WO2017148035A1 - 图像处理方法及装置 - Google Patents

图像处理方法及装置 Download PDF

Info

Publication number
WO2017148035A1
WO2017148035A1 PCT/CN2016/084847 CN2016084847W WO2017148035A1 WO 2017148035 A1 WO2017148035 A1 WO 2017148035A1 CN 2016084847 W CN2016084847 W CN 2016084847W WO 2017148035 A1 WO2017148035 A1 WO 2017148035A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
sub
partition
processing
frame picture
Prior art date
Application number
PCT/CN2016/084847
Other languages
English (en)
French (fr)
Inventor
杨福军
张晓东
Original Assignee
深圳Tcl数字技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳Tcl数字技术有限公司 filed Critical 深圳Tcl数字技术有限公司
Publication of WO2017148035A1 publication Critical patent/WO2017148035A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • 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
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • 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
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image
    • 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
    • G06T2207/20021—Dividing image into blocks, subimages or windows

Definitions

  • the present invention relates to the field of image processing technologies, and in particular, to an image processing method and apparatus.
  • image enhancement is the purpose of emphasizing the overall or local characteristics of an image by analyzing the image information. For example, the original unclear image becomes clear or the contrast difference between different object features in the image is enhanced to improve the visual effect of the image.
  • some common algorithms for image enhancement include: contrast change, spatial filtering, image calculation, and the like.
  • the existing image enhancement method is often based on analyzing and counting the entire frame image content as a statistical area, and using the enhancement algorithm to perform the same image enhancement adjustment on the entire frame image content. In this way, it is difficult to take into account the characteristics of the local area image, and the image adjustment is not fine enough, so that the effect of the image enhancement algorithm is compromised.
  • the main object of the present invention is to provide an image processing method and apparatus for improving image processing effects.
  • an image processing method including:
  • Image feature analysis and statistics are performed on each sub-partition to obtain image feature information of each sub-partition;
  • an enhancement algorithm is applied to each sub-partition to perform image enhancement processing, and the enhanced images are combined to obtain a processed image frame image.
  • the step of performing image feature analysis and statistics on each sub-partition to obtain image feature information of each sub-partition further comprises:
  • an image enhancement algorithm is applied to each sub-partition and the sensitive image region for image enhancement processing, and the enhanced images are combined to obtain a processed image frame image.
  • the performing image enhancement processing comprises: performing image filtering, smoothing processing, and/or brightness and color adjustment.
  • the sub-partition image feature information includes: maximum brightness, minimum brightness, histogram distribution, and color characteristic key information in each sub-partition image picture.
  • the acquiring an image frame picture, and dividing the image frame picture into several sub-partitions comprises: acquiring an image frame picture, and dividing the image frame picture into several sub-partitions according to a resolution of the image frame picture.
  • An embodiment of the present invention further provides an image processing apparatus, including:
  • a dividing module configured to acquire an image frame picture, and divide the image frame picture into several sub-partitions
  • An analysis and statistics module is configured to separately perform image feature analysis and statistics on each sub-partition to obtain image feature information of each sub-partition;
  • the processing module is configured to apply an enhancement algorithm to each sub-partition to perform image enhancement processing according to the image information of each sub-partition, and combine the enhanced images to obtain a processed image frame image.
  • the device further comprises:
  • a sensitive area detecting module configured to perform sensitive area information detection on the image information of each sub-partition, and obtain an area of the sub-partition image that is a sensitive image
  • the processing module is further configured to apply an image enhancement algorithm to each sub-partition and the sensitive image region to perform image enhancement processing according to the feature information of the sub-partition image and the sensitive image region, and combine the enhanced images to be processed.
  • Image frame picture is further configured to apply an image enhancement algorithm to each sub-partition and the sensitive image region to perform image enhancement processing according to the feature information of the sub-partition image and the sensitive image region, and combine the enhanced images to be processed.
  • the processing module performs image enhancement processing, including: performing image filtering, smoothing processing, and/or brightness and color adjustment.
  • the partition image feature information comprises: maximum brightness, minimum brightness, histogram distribution and color characteristic key information in each divided image picture.
  • the image frame picture is acquired, and the image frame picture is divided into several sub-partitions according to the resolution of the image frame picture.
  • An image processing method and device divides an image frame picture into a plurality of sub-partitions by acquiring an image frame picture; performing image feature analysis and statistics on each sub-partition separately, and obtaining segment image feature information;
  • the sub-partition image feature information is applied to each sub-partition by an enhancement algorithm for image enhancement processing to obtain a processed image frame picture.
  • FIG. 1 is a schematic flow chart of a first embodiment of an image processing method according to the present invention.
  • FIG. 2 is a schematic diagram of a conventional image analysis statistical area
  • FIG. 3 is a schematic diagram of an image analysis statistical area according to an embodiment of the present invention.
  • FIG. 4 is a schematic flow chart of a second embodiment of an image processing method according to the present invention.
  • FIG. 5 is a schematic flowchart of an image enhancement algorithm according to an embodiment of the present invention.
  • FIG. 6 is a schematic diagram showing a refinement process of an image enhancement algorithm according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of functional modules of a first embodiment of an image processing apparatus according to the present invention.
  • Figure 8 is a block diagram showing the functional blocks of the second embodiment of the image processing apparatus of the present invention.
  • a first embodiment of the present invention provides an image processing method, including:
  • Step S101 acquiring an image frame picture, and dividing the image frame picture into several sub-partitions
  • Step S102 performing image feature analysis and statistics on each sub-partition separately, and obtaining image information of each sub-partition;
  • the image frame picture is divided into several sub-partitions, so that image feature analysis and statistics are performed on each sub-partition separately, and an enhancement algorithm is applied to each sub-partition for image enhancement processing, so as to take into account the local area image.
  • the image enhancement algorithm first performs statistical analysis on the original image features to obtain image feature information, and then performs enhancement processing on the original image for the obtained image feature information.
  • image features include, for example, maximum brightness, minimum brightness, average brightness, and histogram distribution and color characteristics.
  • Enhance the original image the usual methods are: contrast conversion, improve the brightness of some pixels, reduce the brightness of some pixels. Color conversion, adjustment of certain colors, etc., to improve the layering and color effects of the picture.
  • the enhancement processing of the original image can be equivalent to using the original image signal as an input, adjusting according to a certain enhancement curve, and outputting the adjusted image signal.
  • FIG. 2 is a conventional image analysis statistical area (taking 1920*1080 as an example), and the analysis and statistics of the image enhancement algorithm cover all pixels in the 1920*1080 area, that is, the entire 1920*1080 pixels. For a large area.
  • FIG. 3 is an image analysis statistical area of the present embodiment.
  • the solution in this embodiment divides the 1920*1080 pixels into a plurality of sub-regions, and the analysis statistical region of the image enhancement algorithm is independently performed in units of sub-regions, that is, the object of analysis and statistics is each sub-region.
  • the image feature analysis and statistics are performed on each sub-partition, and the obtained sub-partition image feature information may include: maximum brightness, minimum brightness, histogram distribution, and color characteristic key information in each sub-partition image picture.
  • the characteristics of the local area image can be taken into consideration, and the fine adjustment of the image can be realized, and the image processing effect can be improved.
  • Step S103 Apply an enhancement algorithm to each sub-partition to perform image enhancement processing according to the image information of each sub-partition, and combine the enhanced images to obtain a processed image frame image.
  • performing the image enhancement processing may include: performing image filtering, smoothing processing, and/or brightness, color adjustment, and the like.
  • the existing image enhancement algorithm flow is as follows:
  • a frame of 1920*1080 pixels is statistically analyzed to obtain key information such as maximum brightness, minimum brightness and histogram distribution and color characteristics in the picture, and then the image enhancement algorithm. Based on these key information, an enhanced adjustment curve is obtained, and real-time dynamic enhancement processing is performed on each frame image, such as filtering, smoothing, improving brightness of certain pixels, reducing brightness of certain pixels, enhancing contrast, enhancing certain colors, and the like. .
  • the statistical analysis of these enhancement processes is based on the entire frame, that is, 1920*1080 pixels as an entire unit, and the equivalent enhancement adjustment curve is one.
  • a frame of 1920*1080 pixels is first divided into several sub-partitions according to the image resolution. Then, based on each sub-partition, image analysis and statistics of each sub-partition are performed, and image information of each sub-partition is obtained.
  • the image analysis statistics is to analyze and count each sub-partition, instead of the whole picture.
  • the image analysis and statistics of each sub-partition are the image information of each sub-partition.
  • an enhancement adjustment curve is obtained based on the image information of each sub-partition, and real-time dynamic enhancement processing is performed on each frame image, such as filtering and smoothing, improving brightness of certain pixels, reducing brightness of certain pixels, enhancing contrast, and focusing on certain colors. Enhance and more.
  • the characteristics of the local area image can be taken into consideration, and the fine adjustment of the image can be realized, and the image processing effect can be improved.
  • the second embodiment of the present invention provides an image processing method. Based on the embodiment shown in FIG. 1, the method performs image feature analysis and statistics on each sub-partition to obtain a partition image.
  • the feature information also includes:
  • Step S104 performing sensitive area information detection on the image information of each sub-partition, and obtaining an area of the sub-partition image that is a sensitive image;
  • an image enhancement process is performed by applying an enhancement algorithm to each sub-partition according to the image information of each sub-partition, and the obtained image frame image includes:
  • Step S1031 Apply an image enhancement algorithm to each sub-partition and the sensitive image region to perform image enhancement processing according to the sub-partition image and the feature information of the sensitive image region, and combine the enhanced images to obtain the processed image frame image.
  • the embodiment further includes a solution for detecting and performing corresponding equalization processing on the sensitive area.
  • the present embodiment adopts the sensitive area equalization enhancement processing.
  • the image enhancement algorithm of this embodiment is as follows:
  • a frame of 1920*1080 pixels is first divided into several sub-partitions according to the image resolution. Then, based on each sub-partition, image analysis and statistics of each sub-partition are performed, and image information of each sub-partition is obtained. Image analysis statistics analyze and count each sub-partition, not the entire picture.
  • the image feature information obtained by analyzing and counting each sub-partition image is transmitted to the sensitive area detecting module, and the sensitive area detecting module analyzes the above information to obtain characteristic information of the sensitive image area, for example, for image areas such as skin color and large area blue sky and white clouds. Because the details are too rich and slow, the visual response caused by the change is sensitive, and it is sensitive.
  • the identification of sensitive areas is based on existing technologies, such as skin color recognition, and will not be described in detail herein.
  • the sub-partition image feature information and the feature information of the sensitive image region are further transmitted to the partition equalization image enhancement algorithm module in the processing module.
  • the partition equalization image enhancement algorithm module applies an enhancement algorithm to adjust the image for each sub-partition and the sensitive image region according to the sub-partition image feature information and the feature information of the sensitive image region, and obtains an equivalent enhancement adjustment curve, and uses an enhanced adjustment curve to complete Image enhancement.
  • the purpose of the sensitive area detecting module is to find a sensitive image area and transmit the feature information of the sensitive image area to the partition equalized image enhancement algorithm module.
  • the partitioned equalization image enhancement algorithm module performs special equalization processing on the contents of these sensitive image regions to prevent the mutation and blockiness of the interval.
  • the method of special equalization processing may be a method of merging blocks.
  • image areas such as skin color and large-area blue sky and white clouds
  • these areas should be subjected to special equalization processing, such as merging processing. That is, these regions are analyzed as a whole, and the image features are analyzed to obtain the same enhanced adjustment curve, and the image adjustment curve is used for image enhancement. Otherwise, if the sensitive area is divided into different sub-areas, applying different adjustment curves to each sub-area may cause a block effect, causing excessive unnaturalness and mutation of the above-mentioned sensitive areas.
  • the image enhancement algorithm in this embodiment has an enhanced adjustment curve of a plurality of different curves instead of one.
  • a plurality of different enhancement adjustment curves are used to adjust the respective regions, and finally the regions are merged to obtain an adjusted entire frame.
  • the image frame picture is divided into several sub-partitions; image feature analysis and statistics are performed on each sub-partition to obtain image information of each sub-partition; and image information of each sub-partition image is obtained.
  • Image enhancement processing is applied to each sub-partition for image enhancement processing, and the processed image frame image is obtained.
  • FIG. 6 is a schematic flowchart of a refinement of an image enhancement algorithm according to an embodiment of the present invention.
  • Image input input image signal.
  • Partition by resolution divide the entire picture into several sub-partitions according to the resolution
  • sub-partition image analysis image feature analysis and statistics of each sub-area, to obtain image information of each sub-partition, such as brightness distribution information, color distribution information, etc.;
  • Partition enhancement adjustment According to the obtained image feature information, the enhancement algorithm is used to obtain the enhancement curve of each sub-partition, and the image of each sub-area is enhanced and adjusted, such as filtering and smoothing, improving the brightness of some pixels and reducing Some pixel brightness, enhanced contrast, enhanced for certain colors, and more.
  • the statistical analysis based on these enhancements is based on the entire frame, and the equivalent enhancement adjustment curve is one. After that, the adjusted sub-area image except the sensitive area is output;
  • each sub-partition image feature information is sent to the sensitive area for detection, detecting sensitive image areas;
  • Enhanced adjustment of sensitive areas Combine the same sensitive areas to obtain the same equilibrium adjustment curve, and use an equalization adjustment curve to enhance and adjust the sensitive area image.
  • the enhanced sub-area image is output.
  • Output enhanced image The enhanced sub-partition and the enhanced processed sensitive area image are combined to form a complete enhanced image.
  • the scheme of the embodiment can realize the fine adjustment of the image and improve the image processing effect by using the sub-partition image analysis and the enhancement processing, and the image processing effect can be improved by using the sensitive area equalization enhancement processing, and in some cases,
  • the degree of fine adjustment of the image enhancement algorithm is enhanced by the abrupt effect between the blocks due to the partition processing.
  • the first embodiment of the present invention provides an image processing apparatus, including: a partitioning module 201, an analysis and statistics module 202, and a processing module 203, wherein:
  • a dividing module 201 configured to acquire an image frame picture, and divide the image frame picture into several sub-partitions
  • the analysis and statistics module 202 is configured to separately perform image feature analysis and statistics on each sub-partition to obtain image information of each sub-partition;
  • the processing module 203 is configured to apply an enhancement algorithm to each sub-partition to perform image enhancement processing according to the image information of each sub-partition, and obtain a processed image frame image.
  • the image frame picture is divided into several sub-partitions, so that image feature analysis and statistics are performed separately for each sub-partition, and an enhancement algorithm is applied to each sub-partition for image enhancement processing, so as to take local
  • an enhancement algorithm is applied to each sub-partition for image enhancement processing, so as to take local
  • the image enhancement algorithm first performs statistical analysis on the original image features to obtain image feature information, and then performs enhancement processing on the original image for the obtained image feature information.
  • image features include, for example, maximum brightness, minimum brightness, average brightness, and histogram distribution and color characteristics.
  • Enhance the original image the usual methods are: contrast conversion, improve the brightness of some pixels, reduce the brightness of some pixels. Color conversion, adjustment of certain colors, etc., to improve the layering and color effects of the picture.
  • the enhancement processing of the original image can be equivalent to using the original image signal as an input, adjusting according to a certain enhancement curve, and outputting the adjusted image signal.
  • FIG. 2 is a conventional image analysis statistical area (taking 1920*1080 as an example), and the analysis and statistics of the image enhancement algorithm cover all pixels in the 1920*1080 area, that is, the entire 1920*1080 pixels. For a large area.
  • FIG. 3 is an image analysis statistical area of the present embodiment.
  • the solution in this embodiment divides the 1920*1080 pixels into a plurality of sub-regions, and the analysis statistical region of the image enhancement algorithm is independently performed in units of sub-regions, that is, the object of analysis and statistics is each sub-region.
  • the image feature analysis and statistics are performed on each sub-partition, and the obtained partition image feature information may include: maximum brightness, minimum brightness, histogram distribution, and color characteristic key information in each sub-partition image picture.
  • an enhancement algorithm is applied to each sub-partition to perform image enhancement processing, and the processed image frame picture is obtained.
  • performing the image enhancement processing may include: performing image filtering, smoothing processing, and/or brightness, color adjustment, and the like.
  • the existing image enhancement algorithm flow is as follows:
  • a frame of 1920*1080 pixels is statistically analyzed to obtain key information such as maximum brightness, minimum brightness and histogram distribution and color characteristics in the picture, and then the image enhancement algorithm. Based on these key information, an enhanced adjustment curve is obtained, and real-time dynamic enhancement processing is performed on each frame image, such as filtering, smoothing, improving brightness of certain pixels, reducing brightness of certain pixels, enhancing contrast, enhancing certain colors, and the like. .
  • the statistical analysis of these enhancement processes is based on the entire frame, that is, 1920*1080 pixels as an entire unit, and the equivalent enhancement adjustment curve is one.
  • a frame of 1920*1080 pixels is first divided into several sub-partitions according to the image resolution. Then, based on each sub-partition, image analysis and statistics of each sub-partition are performed, and image information of each sub-partition is obtained.
  • the image analysis statistics is to analyze and count each sub-partition, instead of the whole picture.
  • the image analysis and statistics of each sub-partition are the image information of each sub-partition.
  • an enhancement adjustment curve is obtained based on the image information of each sub-partition, and real-time dynamic enhancement processing is performed on each frame image, such as filtering and smoothing, improving brightness of certain pixels, reducing brightness of certain pixels, enhancing contrast, and focusing on certain colors. Enhance and more.
  • the characteristics of the local area image can be taken into consideration, and the fine adjustment of the image can be realized, and the image processing effect can be improved.
  • the second embodiment of the present invention provides an image processing apparatus, and the apparatus further includes:
  • the sensitive area detecting module 204 is configured to perform sensitive area information detection on the image information of each sub-partition, and obtain an area of the sub-partition image that is a sensitive image;
  • the processing module 203 is further configured to apply an image enhancement algorithm to each sub-partition and the sensitive image region to perform image enhancement processing according to the feature information of the sub-partition image and the sensitive image region, and combine the enhanced images to be processed. After the image frame picture.
  • the embodiment further includes a solution for detecting and performing corresponding equalization processing on the sensitive area.
  • the present embodiment adopts the sensitive area equalization enhancement processing.
  • the image enhancement algorithm of this embodiment is as follows:
  • a frame of 1920*1080 pixels is first divided into several sub-partitions according to the image resolution. Then, based on each sub-partition, image analysis and statistics of each sub-partition are performed, and image information of each sub-partition is obtained. Image analysis statistics analyze and count each sub-partition, not the entire picture.
  • the image feature information obtained by analyzing and counting each sub-partition image is transmitted to the sensitive area detecting module, and the sensitive area detecting module analyzes the above information to obtain characteristic information of the sensitive image area, for example, for image areas such as skin color and large area blue sky and white clouds. Because the details are too rich and slow, the visual response caused by the change is sensitive, and it is sensitive.
  • the identification of sensitive areas is based on existing technologies, such as skin color recognition, and will not be described in detail herein.
  • the sub-partition image feature information and the feature information of the sensitive image region are further transmitted to the partition equalization image enhancement algorithm module in the processing module.
  • the partition equalization image enhancement algorithm module applies an enhancement algorithm to adjust the image for each sub-partition and the sensitive image region according to the sub-partition image feature information and the feature information of the sensitive image region, and obtains an equivalent enhancement adjustment curve, and uses an enhanced adjustment curve to complete Image enhancement.
  • the purpose of the sensitive area detecting module is to find a sensitive image area and transmit the feature information of the sensitive image area to the partition equalized image enhancement algorithm module.
  • the partitioned equalization image enhancement algorithm module performs special equalization processing on the contents of these sensitive image regions to prevent the mutation and blockiness of the interval.
  • the method of special equalization processing may be a method of merging blocks.
  • image areas such as skin color and large-area blue sky and white clouds
  • these areas should be subjected to special equalization processing, such as merging processing. That is, these regions are analyzed as a whole, and the image features are analyzed to obtain the same enhanced adjustment curve, and the image adjustment curve is used for image enhancement. Otherwise, if the sensitive area is divided into different sub-areas, applying different adjustment curves to each sub-area may cause a block effect, causing excessive unnaturalness and mutation of the above-mentioned sensitive areas.
  • the image enhancement algorithm in this embodiment has an enhanced adjustment curve of a plurality of different curves instead of one.
  • a plurality of different enhancement adjustment curves are used to adjust the respective regions, and finally the regions are merged to obtain an adjusted entire frame.
  • the image frame picture is divided into several sub-partitions; image feature analysis and statistics are performed on each sub-partition to obtain image information of each sub-partition; and image information of each sub-partition image is obtained.
  • Image enhancement processing is applied to each sub-partition for image enhancement processing, and the processed image frame image is obtained.
  • FIG. 6 is a schematic flowchart of a refinement of an image enhancement algorithm according to an embodiment of the present invention.
  • Image input input image signal.
  • Partition by resolution divide the entire picture into several sub-partitions according to the resolution
  • sub-partition image analysis image feature analysis and statistics of each sub-area, to obtain image information of each sub-partition, such as brightness distribution information, color distribution information, etc.;
  • Partition enhancement adjustment According to the obtained image feature information, the enhancement algorithm is used to obtain the enhancement curve of each sub-partition, and the image of each sub-area is enhanced and adjusted, such as filtering and smoothing, improving the brightness of some pixels and reducing Some pixel brightness, enhanced contrast, enhanced for certain colors, and more.
  • the statistical analysis based on these enhancements is based on the entire frame, and the equivalent enhancement adjustment curve is one. After that, the adjusted sub-area image except the sensitive area is output;
  • each sub-partition image feature information is sent to the sensitive area for detection, detecting sensitive image areas;
  • Enhanced adjustment of sensitive areas Combine the same sensitive areas to obtain the same equilibrium adjustment curve, and use an equalization adjustment curve to enhance and adjust the sensitive area image.
  • the enhanced sub-area image is output.
  • Output enhanced image The enhanced sub-partition and the enhanced processed sensitive area image are combined to form a complete enhanced image.
  • the scheme of the embodiment can realize the fine adjustment of the image and improve the image processing effect by using the sub-partition image analysis and the enhancement processing, and the image processing effect can be improved by using the sensitive area equalization enhancement processing, and in some cases,
  • the degree of fine adjustment of the image enhancement algorithm is enhanced by the abrupt effect between the blocks due to the partition processing.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Picture Signal Circuits (AREA)

Abstract

一种图像处理方法及装置，其方法包括：获取图像帧画面，将所述图像帧画面分成若干子分区；对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；依据各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果；此外，通过敏感区均衡增强处理，可以避免某些情况下，由于分区处理而产生的区块间的突变效应，提升图像增强算法的精细调整程度。

Description

图像处理方法及装置
技术领域
本发明涉及图像处理技术领域，尤其涉及一种图像处理方法及装置。
背景技术
在图像处理技术中，图像增强是通过对图像信息的分析，从而有目的地强调图像的整体或局部特性。如，将原来不清晰的图像变得清晰或增强图像中不同物体特征之间的对比差别等，以改善图像的视觉效果。目前，图像增强的一些常用算法包括：对比度变、空间滤波、图像运算等。
但是，现有的图像增强方法，往往是基于将整帧图像内容作为一个统计区，进行分析统计，并运用增强算法，对整帧图像内容进行同一图像的增强调整。这种方式，较难兼顾到局部区域图像的特点，对图像调整不够精细，由此使得图像增强算法的效果会打折扣。
发明内容
本发明的主要目的在于提供一种图像处理方法及装置，旨在提升图像处理效果。
为了达到上述目的，本发明提出一种图像处理方法，包括：
获取图像帧画面，将所述图像帧画面分成若干子分区；
对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
优选地，所述对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息的步骤之后还包括：
对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
所述依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，得到处理后的图像帧画面的步骤包括：
依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
优选地，所述进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
优选地，所述各子分区图像特征信息包括：各子分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
优选地，所述获取图像帧画面，将所述图像帧画面分成若干子分区包括：获取图像帧画面，依据所述图像帧画面的分辨率，将所述图像帧画面分成若干子分区。
本发明实施例还提出一种图像处理装置，包括：
划分模块，用于获取图像帧画面，将所述图像帧画面分成若干子分区；
分析统计模块，用于对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
处理模块，用于依据所述各子分区图像特征信息，对各子分区应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
优选地，所述装置还包括：
敏感区检测模块，用于对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
所述处理模块，还用于依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
优选地，所述处理模块进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
优选地，所述分区图像特征信息包括：各分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
优选地，获取图像帧画面，依据所述图像帧画面的分辨率，将所述图像帧画面分成若干子分区。
本发明实施例提出的一种图像处理方法及装置，通过获取图像帧画面，将所述图像帧画面分成若干子分区；对各子分区分别进行图像特征分析统计，得到分区图像特征信息；依据各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，得到处理后的图像帧画面，由此，通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果；此外，通过敏感区均衡增强处理，可以避免某些情况下，由于分区处理而产生的区块间的突变效应，提升图像增强算法的精细调整程度。
附图说明
图1是本发明图像处理方法第一实施例的流程示意图；
图2是现有的图像分析统计区域示意图；
图3是本发明实施例图像分析统计区域示意图；
图4是本发明图像处理方法第二实施例的流程示意图；
图5是本发明实施例中图像增强算法流程示意图；
图6是本发明实施例中图像增强算法的细化流程示意图；
图7是本发明图像处理装置第一实施例的功能模块示意图；
图8是本发明图像处理装置第二实施例的功能模块示意图。
为了使本发明的技术方案更加清楚、明了，下面将结合附图作进一步详述。
具体实施方式
如图1所示，本发明第一实施例提出一种图像处理方法，包括：
步骤S101，获取图像帧画面，将所述图像帧画面分成若干子分区；
步骤S102，对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
为了增强图像处理效果，本实施例将图像帧画面分成若干子分区，以便对各子分区分别进行图像特征分析统计，对各子分区分别应用增强算法进行图像增强处理，以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果。
通常，图像增强算法是先对原始图像特征进行统计分析，得到图像特征信息，之后针对所得到的图像特征信息，对原始图像进行增强处理。其中，图像特征包括诸如：画面最大亮度、最小亮度、平均亮度，以及柱状图分布及色彩特色等等。对原始图像进行增强处理，通常的方法有：对比度变换，提高某些像素亮度，降低某些像素亮度。色彩变换，对某些色彩进行调整等等，以提高画面层次感及色彩效果。对原始图像进行增强处理，可以等效为以原始图像信号为输入，按照一定的增强曲线调整，输出调整后的图像信号。
具体地，如图2所示，图2为现有的图像分析统计区域（以1920*1080为例），图像增强算法的分析统计涵盖1920*1080区域中的所有像素，即整个1920*1080像素为一个大区域。
如图3所示，图3为本实施例图像分析统计区域。由图3可知，本实施例方案是将1920*1080像素分成若干个子区域，而图像增强算法的分析统计区域是以各个子区域为单位分别独立进行，即分析统计的对象是每个子区域。
其中，对各子分区分别进行图像特征分析统计，得到的子分区图像特征信息可以包括：各子分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
由此，通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果。
步骤S103，依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
其中，进行图像增强处理可以包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整等。
具体地，现有的图像增强算法流程如下：
以1920*1080为例，一帧1920*1080像素的内容画面，经统计分析后，得出诸如画面中的最大亮度、最小亮度及柱状图分布及色彩特色等关键信息，而后的图像增强算法，依据这些关键信息，得到增强调整曲线，并对各帧图像进行实时动态增强处理，如，滤波、平滑，提高某些像素亮度，降低某些像素亮度，增强对比度，对某些色彩进行增强等等。这些增强处理的统计分析基础是整帧画面，即以1920*1080像素为一个整个单位，其等效增强调整曲线为一条。
而本实施例的图像增强算法流程如下：
将一帧1920*1080像素的画面，首先按照图象分辨率，分成若干个子分区，之后，基于各个子分区，进行各子分区图像分析统计，得到各子分区图像特征信息。图像分析统计是对各个子分区进行分析统计，而不是整幅画面，各子分区图像分析统计得到的是各子分区图像特征信息。
之后，基于各子分区图像特征信息得到增强调整曲线，并对各帧图像进行实时动态增强处理，如，滤波、平滑，提高某些像素亮度，降低某些像素亮度，增强对比度，对某些色彩进行增强等等。
最后，将增强后的各个图像合并得到处理后的图像帧画面。
由此，通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果。
如图4所示，本发明第二实施例提出一种图像处理方法，基于上述图1所示的实施例，该方法在上述步骤S102：对各子分区分别进行图像特征分析统计，得到分区图像特征信息之后还包括：
步骤S104，对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
上述步骤S103：依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，得到处理后的图像帧画面包括：
步骤S1031，依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
相比上述实施例，本实施例还包括敏感区域检测并进行相应均衡处理的方案。
具体地，为了避免某些情况下，由于分区处理而产生的区块间的突变效应，以提升图像增强算法的精细调整程度，本实施例采用敏感区均衡增强处理。
如图5所示，本实施例的图像增强算法流程如下：
将一帧1920*1080像素的画面，首先按照图象分辨率，分成若干个子分区，之后，基于各个子分区，进行各子分区图像分析统计，得到各子分区图像特征信息。图像分析统计是对各个子分区进行分析统计，而不是整幅画面。各子分区图像分析统计得到的图像特征信息，传递给敏感区检测模块，敏感区检测模块对上述信息进行分析，得到敏感图像区域的特征信息，例如，对于如肤色及大面积蓝天白云等图像区域，因其细节丰富过度缓慢，其变化引起的视觉反应较为敏感，是为敏感区。敏感区的识别是基于已有的技术，如肤色识别等，在此不作详述。
之后将这些子分区图像特征信息及敏感图像区域的特征信息，进一步传递给处理模块中的分区均衡图像增强算法模块。分区均衡图像增强算法模块，依据上述子分区图像特征信息及敏感图像区域的特征信息，对各子分区及敏感图像区域，应用增强算法调整图像，取得等效增强调整曲线，并运用增强调整曲线完成图像增强。
其中敏感区域检测模块的目的是，找出敏感图像区域，将敏感图像区域的特征信息传递给分区均衡图像增强算法模块。分区均衡图像增强算法模块，将这些敏感图像区域内容进行特殊均衡处理，以防止产生区间的突变及块效应。其中，特殊均衡处理的方法，可以是合并区块等方式。
具体地，对于如肤色及大面积蓝天白云等图像区域，应该将这些区域进行特殊均衡处理，如，合并处理。即，将这些区域作为一个整体，分析其图像特征，取得同一条增强调整曲线，并运用图像调整曲线进行图像增强。否则，如果将敏感区分割成不同子区，对各子区运用不同的调整曲线，可能会引起块效应，造成上述敏感区过度不自然，突变等问题。
也就是说，本实施例中的图像增强算法其增强调整曲线为多条不同的曲线，而不是一条。多条不同的增强调整曲线，对各自的区域进行调整，最后通过区域合并，得到经调整后的整帧画面。
本实施例通过上述方案，通过获取图像帧画面，将所述图像帧画面分成若干子分区；对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；依据各子分区图像特征信息，对各子分区分别应用图像增强算法进行图像增强处理，得到处理后的图像帧画面，由此，通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果；此外，通过敏感区均衡增强处理，可以避免某些情况下，由于分区处理而产生的区块间的突变效应，提升图像增强算法的精细调整程度。
参照图6所示，图6为本实施例图像增强算法的一种细化流程示意图。
如图6所示，具体流程如下：
1、图像输入：输入图像信号。
2、按分辨率分区：将整幅画面按分辨率分成若干子分区；
3、子分区图像分析：对各子区域进行图像特征分析统计，得到各子分区图像特征信息，如亮度分布信息，色彩分布信息等；
4、分区增强调整：依据得到的上述图像特征信息，运用增强算法，取得各子分区增强调整曲线，并对各子分区的图像进行增强调整，如，滤波、平滑，提高某些像素亮度，降低某些像素亮度，增强对比度，对某些色彩进行增强等等。这些增强处理的统计分析基础是整帧画面，其等效增强调整曲线为一条。之后，输出除敏感区之外的调整后的子区图像；
5、敏感图检测：各子分区图像特征信息同时送往敏感区检测，检测敏感图像区域；
6、敏感区增强调整：对同一敏感区合并处理，取得同一条均衡调整曲线，并运用一条均衡调整曲线对敏感区域图像进行增强调整。输出增强调整后的子区图像。
7、输出增强后的图像：经增强处理的各子分区及经增强处理后的敏感区域图像，合成完整的整幅增强后的图像。
本实施例方案通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果；此外，通过敏感区均衡增强处理，可以避免某些情况下，由于分区处理而产生的区块间的突变效应，提升图像增强算法的精细调整程度。
对应地，提出本发明图像处理装置实施例。
如图7所示，本发明第一实施例提出一种图像处理装置，包括：划分模块201、分析统计模块202及处理模块203，其中：
划分模块201，用于获取图像帧画面，将所述图像帧画面分成若干子分区；
分析统计模块202，用于对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
处理模块203，用于依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，得到处理后的图像帧画面。
具体地，为了增强图像处理效果，本实施例将图像帧画面分成若干子分区，以便对各子分区分别进行图像特征分析统计，对各子分区分别应用增强算法进行图像增强处理，以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果。
通常，图像增强算法是先对原始图像特征进行统计分析，得到图像特征信息，之后针对所得到的图像特征信息，对原始图像进行增强处理。其中，图像特征包括诸如：画面最大亮度、最小亮度、平均亮度，以及柱状图分布及色彩特色等等。对原始图像进行增强处理，通常的方法有：对比度变换，提高某些像素亮度，降低某些像素亮度。色彩变换，对某些色彩进行调整等等，以提高画面层次感及色彩效果。对原始图像进行增强处理，可以等效为以原始图像信号为输入，按照一定的增强曲线调整，输出调整后的图像信号。
具体地，如图2所示，图2为现有的图像分析统计区域（以1920*1080为例），图像增强算法的分析统计涵盖1920*1080区域中的所有像素，即整个1920*1080像素为一个大区域。
如图3所示，图3为本实施例图像分析统计区域。由图3可知，本实施例方案是将1920*1080像素分成若干个子区域，而图像增强算法的分析统计区域是以各个子区域为单位分别独立进行，即分析统计的对象是每个子区域。
其中，对各子分区分别进行图像特征分析统计，得到的分区图像特征信息可以包括：各子分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
之后，依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，得到处理后的图像帧画面。
其中，进行图像增强处理可以包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整等。
具体地，现有的图像增强算法流程如下：
以1920*1080为例，一帧1920*1080像素的内容画面，经统计分析后，得出诸如画面中的最大亮度、最小亮度及柱状图分布及色彩特色等关键信息，而后的图像增强算法，依据这些关键信息，得到增强调整曲线，并对各帧图像进行实时动态增强处理，如，滤波、平滑，提高某些像素亮度，降低某些像素亮度，增强对比度，对某些色彩进行增强等等。这些增强处理的统计分析基础是整帧画面，即以1920*1080像素为一个整个单位，其等效增强调整曲线为一条。
而本实施例的图像增强算法流程如下：
将一帧1920*1080像素的画面，首先按照图象分辨率，分成若干个子分区，之后，基于各个子分区，进行各子分区图像分析统计，得到各子分区图像特征信息。图像分析统计是对各个子分区进行分析统计，而不是整幅画面，各子分区图像分析统计得到的是各子分区图像特征信息。
之后，基于各子分区图像特征信息得到增强调整曲线，并对各帧图像进行实时动态增强处理，如，滤波、平滑，提高某些像素亮度，降低某些像素亮度，增强对比度，对某些色彩进行增强等等。
最后，将增强后的各个图像合并得到处理后的图像帧画面。
由此，通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果。
如图8所示，本发明第二实施例提出一种图像处理装置，所述装置还包括：
敏感区检测模块204，用于对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
所述处理模块203，还用于依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
相比上述实施例，本实施例还包括敏感区域检测并进行相应均衡处理的方案。
具体地，为了避免某些情况下，由于分区处理而产生的区块间的突变效应，以提升图像增强算法的精细调整程度，本实施例采用敏感区均衡增强处理。
如图5所示，本实施例的图像增强算法流程如下：
将一帧1920*1080像素的画面，首先按照图象分辨率，分成若干个子分区，之后，基于各个子分区，进行各子分区图像分析统计，得到各子分区图像特征信息。图像分析统计是对各个子分区进行分析统计，而不是整幅画面。各子分区图像分析统计得到的图像特征信息，传递给敏感区检测模块，敏感区检测模块对上述信息进行分析，得到敏感图像区域的特征信息，例如，对于如肤色及大面积蓝天白云等图像区域，因其细节丰富过度缓慢，其变化引起的视觉反应较为敏感，是为敏感区。敏感区的识别是基于已有的技术，如肤色识别等，在此不作详述。
之后将这些子分区图像特征信息及敏感图像区域的特征信息，进一步传递给处理模块中的分区均衡图像增强算法模块。分区均衡图像增强算法模块，依据上述子分区图像特征信息及敏感图像区域的特征信息，对各子分区及敏感图像区域，应用增强算法调整图像，取得等效增强调整曲线，并运用增强调整曲线完成图像增强。
其中敏感区域检测模块的目的是，找出敏感图像区域，将敏感图像区域的特征信息传递给分区均衡图像增强算法模块。分区均衡图像增强算法模块，将这些敏感图像区域内容进行特殊均衡处理，以防止产生区间的突变及块效应。其中，特殊均衡处理的方法，可以是合并区块等方式。
具体地，对于如肤色及大面积蓝天白云等图像区域，应该将这些区域进行特殊均衡处理，如，合并处理。即，将这些区域作为一个整体，分析其图像特征，取得同一条增强调整曲线，并运用图像调整曲线进行图像增强。否则，如果将敏感区分割成不同子区，对各子区运用不同的调整曲线，可能会引起块效应，造成上述敏感区过度不自然，突变等问题。
也就是说，本实施例中的图像增强算法其增强调整曲线为多条不同的曲线，而不是一条。多条不同的增强调整曲线，对各自的区域进行调整，最后通过区域合并，得到经调整后的整帧画面。
本实施例通过上述方案，通过获取图像帧画面，将所述图像帧画面分成若干子分区；对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；依据各子分区图像特征信息，对各子分区分别应用图像增强算法进行图像增强处理，得到处理后的图像帧画面，由此，通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果；此外，通过敏感区均衡增强处理，可以避免某些情况下，由于分区处理而产生的区块间的突变效应，提升图像增强算法的精细调整程度。
参照图6所示，图6为本实施例图像增强算法的一种细化流程示意图。
如图6所示，具体流程如下：
1、图像输入：输入图像信号。
2、按分辨率分区：将整幅画面按分辨率分成若干子分区；
3、子分区图像分析：对各子区域进行图像特征分析统计，得到各子分区图像特征信息，如亮度分布信息，色彩分布信息等；
4、分区增强调整：依据得到的上述图像特征信息，运用增强算法，取得各子分区增强调整曲线，并对各子分区的图像进行增强调整，如，滤波、平滑，提高某些像素亮度，降低某些像素亮度，增强对比度，对某些色彩进行增强等等。这些增强处理的统计分析基础是整帧画面，其等效增强调整曲线为一条。之后，输出除敏感区之外的调整后的子区图像；
5、敏感图检测：各子分区图像特征信息同时送往敏感区检测，检测敏感图像区域；
6、敏感区增强调整：对同一敏感区合并处理，取得同一条均衡调整曲线，并运用一条均衡调整曲线对敏感区域图像进行增强调整。输出增强调整后的子区图像。
7、输出增强后的图像：经增强处理的各子分区及经增强处理后的敏感区域图像，合成完整的整幅增强后的图像。
本实施例方案通过子分区图像分析及增强处理，可以兼顾到局部区域图像的特点，实现对图像的精细调整，提高图像处理效果；此外，通过敏感区均衡增强处理，可以避免某些情况下，由于分区处理而产生的区块间的突变效应，提升图像增强算法的精细调整程度。
以上所述仅为本发明的优选实施例，并非因此限制本发明的专利范围，凡是利用本发明说明书及附图内容所作的等效结构或流程变换，或直接或间接运用在其它相关的技术领域，均同理包括在本发明的专利保护范围内。

Claims (15)

  1. 一种图像处理方法，其特征在于，包括：
    获取图像帧画面，将所述图像帧画面分成若干子分区；
    对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
    对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
    依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面；所述子分区图像特征信息包括：各子分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
  2. 根据权利要求1所述的方法，其特征在于，所述进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
  3. 根据权利要求1所述的方法，其特征在于，所述获取图像帧画面，将所述图像帧画面分成若干子分区包括：获取图像帧画面，依据所述图像帧画面的分辨率，将所述图像帧画面分成若干子分区。
  4. 一种图像处理方法，其特征在于，包括：
    获取图像帧画面，将所述图像帧画面分成若干子分区；
    对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
    依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
  5. 根据权利要求4所述的方法，其特征在于，所述对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息的步骤之后还包括：
    对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
    依据所述各子分区图像特征信息，对各子分区分别应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面的步骤包括：
    依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
  6. 根据权利要求4所述的方法，其特征在于，所述进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
  7. 根据权利要求5所述的方法，其特征在于，所述进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
  8. 根据权利要求4所述的方法，其特征在于，所述获取图像帧画面，将所述图像帧画面分成若干子分区包括：获取图像帧画面，依据所述图像帧画面的分辨率，将所述图像帧画面分成若干子分区。
  9. 一种图像处理装置，其特征在于，包括：
    划分模块，用于获取图像帧画面，将所述图像帧画面分成若干子分区；
    分析统计模块，用于对各子分区分别进行图像特征分析统计，得到各子分区图像特征信息；
    处理模块，用于依据所述各子分区图像特征信息，对各子分区应用增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
  10. 根据权利要求9所述的装置，其特征在于，所述装置还包括：
    敏感区检测模块，用于对所述各子分区图像特征信息进行敏感区域信息检测，得到所述各子分区图像中为敏感图像的区域；
    所述处理模块，还用于依据所述子分区图像和敏感图像区域的特征信息，对各子分区及敏感图像区域分别应用图像增强算法进行图像增强处理，将增强后的各个图像合并得到处理后的图像帧画面。
  11. 根据权利要求9所述的装置，其特征在于，所述处理模块进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
  12. 根据权利要求10所述的装置，其特征在于，所述处理模块进行图像增强处理包括：进行图像的滤波、平滑处理，和/或亮度、色彩调整。
  13. 根据权利要求9所述的装置，其特征在于，所述分区图像特征信息包括：各分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
  14. 根据权利要求10所述的装置，其特征在于，所述分区图像特征信息包括：各分区图像画面中的最大亮度、最小亮度、柱状图分布及色彩特色关键信息。
  15. 根据权利要求10所述的装置，其特征在于，
    所述划分模块，还用于获取图像帧画面，依据所述图像帧画面的分辨率，将所述图像帧画面分成若干子分区。
PCT/CN2016/084847 2016-03-01 2016-06-03 图像处理方法及装置 WO2017148035A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610116322.1 2016-03-01
CN201610116322.1A CN105809638A (zh) 2016-03-01 2016-03-01 图像处理方法及装置

Publications (1)

Publication Number Publication Date
WO2017148035A1 true WO2017148035A1 (zh) 2017-09-08

Family

ID=56466505

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/084847 WO2017148035A1 (zh) 2016-03-01 2016-06-03 图像处理方法及装置

Country Status (2)

Country Link
CN (1) CN105809638A (zh)
WO (1) WO2017148035A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108665428A (zh) * 2018-04-26 2018-10-16 青岛海信移动通信技术股份有限公司 图像增强方法、装置、设备及存储介质

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107291413A (zh) * 2017-06-08 2017-10-24 深圳Tcl新技术有限公司 显示终端、画面对比度提高方法及计算机可读存储介质
CN108010491B (zh) * 2017-11-08 2018-09-07 深圳市星王电子有限公司 一种基于瞳孔半径检测的背景光调节方法
CN109035183A (zh) * 2018-08-14 2018-12-18 信利光电股份有限公司 一种亮度调整方法、装置及电子设备
CN111353339A (zh) * 2018-12-21 2020-06-30 厦门歌乐电子企业有限公司 一种物体识别装置及方法
WO2022067489A1 (zh) * 2020-09-29 2022-04-07 深圳市大疆创新科技有限公司 图像处理方法、装置及图像采集设备
CN114449181B (zh) * 2020-11-05 2023-04-21 晶晨半导体(上海)股份有限公司 图像和视频处理方法及其系统、数据处理设备、介质
CN113850186A (zh) * 2021-09-24 2021-12-28 中国劳动关系学院 基于卷积神经网络的智能流媒体视频大数据分析方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050094892A1 (en) * 2003-11-04 2005-05-05 Samsung Electronics Co., Ltd. Method and apparatus for enhancing local luminance of image, and computer-readable recording medium for storing computer program
US20080042927A1 (en) * 2006-08-16 2008-02-21 Samsung Electronics Co., Ltd. Display apparatus and method of adjusting brightness thereof
CN101510302A (zh) * 2009-03-25 2009-08-19 北京中星微电子有限公司 一种图像增强方法和装置
CN102236883A (zh) * 2010-04-27 2011-11-09 株式会社理光 图像增强方法和装置、物体检测方法和装置
CN103024300A (zh) * 2012-12-25 2013-04-03 华为技术有限公司 一种高动态范围图像显示方法及装置
CN104252700A (zh) * 2014-09-18 2014-12-31 电子科技大学 一种红外图像的直方图均衡化方法
CN104580925A (zh) * 2014-12-31 2015-04-29 安科智慧城市技术（中国）有限公司 一种控制图像亮度的方法、装置及摄像机

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050094892A1 (en) * 2003-11-04 2005-05-05 Samsung Electronics Co., Ltd. Method and apparatus for enhancing local luminance of image, and computer-readable recording medium for storing computer program
US20080042927A1 (en) * 2006-08-16 2008-02-21 Samsung Electronics Co., Ltd. Display apparatus and method of adjusting brightness thereof
CN101510302A (zh) * 2009-03-25 2009-08-19 北京中星微电子有限公司 一种图像增强方法和装置
CN102236883A (zh) * 2010-04-27 2011-11-09 株式会社理光 图像增强方法和装置、物体检测方法和装置
CN103024300A (zh) * 2012-12-25 2013-04-03 华为技术有限公司 一种高动态范围图像显示方法及装置
CN104252700A (zh) * 2014-09-18 2014-12-31 电子科技大学 一种红外图像的直方图均衡化方法
CN104580925A (zh) * 2014-12-31 2015-04-29 安科智慧城市技术（中国）有限公司 一种控制图像亮度的方法、装置及摄像机

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108665428A (zh) * 2018-04-26 2018-10-16 青岛海信移动通信技术股份有限公司 图像增强方法、装置、设备及存储介质
CN108665428B (zh) * 2018-04-26 2022-11-11 青岛海信移动通信技术股份有限公司 图像增强方法、装置、设备及存储介质

Also Published As

Publication number Publication date
CN105809638A (zh) 2016-07-27

Similar Documents

Publication Publication Date Title
WO2017148035A1 (zh) 图像处理方法及装置
WO2020080665A1 (en) Methods and apparatuses for performing artificial intelligence encoding and artificial intelligence decoding on image
WO2018103187A1 (zh) 监控装置的监控画面形成方法和系统
EP3868096A1 (en) Artificial intelligence encoding and artificial intelligence decoding methods and apparatuses using deep neural network
WO2017107384A1 (zh) 液晶显示器的图像显示方法及液晶显示器
WO2018223602A1 (zh) 显示终端、画面对比度提高方法及计算机可读存储介质
WO2018192253A1 (zh) 视频信号的传输方法及装置
WO2015009107A1 (en) Method and apparatus for generating 3k-resolution display image for mobile terminal screen
WO2017190452A1 (zh) 背光自适应调节方法及装置
EP3844962A1 (en) Methods and apparatuses for performing artificial intelligence encoding and artificial intelligence decoding on image
WO2016091011A1 (zh) 字幕切换方法及装置
WO2017206368A1 (zh) 高动态范围画面切换方法及装置
WO2018227869A1 (zh) 显示屏画面调整方法、显示终端及可读存储介质
WO2017190445A1 (zh) Rgb图像处理方法及系统
WO2019076087A1 (zh) 电视机及其显示图效控制方法、计算机可读存储介质
WO2021261830A1 (en) Video quality assessment method and apparatus
WO2022025423A1 (en) Video quality assessing method and apparatus
WO2017041537A1 (zh) 视频帧率的调节方法及装置
WO2020080709A1 (en) Artificial intelligence encoding and artificial intelligence decoding methods and apparatuses using deep neural network
WO2016101305A1 (zh) 视频文件处理方法及显示装置
WO2017005062A1 (zh) 图片帧播放方法和装置
WO2022071647A1 (en) Video quality assessment method and apparatus
WO2017113600A1 (zh) 视频播放方法及装置
WO2017107386A1 (zh) 液晶显示终端色域提升方法及液晶显示终端
WO2019031676A1 (ko) 이미지 처리 방법 및 그에 따른 장치

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

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

Ref document number: 16892223

Country of ref document: EP

Kind code of ref document: A1

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

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 21/01/2019)

122 Ep: pct application non-entry in european phase

Ref document number: 16892223

Country of ref document: EP

Kind code of ref document: A1