CN114494066A - Human image sharpening method, device, equipment and medium based on Hessian filter - Google Patents

Human image sharpening method, device, equipment and medium based on Hessian filter Download PDF

Info

Publication number
CN114494066A
CN114494066A CN202210099056.1A CN202210099056A CN114494066A CN 114494066 A CN114494066 A CN 114494066A CN 202210099056 A CN202210099056 A CN 202210099056A CN 114494066 A CN114494066 A CN 114494066A
Authority
CN
China
Prior art keywords
image
original image
label
sharpening
information
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.)
Pending
Application number
CN202210099056.1A
Other languages
Chinese (zh)
Inventor
林奇
周凡
杨铸
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.)
Xiamen Zhenjing Technology Co ltd
Original Assignee
Xiamen Zhenjing Technology 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 Xiamen Zhenjing Technology Co ltd filed Critical Xiamen Zhenjing Technology Co ltd
Priority to CN202210099056.1A priority Critical patent/CN114494066A/en
Publication of CN114494066A publication Critical patent/CN114494066A/en
Pending legal-status Critical Current

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
    • G06T5/20—Image enhancement or restoration using local operators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20221—Image fusion; Image merging

Landscapes

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

Abstract

The invention discloses a human image sharpening method, a human image sharpening device, human image sharpening equipment and a human image sharpening medium based on a Hessian filter, and belongs to the technical field of image processing; the method specifically comprises the following steps: acquiring an original image; carrying out low-pass filtering on the original image to obtain first original image information and obtain a first label image; carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image; filtering the original image Hessian to obtain third original image information and obtain a third label image; carrying out linear fusion on the original image and the second label image according to the third label image to obtain a result image; the method separately processes the structural texture and the skin low-frequency texture, detects the structural texture as a guide mask by using a Hessian filter, and then strengthens the structural texture of the five sense organs to achieve a sharpening effect while not strengthening skin flaws; the method has the advantages that the image noise of the low-frequency skin texture is not enhanced while the structural texture of the five sense organs of the portrait is enhanced, and the effect of improving the sharpness of the portrait is very effective.

Description

Human image sharpening method, device, equipment and medium based on Hessian filter
Technical Field
The invention relates to the technical field of image processing, in particular to a portrait sharpening method, a portrait sharpening device, portrait sharpening equipment and portrait sharpening media based on a Hessian filter.
Background
Image sharpening (image sharpening) is to compensate the outline of an image, enhance the edge of the image and the part with jump gray level, make the image become clear, and is divided into two types, namely spatial domain processing and frequency domain processing. Image sharpening is to highlight edges, contours, or features of some linear target elements of a terrain on an image. This filtering method improves the contrast between the feature edges and the surrounding picture elements and is therefore also referred to as edge enhancement. Some image enhancement sharpening software on the market at present enables post-processing image sharpening to be very fast or convenient, various convenient mobile hard application software also appears on mobile equipment along with the development of mobile internet, but it can be found that algorithms of the mobile hard application software can always sharpen images without distinction, so that flaws of skin are amplified, namely, global sharpening is achieved; although the details of the face are improved, the corresponding facial blemish details are also highlighted. Therefore, the applicant proposes a portrait sharpening method based on a Hessian filter, which technically processes structural textures and low-frequency textures of a portrait separately, detects the structural textures as a guide mask by using the Hessian filter, and then strengthens the structural textures of five sense organs to achieve a sharpening effect while not strengthening skin flaws.
The method and the device search the prior art in the process of item establishment, and search two contrast files for image sharpening of the Nubian technology company Limited.
Comparison document 1: CN201510716021.8 discloses an image sharpening method, comprising: the method comprises the steps that a mobile terminal obtains an original RGB image and converts the original RGB image into a YCbCr space image; acquiring intensity information of each pixel point of a brightness component in the YCbCr space image, and determining a black edge, a white edge and corresponding sharpening intensity according to the intensity information of each pixel point; respectively sharpening the black edge and the white edge according to the sharpening strengths corresponding to the black edge and the white edge; and converting the sharpened YCbCr space image into a new RGB image to obtain a sharpened image. The invention also discloses a mobile terminal for sharpening the image. The invention realizes the sharpening of the black edge and the white edge in a corresponding degree, and improves the sharpening effect of the image. The method is the closest prior art of the application, and sharpening processing is carried out on the image in a YUV format; the brightness channel is also adopted as an incision point to sharpen the image; but the difference is that the application utilizes contrast to determine the texture information of the image, and utilizes the contrast to control the strength of sharpening; the distinguishing technical characteristics of the application are that structural textures are obtained by obtaining texture information of an image, and then the structural textures are used as a guide to strengthen the structural textures of the five sense organs; compared with the comparison document 2, the substantial progress and the salient feature of the application are that the comparison document 2 combines the structural texture and the skin low-frequency texture after extracting the image information by using the brightness channel. According to the method, the portrait structural texture and the skin low-frequency texture are processed separately, the structural texture is detected by using a Hessian filter to serve as a guide mask, and then the five-sense structural texture is enhanced to achieve a sharpening effect while skin flaws are not enhanced.
Comparison document 2: CN201611248625.5 discloses an image sharpening method, which includes: identifying face information contained in a shot image according to a face identification method, and extracting biological feature information in the face information; determining that an image corresponding to biological characteristic information in the shot image is a first sharpened area, and determining that an image except the image corresponding to the biological characteristic information in the shot image is a second sharpened area; receiving a first sharpening instruction aiming at the first sharpening area, and acquiring a first sharpened image; receiving a second sharpening instruction aiming at the second sharpening area, and acquiring a second sharpened image; and fusing the first sharpened image and the second sharpened image to obtain a fused sharpened image. The embodiment of the invention also discloses an image sharpening terminal. The problem of excessive sharpening when the portrait is sharpened is solved, and the reasonability of the portrait sharpening degree is ensured. The contrast discovery actually shows that the method of the application is to obtain the face information, divide the image into a face area and a non-face area by taking the face information as a boundary, sharpen the two parts respectively, and finally fuse the two parts to obtain a sharpening result. The problem of sharpening excessive when the portrait is sharpened is solved, the reasonability of the portrait sharpening degree is ensured, but the problem is that the details of the facial flaws such as acne marks, pockmarks, freckles and the like are correspondingly strengthened, and although the details of the facial flaws are improved, the corresponding facial flaw details are also highlighted. The combination of the comparison document 1 and the comparison document 2 also has difficulty in solving the problem that the blemish of the skin is enlarged.
Disclosure of Invention
Technical scheme (I)
The invention is realized by the following technical scheme: a human image sharpening method based on a Hessian filter specifically comprises the following steps:
acquiring an original image;
carrying out low-pass filtering on a brightness channel of an original image to obtain first original image information and obtain a first label image;
carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image;
filtering the original image Hessian to obtain third original image information and obtain a third label image;
and performing linear fusion on the original image and the second label image according to the third label image to obtain a result image.
As a further explanation of the above scheme, the obtaining mode of the original image includes obtaining by the intelligent terminal device.
As a further illustration of the above scheme, the raw image format comprises a YUV format.
As a further explanation of the above solution, the first original image information includes original image texture structure information.
As a further explanation of the above scheme, the low-pass filtering employs gaussian filtering.
As a further explanation of the above scheme, the second original image information includes original image texture information.
As a further explanation of the above scheme, the original image Hessian filtering calculation method is as follows:
Figure BDA0003491743230000041
wherein G (X, s) is a Gaussian distribution weight,
Figure BDA0003491743230000042
l (X, s) is the image at point XoTaylor expansion of (a).
As a further explanation of the above solution, the third original image information includes structural texture information of the original image.
As a further illustration of the above scheme, the calculation formula of the linear fusion is as follows:
Figure BDA0003491743230000051
in the formula, DstA result graph showing the output, SrcRepresenting an input original image, IDetailRepresenting second original image information, IMaskRepresenting third original image information.
The invention also provides a human image sharpening device based on the Hessian filter, which comprises:
the acquisition unit is used for acquiring an image to be original;
a processing unit: the system comprises a brightness channel low-pass filter, a first label image acquisition module, a second label image acquisition module and a third label image acquisition module, wherein the brightness channel low-pass filter is used for acquiring first original image information of an original image to acquire a first label image; carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image; filtering the original image Hessian to obtain third original image information and obtain a third label image; carrying out linear fusion on the original image and the second label image according to the third label image to obtain a result image;
a preview unit: for previewing the result map output by the processing unit.
The invention also provides human image sharpening equipment based on the Hessian filter, which comprises a processor, a memory and a computer program stored in the memory, wherein the computer program can be executed by the processor to realize the human image sharpening method based on the Hessian filter.
The invention further provides a computer-readable storage medium, which is characterized by comprising a stored computer program, wherein when the computer program runs, a device where the computer-readable storage medium is located is controlled to execute the human image sharpening method based on the Hessian filter.
(III) advantageous effects
Compared with the prior art, the invention has the following beneficial effects: the method separately processes the structural texture and the skin low-frequency texture, detects the structural texture as a guide mask by using a Hessian filter, and then strengthens the structural texture of the five sense organs to achieve a sharpening effect while not strengthening skin flaws; the method has the advantages that the image noise of the low-frequency skin texture is not enhanced while the structural texture of the five sense organs of the portrait is enhanced, and the effect of improving the sharpness of the portrait is very effective. Different information of the picture is processed separately, so that the picture can be processed more finely than a common unified sharpening scheme, and flaws and noise of the face cannot be enhanced.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic flow chart of an embodiment of the present invention;
FIG. 2 is an original image obtained by an embodiment of the present invention;
FIG. 3 is a structural texture image extracted according to an embodiment of the present invention;
FIG. 4 is a graph of an output of an embodiment of the present invention.
Detailed Description
Example 1
Referring to fig. 1, a method for sharpening a human image based on a Hessian filter specifically includes:
obtaining an original image, please refer to fig. 2; it should be further noted that the original image does not limit the format of the picture, the hardware and the software of the picture source. In this embodiment, the original image format is preferably an image in YUV format, which is a color coding method. Are often used in various video processing components. YUV allows for reduced bandwidth of chrominance in view of human perception when encoding photos or videos. YUV is a kind of compiled true-color space (color space), and the proper terms such as Y' UV, YUV, YCbCr, YPbPr, etc. may be called YUV, overlapping with each other. "Y" represents brightness (Luma) or gray scale value, and "U" and "V" represent Chroma (Chroma or Chroma) and are used to describe the color and saturation of the image for specifying the color of the pixel. The YUV coding format has the advantage over the RGB coding format that the unit pixels of the YUV coding format use less memory, but the two representations are identical, and on a device with width w h, RGB represents that the coding occupies bytes w h 3, and YUV represents that the temporary memory is w h + w h/4 w h 3/2.
It needs to be further explained that the obtaining mode of the original image includes the obtaining of the intelligent terminal device; the intelligent terminal device may include a smart phone, a tablet computer, a notebook, a wearable device, a vehicle-mounted intelligent terminal, a video phone, a conference terminal, and the like. The present embodiment assumes that the terminal is a smartphone, but those skilled in the art will appreciate that the configuration according to the embodiment of the present invention can be applied to a fixed-type terminal in addition to elements particularly used for mobile purposes.
Carrying out low-pass filtering on the original image to obtain first original image information and obtain a first label image; the first original image information comprises original image texture structure information; it should be further noted that, in this step, low-pass filtering is specifically performed on the luminance channel of the original image, so as to obtain texture structure information of the image, which is named as ILowThe low-pass filtering is preferably gaussian filtering, and the formula is as follows:
Figure BDA0003491743230000081
in this embodiment, gaussian filtering is preferably used for filtering, each pixel in the image is scanned by using a luminance channel of the original image, and the weighted average gray value of the pixels in the neighborhood determined by the luminance channel is used to replace the value of the central pixel point of the template. Gaussian filtering is very effective for suppressing noise that follows a normal distribution. If a gaussian filter is used, the system function is smooth, avoiding ringing. The original source of the image of the embodiment is the intelligent terminal, which is limited by the volume of the intelligent terminal, and the area of the sensor is small, so that the noise generated by the sensor under the condition of low illumination or high temperature is mostly Gaussian noise. Therefore, it is appropriate to use gaussian filtering in this embodiment.
The image texture structure information is a visual feature reflecting homogeneity phenomenon in the image, and embodies the surface structure organization arrangement attribute with slow change or periodic change on the surface of an object. Texture has three major landmarks: 1) certain local sequences repeat continuously; 2) non-random arrangement; 3) a substantially uniform continuum within the textured area. Unlike image features such as gray scale, color, etc., texture is represented by the gray scale distribution of pixels and their surrounding spatial neighborhood, i.e., local texture information. In addition, the local texture information is repetitive in different degrees, and is the global texture information. While the texture features represent the properties of global features, it also describes the surface properties of the scene to which the image or image region corresponds. The texture structure information of the image obtained through the above steps aims to draw out quality information in the subsequent steps, and the description will be developed in the subsequent steps.
Carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image; the second original image information includes original image texture information. Subtracting the result graph in the step 2 from the original image in the step 1 to obtain texture information of the image, which is named as IDetail(ii) a It should be further noted that the texture information is texture information of the image, including the texture and detail texture. The texture information is different from the texture structure information in the previous step, and the surface properties corresponding to the image region obtained in the previous step include the texture information in the present step and the original texture informationCorresponding gray scale information of the image, etc.; the step is equivalent to extracting data of the original graph and removing redundant data.
Filtering the original image Hessian to obtain third original image information and obtain a third label image; the third original image information comprises structural texture information of the original image; please refer to fig. 3, Hessian filtering is performed on the original image to obtain the structural texture, which is named as IMask. The purpose of this step is to extract features from the texture structure information, so that the edge features of people, scenes and objects in the image are more prominent, and the interference of other information is reduced.
The original image Hessian filtering calculation method comprises the following steps:
Figure BDA0003491743230000091
wherein G (X, s) is a Gaussian distribution weight,
Figure BDA0003491743230000092
l (X, s) is the image at point XoTaylor expansion of (a). D is the two-dimensional image equal to 2, x is the distance, and γ and s are generally equal to 1.
It should be further noted that the above steps are aimed at recording color and light shadow at low frequency and saving texture details to high frequency; the high frequency and the low frequency are independently adjusted and do not interfere with each other. The low frequency layer is used to control the color shade of the image, and the adjustment does not affect the picture details. The high frequency layer is used to control the detail without changing the color. The aim in the embodiment is to sharpen in the high frequency layer, and simultaneously, the information of the low frequency layer is not processed, so that the image noise of the low frequency skin texture is not enhanced while the structural texture of the portrait is enhanced; however, the information of the high frequency layer is much, such as the structure texture and the detail texture, and the structure texture needs to be filtered out through Hessian filtering. The popular understanding of the structure texture is edge feature extraction, please refer to fig. 3, after the steps are completed, it can be found that the face contour and the facial feature contour are extracted, and the detail texture of the face such as skin is excluded. If detail textures are reserved, the same app algorithm as the existing algorithm is prone to indiscriminately sharpening the human image, so that flaws of the skin are amplified, and the effect of beautifying and having contour details cannot be achieved. The texture is used for guidance and further description in the following steps.
Referring to fig. 4, the original image and the second label image are linearly fused according to the third label image to obtain a result graph, and a calculation formula of the linear fusion is as follows:
Figure BDA0003491743230000101
in the formula, DstA result graph showing the output, SrcRepresenting an input original image, IDetailRepresenting original texture information, IMaskRepresenting a structural texture.
It should be further noted that the purpose of linear fusion is to complete the sharpening step, and it can be known from the above steps that the actual purpose is to extract texture information and structural texture in the high-frequency signal respectively, and the texture information of the image is guided and enhanced by using the position information of the structural texture information as a guide, while the detail texture in the original image is retained. Generally speaking, texture information is extracted through filtering and image deduction, the texture information is stored in a high-frequency layer, and corresponding light shadow and color are reserved in a low-frequency layer; the high frequency and the low frequency are separated, so that the image noise of enhancing the low-frequency skin texture in the sharpening process is avoided; the texture information includes structural texture and detail texture, but the part needing sharpening only has structural texture information. Therefore, the invention adopts Hessian filtering to extract the structure texture information in the original image, and uses the structure texture information as the guide to determine the range and the position which need to be sharpened, thereby not only keeping the original skin state, but also enhancing the edge area of the structure of the five sense organs of the portrait, not only ensuring that the flaws of the skin are sharpened, and also improving the stereoscopic impression of the five sense organs of the face.
A Hessian filter based portrait sharpening device, the device comprising:
the acquisition unit is used for acquiring an image to be original;
a processing unit: the system comprises a first label image acquisition module, a second label image acquisition module, a first label acquisition module and a second label acquisition module, wherein the first label image acquisition module is used for acquiring first original image information by low-pass filtering a brightness channel of an original image to acquire a first label image; carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image; filtering the original image Hessian to obtain third original image information and obtain a third label image; carrying out linear fusion on the original image and the second label image according to the third label image to obtain a result image;
a preview unit: for previewing the result map output by the processing unit.
The invention also provides human image sharpening equipment based on the Hessian filter, which comprises a processor, a memory and a computer program stored in the memory, wherein the computer program can be executed by the processor to realize the human image sharpening method based on the Hessian filter.
The invention further provides a computer-readable storage medium, which is characterized by comprising a stored computer program, wherein when the computer program runs, a device where the computer-readable storage medium is located is controlled to execute the human image sharpening method based on the Hessian filter.
Illustratively, the computer program may be divided into one or more units, which are stored in the memory and executed by the processor to accomplish the present invention. The one or more elements may be a series of computer program instruction segments capable of performing certain functions, which are used to describe the execution of the computer program in a Hessian filter based portrait sharpening device.
The Hessian filter based portrait sharpening device may include, but is not limited to, a processor, a memory. Those skilled in the art will appreciate that the schematic diagram is merely an example of a Hessian filter based portrait sharpening device and does not constitute a limitation of a Hessian filter based portrait sharpening device and may include more or fewer components than shown, or combine certain components, or different components, e.g., the Hessian filter based portrait sharpening device may also include an input-output device, a network access device, a bus, etc.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the control center of the Hessian filter based portrait sharpening device connects the various parts of the entire Hessian filter based portrait sharpening device with various interfaces and lines.
The memory may be used to store the computer programs and/or modules, and the processor may implement the various functions of the Hessian filter-based portrait sharpening device by running or executing the computer programs and/or modules stored in the memory and invoking the data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Wherein, the integrated unit of the human image sharpening device based on the Hessian filter can be stored in a computer readable storage medium if the integrated unit is realized in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc.
The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
It should be noted that the above-described device embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines. One of ordinary skill in the art can understand and implement it without inventive effort.
The embodiments in the above embodiments can be further combined or replaced, and the embodiments are only used for describing the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention, and various changes and modifications made to the technical solution of the present invention by those skilled in the art without departing from the design idea of the present invention belong to the protection scope of the present invention.

Claims (10)

1. A human image sharpening method based on a Hessian filter is characterized by specifically comprising the following steps:
acquiring an original image;
low-pass filtering a brightness channel of an original image to obtain first original image information and obtain a first label image;
carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image;
filtering the original image Hessian to obtain third original image information and obtain a third label image;
and performing linear fusion on the original image and the second label image according to the third label image to obtain a result image.
2. The human image sharpening method based on Hessian filter as claimed in claim 1,
the original image obtaining mode comprises the intelligent terminal equipment obtaining.
3. The human image sharpening method based on Hessian filter as claimed in claim 1,
the first original image information comprises original image texture structure information; the low-pass filtering adopts Gaussian filtering.
4. The human image sharpening method based on Hessian filter as claimed in claim 1,
the second original image information includes original image texture information.
5. The human image sharpening method based on Hessian filter as claimed in claim 1,
the original image Hessian filtering calculation method comprises the following steps:
Figure FDA0003491743220000021
wherein G (X, s) is a Gaussian distribution weight,
Figure FDA0003491743220000022
l (X, s) is the image at point XoTaylor expansion of (a).
6. The human image sharpening method based on Hessian filter as claimed in claim 1,
the third original image information includes structural texture information of the original image.
7. The human image sharpening method based on Hessian filter as claimed in claim 1,
the calculation formula of the linear fusion is as follows:
Figure FDA0003491743220000023
in the formula, DstA result graph showing the output, SrcRepresenting an input original image, IDetailRepresenting second original image information, IMaskRepresenting third original image information.
8. A Hessian filter based portrait sharpening device, the device comprising:
the acquisition unit is used for acquiring an image to be original;
a processing unit: the system comprises a first label image acquisition module, a second label image acquisition module, a first label acquisition module and a second label acquisition module, wherein the first label image acquisition module is used for acquiring first original image information by low-pass filtering a brightness channel of an original image to acquire a first label image; carrying out image subtraction on the original image and the first label image to obtain second original image information and obtain a second label image; filtering the original image Hessian to obtain third original image information and obtain a third label image; carrying out linear fusion on the original image and the second label image according to the third label image to obtain a result image;
a preview unit: for previewing the result map output by the processing unit.
9. A Hessian filter-based portrait sharpening device comprising a processor, a memory, and a computer program stored in the memory, the computer program being executable by the processor to implement a Hessian filter-based portrait sharpening method as recited in any one of claims 1 to 7.
10. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls a device on which the computer-readable storage medium is located to perform a method for human image sharpening based on a Hessian filter according to any one of claims 1 to 7.
CN202210099056.1A 2022-01-27 2022-01-27 Human image sharpening method, device, equipment and medium based on Hessian filter Pending CN114494066A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210099056.1A CN114494066A (en) 2022-01-27 2022-01-27 Human image sharpening method, device, equipment and medium based on Hessian filter

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210099056.1A CN114494066A (en) 2022-01-27 2022-01-27 Human image sharpening method, device, equipment and medium based on Hessian filter

Publications (1)

Publication Number Publication Date
CN114494066A true CN114494066A (en) 2022-05-13

Family

ID=81477101

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210099056.1A Pending CN114494066A (en) 2022-01-27 2022-01-27 Human image sharpening method, device, equipment and medium based on Hessian filter

Country Status (1)

Country Link
CN (1) CN114494066A (en)

Similar Documents

Publication Publication Date Title
US10152781B2 (en) Method for image processing using local statistics convolution
Li et al. Single image de-hazing using globally guided image filtering
Zhang et al. Enhancement and noise reduction of very low light level images
EP2852152B1 (en) Image processing method, apparatus and shooting terminal
Yu et al. Fast single image fog removal using edge-preserving smoothing
KR102523505B1 (en) Method and Apparatus for Inverse Tone Mapping
WO2020125631A1 (en) Video compression method and apparatus, and computer-readable storage medium
WO2018099136A1 (en) Method and device for denoising image with low illumination, and storage medium
CN111784605B (en) Image noise reduction method based on region guidance, computer device and computer readable storage medium
US20140079319A1 (en) Methods for enhancing images and apparatuses using the same
KR100846513B1 (en) Method and apparatus for processing an image
CN107172354B (en) Video processing method and device, electronic equipment and storage medium
US7768529B2 (en) Method and system for image editing
CN108090876B (en) Image processing method and device
CN107424137B (en) Text enhancement method and device, computer device and readable storage medium
US20170154437A1 (en) Image processing apparatus for performing smoothing on human face area
Ancuti et al. Image and video decolorization by fusion
Gao et al. A novel UAV sensing image defogging method
Abebe et al. Towards an automatic correction of over-exposure in photographs: Application to tone-mapping
CN111815729B (en) Real-time skin beautifying method, device, equipment and computer storage medium
Hu et al. Kinect depth map based enhancement for low light surveillance image
EP2890109A1 (en) Method for sorting a group of images of a database and method for color correcting an image, corresponding devices, computer program and non-transitory computer-readable medium
US9092862B2 (en) Image processing apparatus, image adjusting apparatus, image processing method and computer readable medium
CN117274085A (en) Low-illumination image enhancement method and device
Kınlı et al. Modeling the lighting in scenes as style for auto white-balance correction

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