceeds 80% of the total number of the pixel points on the bottom surface of the color block, the color block exists on the bottom surface of the color block; finally, counting the number of the bottom surfaces of the color blocks and transmitting the number to a feedback module;
the feedback module judges the analysis result after receiving the analysis results of the fault identification module and the diaphragm identification module, if the analysis result and the analysis result are both 0, the feedback module sends a starting instruction to the reading module, and the reading module sends the preprocessing information of the color image to the analysis module after receiving the starting instruction; if the two are not 0, the analysis process of the digital image quality evaluation system is finished, and the feedback module only receives the analysis results of the fault identification module and the aperture identification module;
after the analysis module receives the preprocessing information of the color image, the definition and the gray scale information entropy of the color image can be calculated and transmitted to the feedback module as an analysis result, and the gray scale information entropy is calculated by a formula II:
the formula II is as follows:
Figure BDA0001600310670000081
wherein H is a gray scale information entropy; t is the number of the gradient domain, and the value is a positive integer; r is the number of gradient domains and takes the value of a positive integer; j is a gray value with a value range of [0, 255]](ii) a S is a gradient sum; i is the probability of pixel appearance, and the value range is [0, 1 ]]；StIs the sum of the gradients of the gradient field numbered t; i (t, j) is the probability of the occurrence of the pixel point with the gray value j in the gradient domain with the number of t, and the value range is [0, 1]；
Calculating the definition by using a formula three:
the formula III is as follows:
Figure BDA0001600310670000082
wherein T is sharpness; z and h are adjustment coefficients which are adjusted by workers according to requirements; t is the number of the gradient domain; r is the number of gradient domains; u is the gradient of the central pixel point in the row direction, v is the gradient of the central pixel point in the column direction, utIs the gradient, v, of the central pixel point of the gradient field with the number t in the row directiontIs the gradient in the column direction of the central pixel point of the gradient field with the number t, u, v, ut、vtAre all rational numbers; s is a gradient sum, StIs the sum of the gradients of the gradient field numbered t; d is the number of lines of pixel points of the color image, F is the number of columns of pixel points of the color image, and D, F are positive integers;
the feedback module is provided with a display screen for displaying the received analysis result; if the analysis results of the fault identification module and the aperture identification module are not both 0, the feedback module only receives the analysis results of the fault identification module and the aperture identification module; and if the analysis results of the fault identification module and the diaphragm identification module are both 0, the feedback module receives the analysis results of the fault identification module, the diaphragm identification module and the analysis module.
The beneficial results of the invention are as follows: the invention provides a digital image quality evaluation system, which is provided with a reading module, a fault identification module, an aperture identification module, an analysis module and a feedback module, wherein the reading module, the fault identification module, the aperture identification module, the analysis module and the feedback module can respectively preprocess a color image acquired by a digital camera, identify color faults and color blocks possibly appearing in the image through an intelligent algorithm, effectively evaluate the definition and the richness of colors of the image, and can make data fully play the value and improve the use efficiency. The evaluation system avoids the image inspection through human vision, reduces man-machine interaction operation, has higher automation degree and more objective and efficient evaluation process.
The above description is only for the preferred embodiment of the present invention, and should not be used to limit the scope of the claims of the present invention. While the foregoing description will be understood and appreciated by those skilled in the relevant art, other equivalents may be made thereto without departing from the scope of the claims.

Claims (1)

1. A digital image quality evaluation system, comprising: the device comprises a reading module, a fault identification module, an aperture identification module, an analysis module and a feedback module;
the reading module and the feedback module are sequentially connected with the fault identification module, the aperture identification module and the analysis module;
the input quantity of the digital image quality evaluation system is a color image shot by a digital camera, and is firstly input into the reading module for preprocessing, and the preprocessing process comprises the following steps: each pixel point on the color image has a unique coordinate (x, y), and the color value of the pixel point is read and converted into a gray value to obtain a gray image; secondly, dividing gradient domains from the gray level image, and numbering each gradient domain, wherein the dividing method comprises the following steps: if the pixel points exist at the upper, lower, left and right positions of the pixel point, the pixel point is defined as a central pixel point, the gradient domain comprises the central pixel point and four adjacent pixel points thereof, the value (| l-c | + | r-c | + | u-c | + | d-c |)/4c is calculated in the gradient domain as a gradient sum, c is the gray value of the central pixel point, u, d, l and r are the gray values of the upper, lower, left and right adjacent pixel points of the central pixel point respectively, and the value ranges of c, l, r, u and d are all [0, 255 ]; finally, 5 cross sections are defined, the serial numbers are respectively 1-5, pixel points with gradient sum ranges of [0.5, 0.6 ], [0.6, 0.7 ], [0.7, 0.8 ], [0.8, 0.9 ], [0.9, + ∞) ] are sequentially and respectively reserved on the cross sections, and the cross sections are transmitted to the fault identification module and the aperture identification module;
the fault identification module is used for counting the number of the color faults of the color image, and the counting process is as follows: the cross sections respectively correspond to a fault coefficient, and the fault coefficients of the cross sections with the numbers of 1-5 sequentially take values of 0.5, 0.6, 0.7, 0.8 and 0.9; firstly, on each cross section, a least square method is utilized to perform linear fitting on all pixel points to obtain a linear function, and a formula I is used for calculating the fault rate:
the formula I is as follows:
Figure FDA0001600310660000011
wherein, delta is the fault rate, is a percentage between 0 and 100 percent and has no unit; gamma is the fault coefficient; i is the serial number of the pixel points on the cross section, and the serial numbers are sequentially numbered from 1 in the row direction, and the value is a positive integer; m is the number of pixel points on the cross section; a is a coefficient of a first order term of the linear function, and b is a constant term of the linear function; the x is the abscissa of the pixel point, and the y is the ordinate of the pixel point; x is the number ofiIs the abscissa, y, of a pixel point with a sequence number i on the cross sectioniIs the ordinate of the pixel point with serial number i on the cross section;
finally, counting the number of the cross sections with the fault rate larger than the fault coefficient and transmitting the number to the feedback module;
the aperture identification module is used for judging the number of color blocks of the color image, and the judgment process is as follows: the cross sections respectively correspond to an aperture coefficient, and the aperture coefficients of the cross sections with the numbers of 1-5 sequentially take values of 0.1, 0.3, 0.5, 0.7 and 0.9; firstly, defining a bottom surface of a color block, wherein the defining method comprises the following steps:comparing the number of pixel points on the cross section with lambda2N, wherein lambda is the aperture coefficient, N is the total number of pixel points of the color image, and the number of the pixel points on the cross section is not less than lambda2The cross section of N is defined as the bottom surface of the color block, if the bottom surface of the color block does not exist, the judgment process of the aperture identification module is finished, and if the bottom surface of the color block exists, the judgment is continued; then, identifying color blocks on the bottom surface of each color block, wherein the identification process is as follows: selecting on the bottom surface of the color block
Figure FDA0001600310660000021
The method comprises the steps that a pixel point with the smallest value is used as a base point, a pixel chain is established by taking the base point as a starting point, the pixel chain is a set of pixel points and is used for storing coordinates of the pixel points, the pixel point which is closest to the last pixel point on the pixel chain is selected on the bottom surface of a color block and is added to the pixel chain, the distance between the two pixel points is calculated by adopting an Euclidean distance formula, when the number of the pixel points on the pixel chain reaches 2 times of the total number of the pixel points on the bottom surface of the color block, the pixel points are stopped being added, the pixel points on the pixel chain are traversed, the unrepeated pixel points are counted as effective pixel points, and if the number of the effective pixel points exceeds 80% of the total number of the pixel points on the bottom surface of the color block, the color block exists on the bottom surface of the color block; finally, counting the number of the bottom surfaces of the color blocks and transmitting the number to the feedback module;
the feedback module judges the analysis results after receiving the analysis results of the fault identification module and the aperture identification module, if the analysis results are both 0, the feedback module sends a starting instruction to the reading module, and the reading module sends the preprocessing information of the color image to the analysis module after receiving the starting instruction; if the two are not 0, the analysis process of the digital image quality evaluation system is finished, and the feedback module only receives the analysis results of the fault identification module and the aperture identification module;
after the analysis module receives the preprocessing information of the color image, the analysis module can calculate the definition and the gray scale information entropy of the color image as analysis results and transmit the analysis results to the feedback module, and the gray scale information entropy is calculated by using a formula II:
the formula II is as follows:
Figure FDA0001600310660000031
wherein H is the gray scale information entropy; t is the number of the gradient domain, and the value is a positive integer; r is the number of gradient domains and takes the value of a positive integer; j is the gray value and the value range is [0, 255]](ii) a S is the sum of the gradients; i is the probability of pixel appearance, and the value range is [0, 1 ]]；StIs the sum of the gradients of the gradient field numbered t; i (t, j) is the probability of the occurrence of the pixel point with the gray value j in the gradient domain with the number of t, and the value range is [0, 1]；
Calculating the sharpness using equation three:
the formula III is as follows:
Figure FDA0001600310660000032
wherein T is the sharpness; z and h are adjustment coefficients which are adjusted by workers according to requirements; t is the number of the gradient domain; r is the number of the gradient domains; u is the gradient of the central pixel point in the row direction, v is the gradient of the central pixel point in the column direction, utIs the gradient, v, of the central pixel point of the gradient field with the number t in the row directiontIs the gradient of the central pixel point of the gradient domain with the number of t in the column direction, the u, the v and the utSaid vtAre all rational numbers; s is the sum of the gradients, StIs the sum of the gradients of the gradient field numbered t; d is the number of rows of pixel points of the color image, F is the number of columns of pixel points of the color image, and both D and F are positive integers;
the feedback module is provided with a display screen and is used for displaying the received analysis result; if the analysis results of the fault identification module and the aperture identification module are not both 0, the feedback module only receives the analysis results of the fault identification module and the aperture identification module; and if the analysis results of the fault identification module and the aperture identification module are both 0, the feedback module receives the analysis results of the fault identification module, the aperture identification module and the analysis module.
CN201810222242.3A 2018-03-19 2018-03-19 Digital image quality evaluation system Active CN108492289B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810222242.3A CN108492289B (en) 2018-03-19 2018-03-19 Digital image quality evaluation system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810222242.3A CN108492289B (en) 2018-03-19 2018-03-19 Digital image quality evaluation system

Publications (2)

Publication Number Publication Date
CN108492289A CN108492289A (en) 2018-09-04
CN108492289B true CN108492289B (en) 2021-09-10

Family

ID=63339712

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810222242.3A Active CN108492289B (en) 2018-03-19 2018-03-19 Digital image quality evaluation system

Country Status (1)

Country Link
CN (1) CN108492289B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112200807B (en) * 2020-12-07 2021-02-12 光谷技术有限公司 Video quality diagnosis method and system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006078928A3 (en) * 2005-01-18 2006-11-23 Trestle Corp System and method for creating variable quality images of a slide
CN104036498A (en) * 2014-05-28 2014-09-10 杭州电子科技大学 Fast evaluation method of OCT image quality based on layer by layer classification
EP2902966A1 (en) * 2014-02-03 2015-08-05 Prosper Creative Co., Ltd. Image inspecting apparatus and image inspecting program
CN106296669A (en) * 2016-08-01 2017-01-04 微梦创科网络科技（中国）有限公司 A kind of image quality evaluating method and device
CN106920231A (en) * 2017-02-17 2017-07-04 北京理工大学 A kind of remote sensing image clouds appraisal procedure based on full-colour image statistical nature
CN107292866A (en) * 2017-05-17 2017-10-24 浙江科技学院 A kind of method for objectively evaluating image quality based on relative gradient
CN107545570A (en) * 2017-08-31 2018-01-05 中国地质大学（武汉） A kind of reconstructed image quality evaluation method of half reference chart

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008137667A1 (en) * 2007-05-04 2008-11-13 Aperio Technologies, Inc. System and method for quality assurance in pathology
US9773307B2 (en) * 2010-01-25 2017-09-26 Amcad Biomed Corporation Quantification and imaging methods and system of the echo texture feature

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006078928A3 (en) * 2005-01-18 2006-11-23 Trestle Corp System and method for creating variable quality images of a slide
EP2902966A1 (en) * 2014-02-03 2015-08-05 Prosper Creative Co., Ltd. Image inspecting apparatus and image inspecting program
CN104036498A (en) * 2014-05-28 2014-09-10 杭州电子科技大学 Fast evaluation method of OCT image quality based on layer by layer classification
CN106296669A (en) * 2016-08-01 2017-01-04 微梦创科网络科技（中国）有限公司 A kind of image quality evaluating method and device
CN106920231A (en) * 2017-02-17 2017-07-04 北京理工大学 A kind of remote sensing image clouds appraisal procedure based on full-colour image statistical nature
CN107292866A (en) * 2017-05-17 2017-10-24 浙江科技学院 A kind of method for objectively evaluating image quality based on relative gradient
CN107545570A (en) * 2017-08-31 2018-01-05 中国地质大学（武汉） A kind of reconstructed image quality evaluation method of half reference chart

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Evaluate mobile video quality with LTE radio access network parameters;王飞 等;《Journal of Beijing Institute of Technology》;20161231;第25卷(第4期);第553页-第561页 *
基于调制传递函数的数码影像质量评价研究;刘佳琪 等;《计算机应用与软件》;20170331;第34卷(第3期);第160页-第164页 *

Also Published As

Publication number Publication date
CN108492289A (en) 2018-09-04

Similar Documents

Publication Publication Date Title
CN100559881C (en) A kind of method for evaluating video quality based on artificial neural net
CN100563303C (en) Image processing equipment, image capture device and image processing method
CN108109147B (en) No-reference quality evaluation method for blurred image
CN101562675B (en) No-reference image quality evaluation method based on Contourlet transform
CN107464222A (en) Based on tensor space without with reference to high dynamic range images method for evaluating objective quality
CN112767385B (en) No-reference image quality evaluation method based on significance strategy and feature fusion
CN112348754B (en) Low-illumination color image enhancement method and device
CN108038839A (en) Twisted-pair feeder lay real-time detection method on a kind of flow production line
Chen et al. Blind quality index for tone-mapped images based on luminance partition
CN112200807B (en) Video quality diagnosis method and system
CN111598837A (en) Full-reference image quality evaluation method and system suitable for visual two-dimensional code
CN107545570A (en) A kind of reconstructed image quality evaluation method of half reference chart
CN110044485B (en) Image type fabric color measuring method
CN108492289B (en) Digital image quality evaluation system
CN112651945A (en) Multi-feature-based multi-exposure image perception quality evaluation method
Triantaphillidou et al. Image quality comparison between JPEG and JPEG2000. II. Scene dependency, scene analysis, and classification
CN111127386B (en) Image quality evaluation method based on deep learning
CN117314787A (en) Underwater image enhancement method based on self-adaptive multi-scale fusion and attention mechanism
CN108389200A (en) Halftone Image quality evaluating method and system based on texture visual characteristic
CN110136105B (en) Method for evaluating definition of same content image based on variance and smoothness
CN112950479B (en) Image gray level region stretching algorithm
CN116091312A (en) Low-contrast image joint enhancement and super-resolution reconstruction method
CN108288267A (en) A kind of scanning electron microscope (SEM) photograph image sharpness based on dark is without with reference to evaluation method
CN114519670A (en) Image super-resolution reconstruction method based on convolutional neural network
CN111076815B (en) Hyperspectral image non-uniformity correction method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
PE01 Entry into force of the registration of the contract for pledge of patent right
PE01 Entry into force of the registration of the contract for pledge of patent right

Denomination of invention: A Digital Image Quality Evaluation System

Effective date of registration: 20230504

Granted publication date: 20210910

Pledgee: Shanghai Rural Commercial Bank Co.,Ltd. Qingpu sub branch

Pledgor: SHANGHAI BAOYI DESIGN AND PRODUCTION Corp.

Registration number: Y2023310000153