WO2019128459A1 - Procédé et dispositif pour l'élimination d'ombre d'image - Google Patents

Procédé et dispositif pour l'élimination d'ombre d'image Download PDF

Info

Publication number
WO2019128459A1
WO2019128459A1 PCT/CN2018/113377 CN2018113377W WO2019128459A1 WO 2019128459 A1 WO2019128459 A1 WO 2019128459A1 CN 2018113377 W CN2018113377 W CN 2018113377W WO 2019128459 A1 WO2019128459 A1 WO 2019128459A1
Authority
WO
WIPO (PCT)
Prior art keywords
shadow
image
color
illumination
initial image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/113377
Other languages
English (en)
Chinese (zh)
Inventor
李革
余晓铭
应振强
王文敏
王荣刚
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.)
Peking University Shenzhen Graduate School
Original Assignee
Peking University Shenzhen Graduate School
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 Peking University Shenzhen Graduate School filed Critical Peking University Shenzhen Graduate School
Publication of WO2019128459A1 publication Critical patent/WO2019128459A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/90—Dynamic range modification of images or parts thereof
    • G06T5/94—Dynamic range modification of images or parts thereof based on local image properties, e.g. for local contrast enhancement
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence

Definitions

  • the present invention relates to the field of image processing, and in particular to an image shadow elimination method and apparatus.
  • Shadows are ubiquitous in video image data. Although shadows can provide auxiliary information for people to perceive scene illumination, depth of field, and object shape, the presence of shadows increases the difficulty of many image processing tasks, such as shadows in image segmentation tasks. Boundary and object boundaries are often difficult to distinguish; in object recognition, target tracking and other tasks, the existence of shadows often changes the material texture of the object, affecting the performance of related algorithms. In addition, for visual perception needs, in the post-production of some image videos, people often want to remove the relevant shadows in the scene that affect the visual experience. For this reason, shadow elimination is an important image processing task.
  • Shadow elimination generally includes two tasks: shadow recognition and illumination recovery.
  • two methods based on region and shadow boundary are generally used.
  • the region-based illumination recovery method uses region matching to find and shadow in the scene.
  • the non-shaded area with similar area material is used to supplement the illumination information, and the illumination restoration method based on the shadow boundary uses the illumination change information on both sides of the shadow boundary to perform illumination restoration on the shadow area. It can be seen that the illumination information mining of the scene is the key to realize image shadow elimination, but the existing shadow elimination methods only fill in the information mining in the local area, can not retain various material characteristics, and the processing efficiency and quality are difficult to satisfy. .
  • the embodiment of the invention provides an image shadow elimination method and device, so as to at least solve the technical problem that only the information mining of the local area information can be complemented in the illumination restoration task in the prior art, and the material characteristics cannot be retained.
  • an image shadow removing method comprising: determining different kinds of color lines in an initial image; performing shadow recognition on the initial image to obtain a shadow recognition result; and different types according to the shadow recognition result The shaded area of the color line is illuminated for illumination to obtain a shadow-removed image.
  • an image shadow removing apparatus including: a first determining module, configured to determine different kinds of color lines in an initial image; and a first identifying module, configured to perform an initial image
  • the shadow recognition obtains the shadow recognition result
  • the first illumination recovery module is configured to perform illumination restoration on the shaded areas of different kinds of color lines according to the shadow recognition result to obtain a shadow elimination image.
  • a storage medium comprising a stored program, wherein the device in which the storage medium is located is controlled to execute the image shadow removal method described above while the program is running.
  • a computer device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, the image shadow removing method being implemented when the processor executes the program .
  • the present invention by determining different kinds of color lines in the initial image; performing shadow recognition on the initial image to obtain a shadow recognition result; according to the shadow recognition result, performing light restoration on the shadow regions of different kinds of color lines to obtain a shadow Eliminate the image and restore the shadow through non-local color lines, thus providing high-quality shadow-removing images, targeted processing of various materials, and retaining the technical effects of various material characteristics, thereby solving the prior art
  • the light recovery task only the information mining of the local area information can be supplemented and the technical problems of various material characteristics cannot be retained.
  • FIG. 1 is a schematic diagram of an image shadow removal method according to an embodiment of the present invention.
  • FIG. 2 is a schematic diagram of an image shadow removal device in accordance with an embodiment of the present invention.
  • a method embodiment of an image shadow removal method is provided, it being noted that the steps illustrated in the flowchart of the figures may be performed in a computer system such as a set of computer executable instructions, and Although the logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than the ones described herein.
  • FIG. 1 is an image shadow removal method according to an embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:
  • Step S102 determining different kinds of color lines in the initial image
  • Step S104 performing shadow recognition on the initial image to obtain a shadow recognition result
  • Step S106 according to the shadow recognition result, performing illumination restoration on the shaded areas of different kinds of color lines to obtain a shadow elimination image.
  • the pixels with the same material in the initial image they have the same reflectivity and different illumination attenuation factors, so these pixels form an over-origin color line in the RGB color space, that is, different in the RGB color space.
  • Color lines represent pixels of different materials, while different points on the same color line indicate different lighting conditions.
  • the present invention by determining different kinds of color lines in the initial image; performing shadow recognition on the initial image to obtain a shadow recognition result; according to the shadow recognition result, performing light restoration on the shadow regions of different kinds of color lines to obtain a shadow Eliminate the image and restore the shadow through non-local color lines, thus providing high-quality shadow-removing images, targeted processing of various materials, and retaining the technical effects of various material characteristics, thereby solving the prior art
  • the light recovery task only the information mining of the local area information can be supplemented and the technical problems of various material characteristics cannot be retained.
  • determining different kinds of color lines in the initial image in step S102 including:
  • Step S202 performing offset correction on the color lines in the initial image to obtain a color line offset correction image
  • Step S204 clustering the color lines in the color line offset correction image to obtain different kinds of color lines.
  • performing offset correction on the color lines in the initial image in step S202 to obtain a color line offset correction image includes:
  • Step S302 calculating an offset of a color line represented by one of the shadow and non-shadow boundary regions in the RGB color space in the initial image;
  • Step S304 calculating a color line offset corresponding to each pixel in the initial image according to the offset amount
  • Step S306 performing offset correction on the color lines in the initial image according to the color line offset to obtain a color line offset corrected image.
  • the main direction of the pixel in the RGB color space may be The mean value of the pixels in the boundary region is obtained, wherein the main direction can be obtained by principal component analysis, assuming that the main direction is represented by v, the mean value is represented by p, and the boundary region is represented by m u , and the color line represented by the boundary region
  • the offset in the RGB color space is represented by D(m u ), which has the following formula 1:
  • ⁇ v ⁇ indicates that modulo is taken for v.
  • step S304 when the color line offset corresponding to each pixel in the initial image is calculated according to the offset amount, specifically, the offset amount of the color line represented by the boundary area in the RGB color space and the pixel of the boundary area are in RGB.
  • the main direction in the color space is calculated. If the initial image is represented by I(x), the color line offset corresponding to each pixel in the initial image I(x) is represented by D(x), and then there is Equation 2:
  • step S306 the color line in the initial image is offset-corrected according to the color line offset, and when the color line offset correction image is obtained, the initial image is subtracted from the color line offset corresponding to each pixel in the initial image.
  • Obtain a color line offset correction image if the color line is offset corrected image Said, there is the following formula 3:
  • Equation 3 Equation 3 above can also be decomposed into the following Equation 4:
  • Equation 4 S(x) represents the illumination attenuation factor map corresponding to the color line offset correction image, L represents the global fixed illumination constant of the scene where the line offset correction image is located, and R(x) represents the reflectance corresponding to each pixel.
  • S(x) takes a value equal to or slightly less than 1, in the penumbra region, S(x) is between 0 and 1, in the umbral area, S(x) is equal to or slightly larger than 0.
  • the color lines in the color line offset correction image are clustered in step S204 to obtain different kinds of color lines, specifically: correcting the RGB of each pixel in the image by using the color line offset.
  • the direction of the color vector clusters the color lines, and the different categories obtained by clustering represent different color lines.
  • the method further includes: step S402, determining a light attenuation factor map of the color line offset correction image; and performing shadow recognition on the initial image in step S104,
  • Obtaining the shadow recognition result includes: Step S502: performing shadow recognition on the initial image by using the illumination attenuation factor map to obtain a shadow recognition result.
  • the pixels with the largest value in each category represent the case where the illumination attenuation factor is 1, and the reflectance of each category is the same, so that each category can be utilized.
  • the ratio of the respective pixels to the pixels having the largest corresponding category value is obtained as a light attenuation factor map of the color line offset corrected image, wherein the light attenuation factor map can be represented by S(x).
  • the shadow recognition result includes identifying the umbral area and the penumbra area in the initial image, and performing light restoration on the shaded areas of the different kinds of color lines according to the shadow recognition result in step S106 includes: Step S602, performing illumination restoration on the umbral area and the penumbra area of different kinds of color lines according to the shadow recognition result.
  • the illumination attenuation factor graph S(x) distinguishes the umbral region, the penumbra region and the non-shadow region in the image, wherein the umbral region can be represented by U, the penumbra region can be represented by P, and the non-shaded region can be represented by N. .
  • step S602 illumination restoration is performed on the umbral region and the penumbra region of different kinds of color lines, wherein when the illumination is performed on the umbral region, the following formula 5 may be adopted:
  • E() represents the mean value
  • SD() represents the standard deviation
  • m represents the index of the color line.
  • a shadow removal image representing an umbral area corresponding to the color line m
  • N m , U m respectively represent a non-shaded area and a pixel set of the umbral area corresponding to the color line m
  • Non-shaded area Represents a color line offset correction image
  • the shading factor K(P m ) of the penumbra region can be linearly interpolated by using the illumination attenuation factor graph S(x) as follows:
  • Equation 7 the illumination recovery result of the penumbra region can be obtained as shown in Equation 7 below.
  • the light recovery results according to the umbra area
  • Light recovery results in penumbra And color line offset correction image
  • Non-shaded area Color line offset correction image Corresponding shadow removal image
  • the method further includes: step S702, performing local smoothing optimization processing on the illumination information after the illumination restoration in the shadow removal image to obtain a illumination optimization image.
  • the illumination recovery part mainly utilizes non-local information in the initial image, and the illumination change tends to have a characteristic of local smooth variation, and the local illumination optimization process can make the illumination change of the image smoother and more natural.
  • the step S702 performs local smoothing optimization processing on the illumination information after the illumination restoration in the shadow removal image, including: Step S802, using the minimized energy equation to restore the image in the shadow removal image after illumination recovery
  • the illumination information is locally smoothed and optimized.
  • Equation 8 the minimum energy equation is as shown in Equation 8.
  • Equation 9 the illumination optimized image I f (x) corresponding to the initial image I(x) can be expressed by the following Equation 9:
  • image shadow removal method of all the above embodiments can be applied to various machine vision fields as image preprocessing.
  • FIG. 2 is an image shadow removing device according to an embodiment of the present invention.
  • the device includes a first determining module and a first identification. a module and a first illumination recovery module, wherein the first determining module is configured to determine different kinds of color lines in the initial image; the first identifying module is configured to perform shadow recognition on the initial image to obtain a shadow recognition result; The module is configured to perform illumination recovery on the shaded areas of different kinds of color lines according to the shadow recognition result to obtain a shadow elimination image.
  • different types of color lines in the initial image are determined by the first determining module; the first identifying module performs shadow recognition on the initial image to obtain a shadow recognition result; and the first illumination recovery module is different according to the shadow recognition result.
  • the shaded area of the color line of the kind is used for light recovery to obtain a shadow-removed image, and the shadow is restored by non-local color lines, thereby realizing high-quality shadow-removing images, and various types of materials are treated in a targeted manner, and various types of materials are retained.
  • the technical effect of the material characteristics further solves the technical problem that the information recovery in the prior art can only complement the local area information and cannot retain various material characteristics.
  • the first determining module, the first identifying module, and the first lighting recovery module correspond to steps S102 to S106 in Embodiment 1, and the modules are the same as the examples and application scenarios implemented in the corresponding steps. However, it is not limited to the content disclosed in the above embodiment 1. It should be noted that the above modules may be implemented as part of a device in a computer system such as a set of computer executable instructions.
  • the first determining module includes a first offset correcting module and a clustering module, wherein the first offset correcting module is configured to perform offset correction on the color lines in the initial image to obtain The color line offset correction image; the clustering module is configured to cluster the color lines in the color line offset correction image to obtain different kinds of color lines.
  • the foregoing first offset correction module and the clustering module correspond to steps S202 to S204 in Embodiment 1, and the foregoing modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the above.
  • the above modules may be implemented as part of a device in a computer system such as a set of computer executable instructions.
  • the first offset correction module includes a first calculation module, a second calculation module, and a first offset correction module, wherein the first calculation module is configured to calculate one of the shadows in the initial image.
  • first computing module corresponds to the steps S302 to S306 in the first embodiment, and the examples and application scenarios implemented by the foregoing modules and corresponding steps. The same, but not limited to, the contents disclosed in the above embodiment 1.
  • the above modules may be implemented as part of a device in a computer system such as a set of computer executable instructions.
  • the device further includes a second determining module, configured to determine a light attenuation factor map of the color line offset correction image after the clustering module obtains different kinds of color lines; the first identifying module, The second identification module is configured to perform shadow recognition on the initial image by using a light attenuation factor map to obtain a shadow recognition result.
  • a second determining module configured to determine a light attenuation factor map of the color line offset correction image after the clustering module obtains different kinds of color lines
  • the second identification module is configured to perform shadow recognition on the initial image by using a light attenuation factor map to obtain a shadow recognition result.
  • the foregoing second determining module and the second identifying module correspond to step S402 to step S502 in the first embodiment, and the foregoing modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the foregoing implementation.
  • the shadow recognition result includes identifying an umbra area and a penumbra area in the initial image
  • the first illumination recovery module includes: a second illumination recovery module, configured to Illumination recovery is performed in the umbral area and the penumbra area of the color line of the type.
  • the foregoing second illumination recovery module corresponds to step S602 in Embodiment 1, and the foregoing modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the content disclosed in Embodiment 1 above. It should be noted that the above modules may be implemented as part of a device in a computer system such as a set of computer executable instructions.
  • the device further includes: a first optimization module, configured to perform local smoothing optimization on the illumination information after the illumination restoration in the shadow removal image after the first illumination recovery module obtains the shadow removal image. , get the light optimized image.
  • a first optimization module configured to perform local smoothing optimization on the illumination information after the illumination restoration in the shadow removal image after the first illumination recovery module obtains the shadow removal image. , get the light optimized image.
  • the foregoing first optimization module corresponds to step S702 in Embodiment 1, and the foregoing modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the content disclosed in Embodiment 1 above. It should be noted that the above modules may be implemented as part of a device in a computer system such as a set of computer executable instructions.
  • the first optimization module includes a second optimization module, configured to perform local smoothing optimization on the illumination information after the illumination restoration in the shadow removal image using the minimized energy equation.
  • the foregoing second optimization module corresponds to step S802 in Embodiment 1, and the foregoing modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the content disclosed in Embodiment 1 above. It should be noted that the above modules may be implemented as part of a device in a computer system such as a set of computer executable instructions.
  • a product embodiment of a storage medium comprising a stored program, wherein the device in which the storage medium is located is controlled to perform the image shadow removal method described above while the program is running.
  • a product embodiment of a processor for running a program is executed, wherein the image shadow removal method described above is executed while the program is running.
  • a product embodiment of a computer device comprising: a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the image is implemented when the processor executes the program Shadow elimination method.
  • a product embodiment of a terminal includes a first determining module, a first identifying module, a first lighting recovery module, and a processor, wherein the first determining module is configured to determine an initial image. Different types of color lines; a first identification module for performing shadow recognition on the initial image to obtain a shadow recognition result; and a first illumination recovery module for illuminating the shaded areas of different kinds of color lines according to the shadow recognition result Recovering, obtaining a shadow removal image; the processor, the processor running the program, wherein the program execution performs the image shadow removal method on the data output from the first determining module, the first recognition module, and the first illumination recovery module.
  • a product embodiment of a terminal includes a first determining module, a first identifying module, a first lighting recovery module, and a storage medium, wherein the first determining module is configured to determine an initial image. Different types of color lines; a first identification module for performing shadow recognition on the initial image to obtain a shadow recognition result; and a first illumination recovery module for illuminating the shaded areas of different kinds of color lines according to the shadow recognition result Recovering, obtaining a shadow removal image; a storage medium for storing the program, wherein the program performs the image shadow removal method on the data output from the first determination module, the first recognition module, and the first illumination recovery module at runtime.
  • the disclosed technical contents may be implemented in other manners.
  • the device embodiments described above are only schematic.
  • the division of the unit may be a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or may be Integrate into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, unit or module, and may be electrical or otherwise.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • the technical solution of the present invention which is essential or contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product stored in a storage medium.
  • a number of instructions are included to cause a computer device (which may be a personal computer, server or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes: a U disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk, and the like. .

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

L'invention concerne un procédé et un dispositif pour l'élimination d'ombre d'image. Le procédé consiste à : déterminer différents types de lignes de couleur dans une image initiale; effectuer une reconnaissance d'ombre par rapport à l'image initiale pour produire un résultat de reconnaissance d'ombre; la réalisation d'une récupération d'éclairage par rapport à des zones d'ombre des différents types de lignes de couleur pour produire une image sans ombre. La présente invention résout le problème technique de l'état de la technique selon lequel l'extraction et la complémentation d'informations ne peuvent être réalisées que pour des informations de zone locale et diverses caractéristiques de matériau ne peuvent pas être conservées dans une tâche de récupération d'éclairage.
PCT/CN2018/113377 2017-12-28 2018-11-01 Procédé et dispositif pour l'élimination d'ombre d'image Ceased WO2019128459A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711454482.8A CN108305217A (zh) 2017-12-28 2017-12-28 图像阴影消除方法和装置
CN201711454482.8 2017-12-28

Publications (1)

Publication Number Publication Date
WO2019128459A1 true WO2019128459A1 (fr) 2019-07-04

Family

ID=62867585

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/113377 Ceased WO2019128459A1 (fr) 2017-12-28 2018-11-01 Procédé et dispositif pour l'élimination d'ombre d'image

Country Status (2)

Country Link
CN (1) CN108305217A (fr)
WO (1) WO2019128459A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117952955A (zh) * 2024-03-06 2024-04-30 济宁鲁英机械制造有限公司 一种丝杠表面缺陷智能检测方法及装置
CN120070263A (zh) * 2025-01-17 2025-05-30 武汉大学 基于细节增强与边缘重建的遥感影像阴影去除方法与装置

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108305217A (zh) * 2017-12-28 2018-07-20 北京大学深圳研究生院 图像阴影消除方法和装置
CN110427950B (zh) * 2019-08-01 2021-08-27 重庆师范大学 紫色土土壤图像阴影检测方法
CN111667432B (zh) * 2020-06-09 2022-08-02 中国电子科技集团公司第五十四研究所 一种基于物理模型的遥感影像阴影去除方法
CN112862714B (zh) * 2021-02-03 2024-12-17 维沃移动通信有限公司 图像处理方法及装置
CN114897708B (zh) * 2022-03-02 2025-07-04 中国科学院信息工程研究所 一种基于深度学习和反射率的阴影去除方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104599253A (zh) * 2015-01-30 2015-05-06 武汉大学 一种自然图像阴影消除方法
KR20160037481A (ko) * 2014-09-29 2016-04-06 에스케이텔레콤 주식회사 영상 인식을 위한 그림자 제거 방법 및 이를 위한 그림자 제거 장치
CN108305217A (zh) * 2017-12-28 2018-07-20 北京大学深圳研究生院 图像阴影消除方法和装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20160037481A (ko) * 2014-09-29 2016-04-06 에스케이텔레콤 주식회사 영상 인식을 위한 그림자 제거 방법 및 이를 위한 그림자 제거 장치
CN104599253A (zh) * 2015-01-30 2015-05-06 武汉大学 一种自然图像阴影消除方法
CN108305217A (zh) * 2017-12-28 2018-07-20 北京大学深圳研究生院 图像阴影消除方法和装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
YU , XIAOMING: "A New Shadow Removal Method Using Color-Lines", CAIP 2017: INTERNATIONAL CONFERENCE ON COMPUTER ANALYSIS OF IMAGES AND PA- TTERNS, 24 August 2017 (2017-08-24), XP047422910, ISSN: 1611-3349 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117952955A (zh) * 2024-03-06 2024-04-30 济宁鲁英机械制造有限公司 一种丝杠表面缺陷智能检测方法及装置
CN120070263A (zh) * 2025-01-17 2025-05-30 武汉大学 基于细节增强与边缘重建的遥感影像阴影去除方法与装置

Also Published As

Publication number Publication date
CN108305217A (zh) 2018-07-20

Similar Documents

Publication Publication Date Title
WO2019128459A1 (fr) Procédé et dispositif pour l'élimination d'ombre d'image
US20260057536A1 (en) Systems and Methods for Hybrid Depth Regularization
CN109191395B (zh) 图像对比度增强方法、装置、设备及存储介质
EP3385903B1 (fr) Procédé et appareil pour améliorer une image de profondeur
Newson et al. Non-local patch-based image inpainting
CN101783012B (zh) 一种基于暗原色的自动图像去雾方法
US9396530B2 (en) Low memory content aware image modification
CN105303599B (zh) 用于在渲染图像时使用的可重照纹理
CN106133788B (zh) 处理数字图像的图像处理装置
US9196021B2 (en) Video enhancement using related content
CN108876743A (zh) 一种图像快速去雾方法、系统、终端及存储介质
CN108431751B (zh) 背景去除
Hou et al. Underwater image dehazing and denoising via curvature variation regularization
WO2012075227A1 (fr) Suppression de variation d'éclairement dans des images
CN114998320B (zh) 视觉显著性检测的方法、系统、电子装置和存储介质
CN113538569A (zh) 一种弱纹理物体位姿估计方法和系统
Tan et al. Stereo disparity through cost aggregation with guided filter
Xu et al. Depth-aided exemplar-based hole filling for DIBR view synthesis
CN107808366A (zh) 一种基于块匹配的自适应光转移单幅阴影去除方法
CN112561919B (zh) 一种图像分割方法、设备和计算机可读存储介质
Mukherjee et al. A hybrid algorithm for disparity calculation from sparse disparity estimates based on stereo vision
Kuzovkin et al. Descriptor-based image colorization and regularization
US9165347B2 (en) Method of and apparatus for local optimization texture synthesis 3-D inpainting
Li et al. DR-Net: Transmission steered single image dehazing network with weakly supervised refinement
CN115358963B (zh) 基于扩展高斯差分和引导滤波的图像融合方法

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: 18893606

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: 18893606

Country of ref document: EP

Kind code of ref document: A1