WO2014106445A1 - 逆光检测方法及设备 - Google Patents

逆光检测方法及设备 Download PDF

Info

Publication number
WO2014106445A1
WO2014106445A1 PCT/CN2013/090806 CN2013090806W WO2014106445A1 WO 2014106445 A1 WO2014106445 A1 WO 2014106445A1 CN 2013090806 W CN2013090806 W CN 2013090806W WO 2014106445 A1 WO2014106445 A1 WO 2014106445A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
brightness
image block
detected
area
Prior art date
Application number
PCT/CN2013/090806
Other languages
English (en)
French (fr)
Inventor
周华
罗巍
邓斌
杜成
Original Assignee
华为终端有限公司
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 华为终端有限公司 filed Critical 华为终端有限公司
Priority to JP2015527788A priority Critical patent/JP5988119B2/ja
Priority to EP13870104.0A priority patent/EP2849431B1/en
Publication of WO2014106445A1 publication Critical patent/WO2014106445A1/zh
Priority to US14/566,020 priority patent/US9390475B2/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/11Region-based segmentation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/136Segmentation; Edge detection involving thresholding
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/194Segmentation; Edge detection involving foreground-background segmentation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70Circuitry for compensating brightness variation in the scene
    • H04N23/71Circuitry for evaluating the brightness variation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30168Image quality inspection

Definitions

  • Backlight detection method and device The present application claims priority to Chinese Patent Application No. CN201310005069.9, entitled “Backlight Detection Method and Equipment", filed on January 7, 2013, the entire contents of which are incorporated by reference. Combined in this application.
  • the present invention relates to the field of information technologies, and in particular, to a backlight detection method and device.
  • Background Art In an image forming process such as a digital camera or a video camera, in the case of backlight shooting, there is often a result that the background portion is too bright and the subject that is actually focused, that is, the object portion is too dark.
  • Backlight detection can be widely applied to intelligent scene recognition and automatic exposure control of imaging devices such as digital cameras and video cameras.
  • the backlight detection first divides the image into a plurality of rectangular blocks, and then searches for a continuous rectangular block whose luminance value is smaller than the brightness threshold, and counts the number of rectangular blocks satisfying the condition, if the total number of the rectangular blocks is greater than a preset threshold, and If the color saturation variance of the entire image is greater than the preset saturation threshold, then the image is determined to be a backlit scene.
  • Embodiments of the present invention provide a backlight detection method and device for improving the accuracy of backlight detection.
  • an embodiment of the present invention provides a backlight detection method, including:
  • the to-be-detected image includes an M-row and N-column image blocks, where M and N are positive integers; and determining, according to the brightness values of the image blocks,
  • the brightness relationship between adjacent image blocks includes:
  • the difference between the brightness value of the two adjacent image blocks and the first preset brightness threshold a relationship, determining a brightness relationship between the two adjacent image blocks including:
  • the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block.
  • the first preset brightness threshold is all the two adjacent ones in the to-be-detected image.
  • the determining the image to be detected according to a brightness relationship between adjacent image blocks Dark areas and bright areas in including:
  • the image with a small brightness is The block is a dark area
  • the brightness of one of the two adjacent image blocks is smaller than the brightness of the other image block, and wherein the brightness value of the image block having a small brightness value is greater than or equal to the second predetermined brightness threshold, the brightness is Small image blocks are bright and dark transition areas;
  • the image with the high brightness is The block is a bright area
  • the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block, and wherein the brightness value of the image block having the large brightness value is less than or equal to the third predetermined brightness threshold, the brightness is Large image blocks are bright and dark transition areas.
  • the method further includes: according to the relationship between the difference between the luminance value difference of the two adjacent image blocks and the first preset brightness threshold in each row of the image to be detected, the method further includes:
  • the determining, by the brightness relationship between adjacent image blocks, determining a dark area and a bright area in the image to be detected also includes: for each image block in each row or column, if the order from the first image block to the last image block of each row or column and the last image block to the first image block If the order is determined to be a bright area or a dark area, the image block is determined to be a bright area or a dark area; or
  • the order is determined to be a bright area, and the other order is determined to be a dark area, and the image block is determined to be a bright-dark transition area;
  • the order is determined to be a dark area or a bright area
  • the other order is determined to be a light dark transition area
  • the image block is determined to be a dark area or a bright area.
  • the determining, by the dark area and the bright area, whether the image to be detected is a backlight scene includes:
  • the image to be detected is determined to be a backlight scene, and the light-dark preset relationship satisfies any one or more of the following conditions. a combination of conditions: a ratio of a total area of the dark area to an area of the image to be detected is greater than a preset ratio; an average value of the brightness value of the dark area is less than a preset average value; and an average value of the brightness value of the bright area The difference from the average value of the brightness values of the dark areas is greater than the preset difference.
  • the determining, according to the dark area and the bright area, whether the image to be detected is a backlight scene also includes: Determining the backlighting scene, if the average gradient of the pixel points included in the dark area of the image to be detected is greater than a preset gradient threshold, determining that the image to be detected is a backlight scene, if the image to be detected is in the image to be detected The average gradient of the pixel points included in the dark region is less than the preset gradient threshold, and then the image to be detected is determined to be a non-backlight field.
  • an embodiment of the present invention provides a backlight detection device, including:
  • Obtaining module configured to acquire a brightness value of each image block in the image to be detected
  • Determining a module determining, according to the brightness value of each image block, a brightness relationship between adjacent image blocks; and processing module: configured to determine the to-be-detected according to a brightness relationship between adjacent image blocks a dark area and a bright area in the image, and determining whether the image to be detected is a backlight scene according to the dark area and the bright area.
  • the image to be detected includes an M-row and N-column image blocks, and M and N are both positive integers; the determining module is specifically configured to:
  • the determining module is specifically configured to: if the difference in luminance values of the two adjacent image blocks is greater than the a first preset brightness threshold, determining that the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block; or, if the brightness values of the two adjacent image blocks are If the absolute value of the difference is less than or equal to the first preset brightness threshold, it is determined that the brightness of the two adjacent image blocks is the same.
  • the first preset brightness threshold is any of the two adjacent ones in the image to be detected.
  • the processing module is specifically configured to:
  • the brightness of one of the two adjacent image blocks is smaller than the brightness of the other image block, and wherein the brightness value of the image block having a small brightness value is smaller than the second predetermined brightness threshold, determining that the brightness is small
  • the image block is a dark area; If the brightness of one of the two adjacent image blocks is smaller than the brightness of the other image block, and wherein the brightness value of the image block having a small brightness value is greater than or equal to the second predetermined brightness threshold, determining An image block with a small brightness is a bright and dark transition region;
  • the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block, and wherein the brightness value of the image block having the large brightness value is greater than the third predetermined brightness threshold, determining that the brightness is large
  • the image block is a bright area
  • the image block with high brightness is a bright and dark transition area.
  • the determining module is further configured to:
  • the processing module is further configured to: for each of the image blocks in each row or column, along each row or each The order of the first image block to the last image block of the column and the order of the last image block to the first image block are both determined as a bright area or a dark area, and then the image block is determined to be a bright area or a dark area; Or,
  • the order is determined to be a bright area, and the other order is determined to be a dark area, and then the image block is determined to be a light-dark transition area;
  • the order is determined to be a dark area or a bright area, and the other order is determined to be a bright-dark transition area, and it is determined that the image block is a dark area or a bright area.
  • the processing module is specifically configured to:
  • the image is a backlight scene, and the light-dark preset relationship satisfies any one of the following conditions or a combination of the plurality of conditions: a ratio of a total area of the dark area to an area of the image to be detected is greater than a preset ratio; The average value of the brightness value of the dark area is less than the preset average value; the difference between the average value of the brightness value of the bright area and the average value of the brightness value of the dark area is greater than the preset difference value.
  • the processing module is further configured to:
  • Determining the backlighting scene if the average gradient of the pixel points included in the dark area of the image to be detected is greater than a preset gradient threshold, determining that the image to be detected is a backlight scene, if the image to be detected is in the image to be detected The average gradient of the pixel points included in the dark region is less than the preset gradient threshold, and then the image to be detected is determined to be a non-backlit scene.
  • an embodiment of the present invention provides a backlight detection apparatus, including: a processor and a memory, where the memory stores an execution instruction, and when the backlight detection device is in operation, the processor communicates with the memory, The processor executing the execution instructions causes the backlighting detection device to perform any of the first to eighth possible implementations of the first aspect and the first aspect.
  • the backlight detection method and device provided by the embodiment of the present invention, the method obtains the brightness value of each image block in the image to be detected; determines the brightness relationship between adjacent image blocks according to the brightness value of each image block; The brightness relationship between the image blocks determines the dark area and the bright area in the image to be detected, and determines whether the image to be detected is a backlight scene according to the dark area and the bright area, thereby improving the accuracy of the backlight scene detection.
  • FIG. 1 is a flow chart of a first embodiment of a backlight detection method according to the present invention
  • FIG. 2 is a schematic diagram of luminance values of image blocks of a second embodiment of the backlight detection method of the present invention
  • FIG. 3B is a second schematic view of a bright and dark region along the row direction of the second embodiment of the backlight detection method of the present invention
  • FIG. 1 is a flow chart of a first embodiment of a backlight detection method according to the present invention
  • FIG. 2 is a schematic diagram of luminance values of image blocks of a second embodiment of the backlight detection method of the present invention
  • FIG. 3B is a second schematic view of a bright and dark region along the row direction of the second embodiment of the backlight detection method of the present invention
  • FIG. 3C is a schematic diagram of a light-dark region along a column direction according to a second embodiment of the backlight detection method of the present invention
  • FIG. 3D is a second schematic diagram of a light-dark region along the column direction of the second embodiment of the backlight detection method of the present invention
  • FIG. 5A is a schematic diagram of a non-backlighting scene according to a second embodiment of the method for detecting backlights according to the second embodiment of the present invention
  • FIG. FIG. 7 is a schematic structural diagram of the implementation of the back light detecting device of the present invention.
  • FIG. 1 is a flow chart of a first embodiment of a backlight detection method according to the present invention.
  • the present embodiment provides a backlight detection method, which can be performed by any device that performs a backlight detection method, and the device can be implemented by software and/or hardware.
  • the device may specifically be a terminal device such as a mobile phone, a video camera, or a camera having an imaging function, and the embodiment is not particularly limited herein.
  • the method in this embodiment may include: Step 101: Acquire a luminance value of each image block in an image to be detected; in a specific implementation process, the image to be detected may be uniformly divided into MXN images.
  • Block, the brightness of each image block can be calculated by using various existing image block brightness calculation methods, wherein M and N are positive integers, and the brightness value can be selected to a certain value range, for example: the value range is 0- 255, the intermediate brightness value is 128.
  • Lij represents the image block corresponding to the jth column of the i-th row, and the value of Lij represents the luminance value of the image block Lij. Where l ⁇ i ⁇ M,
  • Step 102 Determine, according to the brightness value of each image block, a brightness relationship between adjacent image blocks.
  • the image to be detected includes M rows and N columns. Therefore, according to the brightness values of the image blocks,
  • the brightness relationship between adjacent image blocks may specifically include two possible implementations: One possible implementation, determined according to the row of the image to be detected. Specifically, determining, according to the magnitude relationship between the difference between the luminance values of any two adjacent image blocks in each row of the image to be detected and the first preset luminance threshold, determining two adjacent images in each row of the image to be detected. The brightness relationship between the blocks. Another possible implementation is determined by the column of the image to be detected.
  • determining, according to the magnitude relationship between the difference between the luminance values of any two adjacent image blocks in each column of the image to be detected and the first preset brightness threshold determining two adjacent images in each column of the image to be detected.
  • the brightness relationship between the blocks It can be understood by those skilled in the art that the above two possible implementation manners may be determined only according to the row of the image to be detected, or only according to the column of the image to be detected, that is, determined by row or column, and may also be determined according to rows and columns. Also determined.
  • the first preset brightness threshold may be determined empirically, and may also be determined according to an average value of luminance difference values of adjacent image blocks. For the specific value of the first preset brightness threshold, the embodiment is not particularly limited herein.
  • the backlight detection method provided in this embodiment can determine the brightness relationship between two adjacent image blocks according to the row and/or column of the image to be detected, and can determine the position between adjacent image blocks in the image to be detected from multiple directions.
  • the brightness relationship makes the backlight detection accuracy high.
  • the luminance relationship between adjacent image blocks includes three cases, in the case of the i-th behavior, adjacent image blocks.
  • Li j and the image block Lij+1 are taken as an example, respectively, the brightness of the image block Lij+1 is smaller than the brightness of the image block Lij, or the brightness of the image block Lij+1 is greater than the brightness of the image block Lij, or the image block Lij+1 The brightness is equal to the brightness of the image block Lij. It should be noted that there is no limitation on the specific positional relationship between the image block Lij and the image block Lij+1 in the row, and the image block Lij may be closer to the first image block in the row or in the row. In the first image block, the image block Lij+1 may be the first image block closer to the row or the first image block in the row.
  • Step 103 Determine a dark area and a bright area in the image to be detected according to a brightness relationship between adjacent image blocks, and determine whether the image to be detected is a backlight scene according to the dark area and the bright area.
  • the dark area and the bright area in the image to be detected may be determined according to the brightness relationship between adjacent image blocks.
  • the image block Lij+1 is a dark area, if the image block If the brightness of Lij+1 is greater than the second preset brightness threshold, the image block Lij+1 is a bright-dark transition region; if the brightness of the image block Lij+1 is greater than the brightness of the image block Lij, and the brightness of the image block Lij+1 is greater than For the third preset brightness threshold, the image block Lij+1 is a bright area. If the brightness of the image block Lij+1 is less than the third preset brightness threshold, the image block Lij+1 is bright. Dark transition zone.
  • the second preset brightness threshold can avoid determining an image block with a large luminance value as a dark region
  • the third preset luminance threshold can avoid determining an image block having a small luminance value as a bright region.
  • the second preset brightness threshold and the third brightness threshold may be selected according to an empirical value, and the value range is larger and more flexible. Generally, the value of the second preset brightness threshold is less than the intermediate brightness value, and the value of the third preset brightness threshold is greater than the intermediate brightness value.
  • the image to be detected is a backlight scene.
  • the light-dark preset relationship satisfies any one of the following conditions or a combination of the plurality of conditions: the ratio of the total area of the dark area to the area of the image to be detected is greater than the preset ratio; the average value of the brightness value of the dark area is less than the preset average Value; The difference between the average value of the brightness value of the bright area and the average value of the brightness value of the dark area is greater than the preset difference.
  • the backlight detection method provided by the embodiment of the present invention obtains the brightness value of each image block in the image to be detected; determines the brightness relationship between adjacent image blocks according to the brightness value of each image block; The brightness relationship determines the dark area and the bright area in the image to be detected, and determines whether the image to be detected is a backlight scene according to the dark area and the bright area, which can not only improve the accuracy of the backlight scene detection when the degree of backlighting of the subject is greatly changed. It also improves the accuracy of the backlight scene when the position, area, shape, and the like of the subject are greatly changed.
  • the embodiment of the present invention will be described in detail below by taking a specific embodiment as an example. FIG.
  • FIG. 2 is a schematic diagram of luminance values of each image block in Embodiment 2 of the backlight detection method of the present invention.
  • the image to be detected is first divided into 11 rows and 20 columns, and 11 X 20 image blocks are shared, and the brightness values of the image blocks are determined.
  • the brightness value of each image block is indicated.
  • the luminance value difference values of any two adjacent image blocks are acquired. Specifically, it may be described in the direction of the row or in the direction of the column, respectively.
  • the difference in luminance values of any two adjacent image blocks in each row in the direction of the row that is, in the order of each row of the image to be detected from the first image block to the last image block, that is, from left to In the order of right, along the order of each line of the image to be detected from the last image block to the first image block, the difference in luminance values of any two adjacent image blocks in each line is obtained, that is, from right to left. order.
  • the left-to-right order is taken as an example, that is, the j direction indicated by the horizontal arrow shown in FIG. 2, and the luminance value difference of any two adjacent image blocks in the i-th row is obtained, that is, obtained.
  • the difference between Lij+1 and Lij is taken as an example, that is, the j direction indicated by the horizontal arrow shown in FIG.
  • the image block Lij may be closer to the first image block in the row or For the first image block in the row, the image block Lij+1 may be the first image block closer to the row or the first image block in the row.
  • the difference in luminance values of any two adjacent image blocks in each column along the direction of the column that is, in the order of each column of the image to be detected from the first image block to the last image block, that is, from top to bottom
  • the brightness value difference of any two adjacent image blocks in each column is obtained along the order of each column of the image to be detected from the last image block to the first image block, that is, in order from bottom to top
  • the order of the top to bottom is taken as an example, that is, the i direction indicated by the vertical arrow shown in FIG. 2, and the difference in luminance values of any two adjacent image blocks in the jth column is obtained, that is, Obtain the difference between Li+lj and Lij.
  • determining, according to the magnitude relationship between the difference between the brightness value of any two adjacent image blocks in each row of the image to be detected and the first preset brightness threshold determining two adjacent two images in each row of the image to be detected. Determining the relationship between the luminance values of the adjacent two image blocks in each column of the image to be detected and the first preset brightness threshold, determining any adjacent ones in each column of the image to be detected.
  • the first preset brightness threshold is an average value of the absolute values of the luminance value differences of all the two adjacent image blocks in the image to be detected. It can be understood by those skilled in the art that any two adjacent nodes are determined according to the row and the column. The manner of brightness relationship between image blocks is similar.
  • the present embodiment uses the behavior example, and, in the direction of the line, the order from the first image block to the last image block along each line of the image to be detected.
  • the manner of determining the brightness relationship between any two adjacent image blocks in the order from the last image block to the first image block along each line of the image to be detected is similar, and therefore, the embodiment of the present invention is hereby
  • the order in which each line of the image is detected from the first image block to the last image block is taken as an example to determine the brightness relationship between any two adjacent image blocks.
  • the first preset brightness threshold is equal to ⁇ £
  • the first preset brightness threshold is equal to 34.
  • the luminance relationship between the image block Lij+1 and the image block Lij is determined according to the difference between the difference between the image block Lij+1 and the image block Lij and the size of the AL. It can be understood by those skilled in the art that the values of the first preset brightness thresholds of different images to be detected are different.
  • the dark area and the bright area in the image to be detected are determined according to the brightness relationship between adjacent image blocks, and the order of the lines is also taken as an example here.
  • the order of the columns similar to the rows, this embodiment will not be described herein.
  • an image block of the first nine columns of the fourth row in FIG. 2 is taken as an example.
  • Table 1 shows the values of the corresponding Lij and Lij+1-Lij.
  • Ldark can take 120, which can avoid determining the image block with small brightness value as the third preset brightness threshold of the bright area.
  • Llight can take 130.
  • the dark area and the bright area may be identified, and for the bright and dark transition area, or the image block of the bright and dark area cannot be determined, no identification is performed.
  • the dark area and bright area identification process of the image to be detected the dark area is identified as a white dot, and the bright area is identified as a black fork.
  • L42-L41 -68 ⁇ -34, and L42 is smaller than Ldark, indicating that L42 is a dark area, and image block L42 is identified.
  • FIG. 3A is a schematic diagram of a bright and dark area along the row direction according to Embodiment 2 of the backlight detection method of the present invention.
  • L46-L45 5 ⁇ 34, then the brightness of L46 and L45 is the same, L46 is the dark area;
  • FIG. 3B is a second schematic diagram of a bright and dark region along the row direction according to Embodiment 2 of the backlight detection method of the present invention.
  • Figure 3B is in the row direction opposite to the row direction of Figure 3A.
  • FIG. 3C is a schematic diagram of a bright and dark area along the column direction according to the second embodiment of the backlight detection method of the present invention, that is, in the order from the top to the bottom of the column.
  • FIG. 3D is a second schematic diagram of the bright and dark regions along the column direction according to the second embodiment of the backlight detection method of the present invention, that is, the sequence from bottom to top along the column.
  • the acquisition method of FIG. 3B to FIG. 3D is similar to that of FIG. 3A, and details are not described herein again.
  • the merging of FIG. 3A to FIG. 3D is merged into FIG. 4, and the final dark area and the bright area of the image to be detected are obtained.
  • FIG. 3C is a schematic diagram of a bright and dark area along the column direction according to the second embodiment of the backlight detection method of the present invention, that is, in the order from the top to the bottom of the column.
  • FIG. 3D is a second schematic diagram of the bright and dark regions along the column direction according
  • FIGS. 3A and 3B are first determined according to FIGS. 3A and 3B, and the dark area and the bright area of the image to be detected along the column direction are determined according to FIGS. 3C and 3D.
  • the dark area and the bright area of the final image to be detected are determined according to the determination results of FIGS. 3A and 3B and the determination results of FIGS. 3C and 3D.
  • any one of the image blocks in each row or column is determined to be bright if the order from the first image block to the last image block of each row or column and the order from the last image block to the first image block are For a region or a dark region, the image block is determined to be a bright region or a dark region; or, for each image block in each row or column, if the first image block to the last image block along each row or column
  • the order of the sequence and the order of the last image block to the first image block is determined to be a bright area, and the other order is determined to be a dark area, and the image block is determined to be a bright-dark transition area; or, for each line or Any one of the image blocks in each column is determined to be dark if the order of the first image block to the last image block of each row or column and the order of the last image block to the first image block are dark.
  • the image block is determined as a bright region or a dark region.
  • the bright area and the dark area of the image to be detected are determined, and the light and dark relationship of all the image blocks can be comprehensively considered, and the bright area and the dark area of the detected image are corrected to improve the accuracy of backlight detection.
  • Figure 4 obtained according to the above method is the final dark area and bright area of the image to be detected according to the embodiment of the present invention. Whether the image to be detected is a backlit scene can be determined according to the dark area and the bright area shown in FIG. When the dark area and the bright area in the image to be detected satisfy the preset relationship, it is determined that the image to be detected is a backlight scene.
  • the light-dark preset relationship satisfies any one of the following conditions or a combination of the plurality of conditions: the ratio of the total area of the dark area to the area of the image to be detected is greater than the preset ratio; the average value of the brightness value of the dark area is less than the preset average Value; the average value of the brightness value of the bright area and the dark area The difference in the average value of the luminance values is greater than the preset difference.
  • the average gradient of the pixel included in the dark region in the image to be detected is greater than the preset gradient threshold, it is determined that the image to be detected is a backlight scene, and the pixel included in the dark region in the image to be detected If the average gradient is less than the preset gradient threshold, it is determined that the subject is a black object, and the image to be detected is a non-backlit scene.
  • the specific value of the preset gradient threshold can be set according to experience. In this embodiment, the preset gradient threshold is set to 4, and the average gradient of the pixel points included in the dark region in FIG. 4 is 6.1.
  • FIG. 5A is a schematic diagram 1 of a non-backlighting scene according to a second embodiment of the backlight detection method of the present invention
  • FIG. 5B is a second schematic diagram of a non-backlighting scene according to a second embodiment of the backlight detection method of the present invention.
  • the subject of the image to be detected is a black object
  • the bright area and the dark area of the image to be detected are shown in FIG. 5B
  • the average gradient of the pixel points included in the dark area in FIG. 5B is calculated as 0.
  • the backlight detection method determines the dark region and the bright region in the image to be detected according to the brightness relationship between adjacent image blocks, and does not need to set an accurate threshold value for determining the dark region.
  • FIG. 6 is a schematic structural view of a first embodiment of the backlight detecting device of the present invention.
  • the backlight detecting device provided in this embodiment may be a terminal device such as a mobile phone, a camera, a video camera, a computer, or the like including an imaging lens.
  • the image to be detected is obtained by the lens, and each module pair in the backlight detecting device shown in FIG. 6 provided by the embodiment of the present invention detects whether the image to be detected is a backlight scene.
  • the backlight detection device provided by the embodiment of the present invention includes an obtaining module 601, a determining module 602, and a processing module 603.
  • the obtaining module 601 is configured to obtain a brightness value of each image block in the image to be detected.
  • the determining module 602 is configured to determine a brightness relationship between the adjacent image blocks according to the brightness values of the image blocks, and the processing module 603 is configured to determine the to-be-detected according to a brightness relationship between the adjacent image blocks.
  • the image to be detected includes M rows and N columns of image blocks, and M and N are both positive integers; the determining module 602 is specifically configured to: according to any adjacent two of each row of the image to be detected.
  • the determining module 602 is specifically configured to: if the difference in luminance values of the two adjacent image blocks is greater than the first preset brightness threshold, determine the two adjacent image blocks The brightness of one of the image blocks is greater than the brightness of the other image block; or, if the absolute value of the brightness value difference of the two adjacent image blocks is less than or equal to the first preset brightness threshold, The brightness of any two adjacent image blocks is the same.
  • the first preset brightness threshold is an average value of absolute values of luminance value differences of all the two adjacent image blocks in the image to be detected.
  • the processing module 603 is specifically configured to: if a brightness of one image block of the two adjacent image blocks is smaller than a brightness of another image block, and wherein a brightness value of the image block having a small brightness value is smaller than the first Two preset brightness thresholds are determined, wherein the image block in which the brightness is small is determined to be a dark area; if the brightness of one of the two adjacent image blocks is smaller than the brightness of the other image block, and the image in which the brightness value is small If the brightness value of the block is greater than or equal to the second preset brightness threshold, determining that the image block having a small brightness is a light-dark transition region; If the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block, and wherein the brightness value of the image block having the large brightness value is greater than the third predetermined brightness threshold, determining that the brightness is large The image block is a bright area; if the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block, and the brightness value of the image
  • the determining module 602 is further configured to: acquire, according to an order of the first image block to the last image block, each row of the image to be detected, obtain brightness values of any two adjacent image blocks in each row. a difference value, along the order of each line of the image to be detected from the last image block to the first image block, obtaining a luminance value difference value of any two adjacent image blocks in each row; along the image to be detected The order of the brightness values of any two adjacent image blocks in each column is obtained from the order of the first image block to the last image block, along each column of the image to be detected from the last image block to The order of the first image block is to obtain the difference in luminance values of any two adjacent image blocks in each column.
  • the processing module 603 is further configured to: for each image block in each row or column, the order of the first image block to the last image block along each row or column and the last image along the last image
  • the order of the block to the first image block is determined to be a bright area or a dark area, and then the image block is determined to be a bright area or a dark area; or, for each image block in each row or column, if each line is along each line Or the order of the first image block to the last image block of each column and the order of the last image block to the first image block are determined as a bright area, and the other order is determined to be a dark area, then the determination is made
  • the image block is a bright-dark transition region; or, for each image block in each row or column, the order from the first image block to the last image block along each row or column and along the last image block
  • One of the order to the first image block is determined to be a dark area or a bright area, and the other order is determined to be a light-dark
  • the processing module 603 is specifically configured to: if the dark area and the bright area in the to-be-detected image meet the light-dark preset relationship, determine that the image to be detected is a backlight scene, and the light-dark pre- The set relationship satisfies any one of the following conditions or a combination of the following conditions: the ratio of the total area of the dark area to the area of the image to be detected is greater than a preset ratio; the average value of the dark area is smaller than the preset An average value; the average value of the brightness value of the bright area is larger than the average value of the brightness value of the dark area The preset difference.
  • processing module 603 is further configured to: verify the backlighting scenario, if the average gradient of the pixel points included in the dark region in the image to be detected is greater than a preset gradient threshold, determine the to-be-detected The image is a backlit scene, and if the average gradient of the pixel points included in the dark region in the image to be detected is less than a preset gradient threshold, determining that the image to be detected is a non-backlight field
  • FIG. 7 is a schematic structural view of the second implementation of the backlight detecting device of the present invention.
  • the backlight detecting device provided in this embodiment may be a terminal device such as a mobile phone, a camera, a video camera, a computer, or the like including an imaging lens.
  • the image to be detected can be obtained by the lens, and the components in the backlight detection device shown in FIG. 7 provided by the embodiment of the present invention are used to detect whether the image to be detected is a backlight scene. As shown in FIG.
  • the backlight detecting apparatus 70 includes a processor 701 and a memory 702.
  • the backlight detecting device 70 further includes a lens 703, and the lens 703 can be connected to the processor 701.
  • the processor 701 can obtain an image to be detected through the lens 703, divide the image to be detected into image blocks, and calculate a brightness value of each image block.
  • the memory 702 stores execution instructions.
  • the processor 701 communicates with the memory 702, and the processor 701 calls an execution instruction in the memory 702 for performing the following operations - acquiring each image block in the image to be detected.
  • a brightness value determining, according to the brightness value of each image block, a brightness relationship between adjacent image blocks; determining a dark area in the image to be detected according to a brightness relationship between adjacent image blocks Brightening the area, and determining, according to the dark area and the bright area, whether the image to be detected is a backlight scene.
  • the image to be detected includes M rows and N columns of image blocks, and M and N are positive integers; and determining, according to the brightness values of the image blocks, brightness relationships between adjacent image blocks, Determining - determining, according to a magnitude relationship between a difference in luminance value of any two adjacent image blocks in each row of the image to be detected and a first preset brightness threshold, any adjacent one of each row of the image to be detected a brightness relationship between the two image blocks; and/or, according to the magnitude relationship between the difference between the brightness values of any two adjacent image blocks in each column of the image to be detected and the first preset brightness threshold, Describe the brightness between any two adjacent image blocks in each column of the detected image Department.
  • determining, according to a magnitude relationship between a difference value of the luminance values of the two adjacent image blocks and a first preset brightness threshold determining a brightness relationship between the two adjacent image blocks, including And if the luminance value difference of the two adjacent image blocks is greater than the first preset brightness threshold, the brightness of one of the two adjacent image blocks is greater than the other image block. Or the brightness of the adjacent two image blocks is the same if the absolute value of the luminance value difference of the two adjacent image blocks is less than or equal to the first preset brightness threshold.
  • the first preset brightness threshold is an average value of absolute values of luminance value differences of all the two adjacent image blocks in the image to be detected.
  • the determining the dark area and the bright area in the image to be detected according to the brightness relationship between the adjacent image blocks includes: if the brightness of one of the two adjacent image blocks is If the brightness of the image block is smaller than the brightness of the other image block, and the brightness value of the image block having the small brightness value is smaller than the second predetermined brightness threshold, the image block with the small brightness is a dark area; if the image block is any adjacent two If the brightness of one image block is smaller than the brightness of another image block, and the brightness value of the image block having a small brightness value is greater than or equal to the second predetermined brightness threshold, the image block with a small brightness is a bright-dark transition region; The brightness of one of the two adjacent image blocks is greater than the brightness of the other image block, and wherein the brightness value of the image block having the large brightness value is greater than the third predetermined brightness threshold, wherein the image block with the large brightness is a bright area; if the brightness of one of the two adjacent image blocks is greater than the brightness of the other image block, and the brightness
  • the method further includes: along the image to be detected The order of the brightness values of any two adjacent image blocks in each row is obtained from the order of the first image block to the last image block, along each line of the image to be detected from the last image block to the first The order of an image block, obtaining the difference in luminance values of any two adjacent image blocks in each row; Determining, according to a magnitude relationship between a difference in luminance value of any two adjacent image blocks in each column of the image to be detected and a first preset brightness threshold, determining any adjacent one of each column of the image to be detected Before the brightness relationship between the two image blocks, the method further includes: acquiring, according to the order of each column of the image to be detected from the first image block to the last image block, acquiring two adjacent image blocks in each column The luminance value difference is obtained by acquiring the luminance value difference value of any two adjacent image blocks in each column along the order of
  • the determining the dark area and the bright area in the image to be detected according to the brightness relationship between adjacent image blocks further includes: for each image block in each row or column, if each edge The order of the first image block to the last image block of the row or each column and the order of the last image block to the first image block are determined to be a bright region or a dark region, and the image block is determined to be a bright region or Dark area; or, for each image block in each row or column, the order from the first image block to the last image block in each row or column and the last image block to the first image block One of the sequences is determined to be a bright area, and the other order is determined to be a dark area, and the image block is determined to be a bright-dark transition area; or, for each image block in each row or column, if each line is along each line Or the order of the first image block to the last image block of each column and the order of the last image block to the first image block are determined as a dark area or a bright area, and the other order is
  • determining, according to the dark area and the bright area, whether the image to be detected is a backlight scene including: if the dark area and the bright area in the image to be detected satisfy a light-dark preset relationship, determining The image to be detected is a backlight scene, and the light-dark preset relationship satisfies any one of the following conditions or a combination of the plurality of conditions: a ratio of a total area of the dark area to an area of the image to be detected is greater than a preset The ratio of the brightness value of the dark area is less than a preset average value; the difference between the average value of the brightness value of the bright area and the average value of the brightness value of the dark area is greater than a preset difference.
  • the method further includes: verifying the backlight scene, if the pixel included in the dark area in the image to be detected If the average gradient of the point is greater than the preset gradient threshold, determining that the image to be detected is a backlight scene, if the dark area in the image to be detected The average gradient of the pixel points included in the domain is less than a preset gradient threshold, and then determining that the image to be detected is a non-backlight field
  • the backlight detection device of this embodiment may be used to implement the technical solution of the foregoing method embodiment, and the implementation principle and the technical effect are similar, and details are not described herein again.
  • the disclosed apparatus and method can be implemented in other ways.
  • the device embodiments described above are only schematic.
  • the division of the unit or module is only a logical function division.
  • there may be another division manner for example, multiple units or modules may be used. Combined or can be integrated 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, device or module, and may be electrical, mechanical or otherwise.
  • the modules described as separate components may or may not be physically separate.
  • the components displayed as modules may or may not be physical modules, that is, may be located in one place, or may be distributed to multiple network units.
  • modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. It will be understood by those skilled in the art that all or part of the steps of implementing the above method embodiments may be performed by hardware associated with the program instructions.
  • the aforementioned program can be stored in a computer readable storage medium. When the program is executed, the steps including the foregoing method embodiments are performed; and the foregoing storage medium includes: various media that can store program codes, such as wake up, wake up, disk or optical disk.

Abstract

本发明实施例提供一种逆光检测方法及设备,该方法包括获取待检测图像中的各图像块的亮度值;根据所述各图像块的亮度值,确定相邻所述图像块之间的亮度关系;根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区域和亮区域,并根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。本发明实施例提供的逆光检测方法及设备可以提高逆光检测的准确性。

Description

逆光检测方法及设备 本申请要求于 2013年 01月 07日提交中国专利局、 申请号为 CN201310005069. 9、 发明名称为 "逆光检测方法及设备"的中国专利申请的优先权, 其全部内容通过引用结 合在本申请中。 技术领域 本发明实施例涉及信息技术领域, 尤其涉及一种逆光检测方法及设备。 背景技术 在数字照相机或摄像机等成像设备进行成像的过程中, 在逆光拍摄的情况下, 往往 出现背景部分过亮而真正所关注的被摄物体即对象部分过暗的结果。逆光检测可以广泛 应用于数字照相机及摄像机等成像装置的智能场景识别及自动曝光控制等功能。
现有技术中的逆光检测, 先将图像划分为多个矩形块, 然后搜索亮度值小于亮度阈 值的连续矩形块, 统计满足条件的矩形块的数量, 若矩形块的总数大于预设门限, 且整 个图像的颜色饱和度方差大于预设饱和度门限, 则确定该图像为逆光场景。
然而, 现有技术的逆光检测准确性低。
发明内容 本发明实施例提供一种逆光检测方法及设备, 用以提高逆光检测的准确性。
第一方面, 本发明实施例提供一种逆光检测方法, 包括:
获取待检测图像中的各图像块的亮度值;
根据所述各图像块的亮度值, 确定相邻所述图像块之间的亮度关系;
根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区域和亮区域, 并 根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。
结合第一方面, 在第一种可能的实现方式中, 所述待检测图像中包括 M行 N列图像 块, M和 N均为正整数; 所述根据所述各图像块的亮度值, 确定相邻所述图像块之间的 亮度关系, 包括:
根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每行中任意相邻的两个图像块之间的亮度关 系; 和 /或,
根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮度关 系。
结合第一方面的第一种可能的实现方式, 在第二种可能的实现方式中, 所述根据所 述任意相邻的两个图像块的亮度值差值与第一预设亮度阈值的大小关系,确定所述任意 相邻的两个图像块之间的亮度关系, 包括:
若所述任意相邻的两个图像块的亮度值差值大于所述第一预设亮度阈值, 则所述任 意相邻的两个图像块中的一个图像块的亮度大于另一个图像块的亮度; 或者, 若所述任 意相邻的两个图像块的亮度值差值的绝对值小于或等于第一预设亮度阈值, 则所述任意 相邻的两个图像块的亮度相同。
结合第一方面的第一种或第二种可能的实现方式, 在第三种可能的实现方式中, 所 述第一预设亮度阈值为所述待检测图像中所有所述任意相邻的两个图像块的亮度值差 值绝对值的平均值。
结合第一方面的第一种至第三种任一种可能的实现方式,在第四种可能的实现方式 中, 所述根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区域和亮区 域, 包括:
若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值小于第二预设亮度阈值, 则其中亮度小的图像块为暗区 域;
若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值大于等于所述第二预设亮度阈值, 则其中亮度小的图像块 为亮暗过渡区域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值大于第三预设亮度阈值, 则其中亮度大的图像块为亮区 域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值小于等于所述第三预设亮度阈值, 则其中亮度大的图像块 为亮暗过渡区域。
结合第一方面的第一种至第四种任一种可能的实现方式,在第五种可能的实现方式 中,所述根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设 亮度阈值的大小关系之前, 还包括:
沿所述待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获取每行中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每行从最后一个图像块至第一 个图像块的顺序, 获取每行中任意相邻的两个图像块的亮度值差值;
所述根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预 设亮度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮 度关系之前, 还包括:
沿所述待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获取每列中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每列从最后一个图像块至第一 个图像块的顺序, 获取每列中任意相邻的两个图像块的亮度值差值。
结合第一方面的第五种可能的实现方式, 在第六种可能的实现方式中, 所述根据相 邻所述图像块之间的亮度关系确定所述待检测图像中的暗区域和亮区域, 还包括: 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 所述图像块确定为亮区域或暗区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则所述图像块确定为亮暗过渡区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则所述图像块确定为暗区域或亮区域。
结合第一方面的第六种可能的实现方式, 在第七种可能的实现方式中, 所述根据所 述暗区域和亮区域确定所述待检测图像是否为逆光场景, 包括:
若所述待检测图像中的所述暗区域和亮区域满足亮暗预设关系, 则确定所述待检测 图像为逆光场景, 所述亮暗预设关系满足如下中的任一条件或多种条件的组合: 所述暗 区域的总面积占所述待检测图像的面积的比例大于预设比值; 所述暗区域的亮度值平均 值小于预设平均值; 所述亮区域的亮度值平均值与所述暗区域的亮度值平均值的差值大 于预设差值。
结合第一方面的第一种至第七种任一种可能的实现方式,在第八种可能的实现方式 中, 所述根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景之后, 还包括: 对所述逆光场景进行验证,若所述待检测图像中的暗区域所包含的像素点的平均梯 度大于预设梯度阈值, 则确定所述待检测图像为逆光场景, 若所述待检测图像中的暗区 域所包含的象素点的平均梯度小于预设梯度阈值, 则确定所述待检测图像为非逆光场 牙、。
第二方面, 本发明实施例提供一种逆光检测设备, 包括:
获取模块: 用于获取待检测图像中的各图像块的亮度值;
确定模块:用于根据所述各图像块的亮度值,确定相邻所述图像块之间的亮度关系; 处理模块: 用于根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区 域和亮区域, 并根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。
结合第二方面, 在第二种可能的实现方式中, 所述待检测图像中包括 M行 N列图像 块, M和 N均为正整数; 所述确定模块具体用于:
根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每行中任意相邻的两个图像块之间的亮度关 系; 和 /或,
根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮度关 系。
结合第二方面的第一种可能的实现方式, 在第二种可能的实现方式中, 所述确定模 块具体用于- 若所述任意相邻的两个图像块的亮度值差值大于所述第一预设亮度阈值, 则确定所 述任意相邻的两个图像块中的一个图像块的亮度大于另一个图像块的亮度; 或者, 若所 述任意相邻的两个图像块的亮度值差值的绝对值小于或等于第一预设亮度阈值, 则确定 所述任意相邻的两个图像块的亮度相同。
结合第二方面的第一种或第二种可能的实现方式, 在第三种可能的实现方式中, 所 述第一预设亮度阈值为所述待检测图像中所有所述任意相邻的两个图像块的亮度值差 值绝对值的平均值。
结合第二方面的第一种至第三种任一种可能的实现方式,在第四种可能的实现方式 中, 所述处理模块具体用于:
若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值小于第二预设亮度阈值, 则确定其中亮度小的图像块为暗 区域; 若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值大于等于所述第二预设亮度阈值, 则确定其中亮度小的图 像块为亮暗过渡区域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值大于第三预设亮度阈值, 则确定其中亮度大的图像块为亮 区域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值小于等于所述第三预设亮度阈值, 则确定其中亮度大的图 像块为亮暗过渡区域。
结合第二方面的第一种至第四种任一种可能的实现方式,在第五种可能的实现方式 中, 所述确定模块还用于:
沿所述待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获取每行中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每行从最后一个图像块至第一 个图像块的顺序, 获取每行中任意相邻的两个图像块的亮度值差值;
沿所述待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获取每列中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每列从最后一个图像块至第一 个图像块的顺序, 获取每列中任意相邻的两个图像块的亮度值差值。
结合第二方面的第五种可能的实现方式, 在第六种可能的实现方式中, 所述处理模 块还用于- 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 确定所述图像块为亮区域或暗区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则确定所述图像块为亮暗过渡区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则确定所述图像块为暗区域或亮区域。
结合第二方面的第六种可能的实现方式, 在第七种可能的实现方式中, 所述处理模 块具体用于:
若所述待检测图像中的所述暗区域和亮区域满足亮暗预设关系, 则确定所述待检测 图像为逆光场景, 所述亮暗预设关系满足如下中的任一条件或多种条件的组合: 所述暗 区域的总面积占所述待检测图像的面积的比例大于预设比值; 所述暗区域的亮度值平均 值小于预设平均值; 所述亮区域的亮度值平均值与所述暗区域的亮度值平均值的差值大 于预设差值。
结合第二方面的第一种至第七种任一种可能的实现方式,在第八种可能的实现方式 中, 所述处理模块还用于:
对所述逆光场景进行验证,若所述待检测图像中的暗区域所包含的像素点的平均梯 度大于预设梯度阈值, 则确定所述待检测图像为逆光场景, 若所述待检测图像中的暗区 域所包含的象素点的平均梯度小于预设梯度阈值, 则确定所述待检测图像为非逆光场 景。
第三方面, 本发明实施例提供一种逆光检测设备, 包括: 处理器和存储器, 所述存 储器存储执行指令, 当所述逆光检测设备运行时, 所述处理器与所述存储器之间通信, 所述处理器执行所述执行指令使得所述逆光检测设备执行如第一方面以及第一方面的 第一种至第八种任一种可能的实现方式。
本发明实施例提供的逆光检测方法及设备, 该方法通过获取待检测图像中的各图像 块的亮度值; 根据各图像块的亮度值, 确定相邻图像块之间的亮度关系; 根据相邻图像 块之间的亮度关系确定待检测图像中的暗区域和亮区域,根据暗区域和亮区域确定待检 测图像是否为逆光场景, 提高逆光场景检测的准确性。
附图说明 为了更清楚地说明本发明实施例或现有技术中的技术方案, 下面将对实施例或现有 技术描述中所需要使用的附图作一简单地介绍, 显而易见地, 下面描述中的附图是本发 明的一些实施例, 对于本领域普通技术人员来讲, 在不付出创造性劳动性的前提下, 还 可以根据这些附图获得其他的附图。 图 1为本发明逆光检测方法实施例一的流程图; 图 2为本发明逆光检测方法实施例二的各图像块的亮度值示意图; 图 3A为本发明逆光检测方法实施例二的沿行方向亮暗区域示意图一; 图 3B为本发明逆光检测方法实施例二的沿行方向亮暗区域示意图二; 图 3C为本发明逆光检测方法实施例二的沿列方向亮暗区域示意图一; 图 3D为本发明逆光检测方法实施例二的沿列方向亮暗区域示意图二; 图 4为本发明逆光检测方法实施例二的亮暗区域示意图一; 图 5A为本发明逆光检测方法实施例二的非逆光场景示意图一; 图 5B为本发明逆光检测方法实施例二的非逆光场景示意图二; 图 6为本发明逆光检测设备实施一的结构示意图; 图 7为本发明逆光检测设备实施二的结构示意图。
具体实施方式 为使本发明实施例的目的、 技术方案和优点更加清楚, 下面将结合本发明实施例中 的附图, 对本发明实施例中的技术方案进行清楚、 完整地描述, 显然, 所描述的实施例 是本发明一部分实施例, 而不是全部的实施例。 基于本发明中的实施例, 本领域普通技 术人员在没有做出创造性劳动前提下所获得的所有其他实施例, 都属于本发明保护的范 围。 图 1为本发明逆光检测方法实施例一的流程图。本实施例提供了一种逆光检测方法, 该方法可以由任意执行逆光检测方法的设备来执行, 该设备可以通过软件和 /或硬件实 现。 该设备具体可以为具有成像功能的手机、 摄像机、 照相机等终端设备, 本实施例在 此不作特别限制。 如图 1所示, 本实施例的方法可以包括: 步骤 101 : 获取待检测图像中的各图像块的亮度值; 在具体实现过程中, 通常可以将该待检测图像均匀的划分为 MXN个图像块, 可以 采用现有的各种图像块亮度计算方法来计算各图像块的亮度,其中, M和 N均为正整数, 亮度值可以选取一定的取值范围, 例如: 取值范围为 0-255, 中间亮度值为 128。 Lij 代表第 i行第 j列对应的图像块, Lij的取值代表图像块 Lij的亮度值。其中, l^ i^M,
步骤 102: 根据各图像块的亮度值, 确定相邻图像块之间的亮度关系; 在实际应用过程中, 待检测图像包括 M行 N列, 因此, 根据各图像块的亮度值, 确 定相邻图像块之间的亮度关系具体可以包括两种可能的实现情况: 一种可能的实现情况, 根据待检测图像的行确定。 具体地, 根据待检测图像的每行 中任意相邻的两个图像块的亮度值差值与第一预设亮度阈值的大小关系,确定待检测图 像的每行中任意相邻的两个图像块之间的亮度关系。 另一种可能的实现情况, 根据待检测图像的列确定。 具体地, 根据待检测图像的每 列中任意相邻的两个图像块的亮度值差值与第一预设亮度阈值的大小关系,确定待检测 图像的每列中任意相邻的两个图像块之间的亮度关系。 本领域技术人员可以理解, 上述两种可能的实现方式, 可以仅根据待检测图像的行 确定, 或仅根据待检测图像的列确定, 即行或列择一的方式确定, 还可以根据行和列同 时确定。 第一预设亮度阈值可以根据经验确定, 还可以根据相邻图像块亮度差值的平均 值确定。 对于第一预设亮度阈值的具体值, 本实施例在此不做特别限制。 本实施例提供的逆光检测方法, 通过根据待检测图像的行和 /或列确定任意相邻的 两个图像块之间的亮度关系, 能够从多方位确定待检测图像中相邻图像块之间的亮度关 系, 使逆光检测准确性高。 特别地, 相邻图像块之间的亮度关系包括三种情况, 以第 i行为例, 相邻的图像块
Li j与图像块 Lij+1为例, 分别为图像块 Lij+1的亮度小于图像块 Lij的亮度, 或者, 图像块 Lij+1的亮度大于图像块 Lij的亮度,或者,图像块 Lij+1的亮度等于图像块 Lij 的亮度。 需要说明的是, 这里对图像块 Lij与图像块 Lij+1在所在行中的具体位置关系 并不做出限制, 图像块 Lij可以是更靠近所在行中的第一个图像块或者为所在行中的第 一个图像块, 图像块 Lij+1可以是更靠近所在行中的第一个图像块或者为所在行中的第 一个图像块。 步骤 103: 根据相邻图像块之间的亮度关系确定待检测图像中的暗区域和亮区域, 并根据暗区域和亮区域确定待检测图像是否为逆光场景。 在实际应用过程中, 可根据相邻图像块之间的亮度关系, 确定待检测图像中的暗区 域和亮区域。 在具体实现过程中, 若图像块 Lij+1的亮度小于图像块 Lij的亮度, 且图 像块 Lij+1的亮度小于第二预设亮度阈值, 则图像块 Lij+1为暗区域, 若图像块 Lij+1 的亮度大于第二预设亮度阈值, 则图像块 Lij+1为亮暗过渡区域; 若图像块 Lij+1的亮 度大于图像块 Lij 的亮度, 且图像块 Lij+1 的亮度大于第三预设亮度阈值, 则图像块 Lij+1为亮区域, 若图像块 Lij+1的亮度小于第三预设亮度阈值, 则图像块 Lij+1为亮 暗过渡区域。 本领域技术人员可以理解,第二预设亮度阈值可以避免将亮度值很大的图像块确定 为暗区域, 第三预设亮度阈值可以避免将亮度值很小的图像块确定为亮区域。 其中, 第 二预设亮度阈值与第三亮度阈值可以根据经验值进行选取, 且取值范围较大, 较灵活。 一般情况下, 第二预设亮度阈值的取值小于中间亮度值, 第三预设亮度阈值的取值大于 中间亮度值。 在确定待检测图像中的暗区域和亮区域后,根据暗区域和亮区域确定待检测图像是 否为逆光场景。 具体地, 若待检测图像中的暗区域和亮区域满足亮暗预设关系, 则确定 待检测图像为逆光场景。其中,亮暗预设关系满足如下中的任一条件或多种条件的组合: 暗区域的总面积占待检测图像的面积的比例大于预设比值; 暗区域的亮度值平均值小于 预设平均值; 亮区域的亮度值平均值与暗区域的亮度值平均值的差值大于预设差值。 本发明实施例提供的逆光检测方法, 通过获取待检测图像中的各图像块的亮度值; 根据各图像块的亮度值, 确定相邻图像块之间的亮度关系; 根据相邻图像块之间的亮度 关系确定待检测图像中的暗区域和亮区域,根据暗区域和亮区域确定待检测图像是否为 逆光场景, 不仅能够提高拍摄主体的逆光程度发生较大变化时的逆光场景检测的准确 性, 还能够提高拍摄主体的位置、 面积、 形状等发生较大变化时逆光场景的准确性。 下面以一个具体实施例为例, 对本发明实施例做详细说明。 图 2为本发明逆光检测方法实施例二的各图像块的亮度值示意图。 如图 2所示, 先 将待检测图像划分为 11行, 20列, 共有 11 X 20个图像块, 并确定各图像块的亮度值, 本领域技术人员可以理解, 为了便于说明, 在各图像块中, 标明各图像块的亮度值。 在获得各图像块的亮度值之后,获取任意相邻的两个图像块的亮度值差值。具体地, 可以沿行的方向, 也可以沿列的方向, 以下分别说明。 沿行的方向, 即沿待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获 取每行中任意相邻的两个图像块的亮度值差值, 即按照从左到右的顺序, 沿待检测图像 的每行从最后一个图像块至第一个图像块的顺序, 获取每行中任意相邻的两个图像块的 亮度值差值, 即按照从右到左的顺序。 本实施例在此以从左到右的顺序为例, 即图 2所 示的水平箭头所示的 j方向, 获取第 i行中任意相邻的两个图像块的亮度值差值, 即获 取 Lij+1与 Lij的差值。 需要说明的是, 这里对图像块 Li j与图像块 Li j+1在所在行中 的具体位置关系并不做出限制, 图像块 Lij可以是更靠近所在行中的第一个图像块或者 为所在行中的第一个图像块, 图像块 Lij+1可以是更靠近所在行中的第一个图像块或者 为所在行中的第一个图像块。 沿列的方向, 即沿待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获 取每列中任意相邻的两个图像块的亮度值差值, 即按照从上到下的顺序, 沿待检测图像 的每列从最后一个图像块至第一个图像块的顺序, 获取每列中任意相邻的两个图像块的 亮度值差值, 即按照从下至上的顺序。 本实施例在此以从上至下的顺序为例, 即图 2所 示的竖直箭头所示的 i方向, 获取第 j列中任意相邻的两个图像块的亮度值差值, 即获 取 Li+lj与 Lij的差值。 再根据待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮度 阈值的大小关系, 确定待检测图像的每行中任意两个相邻的两个图像之间的亮度关系, 根据待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮度阈值的 大小关系, 确定待检测图像的每列中任意相邻的两个图像块之间的亮度关系。 其中, 第 一预设亮度阈值为待检测图像中所有任意相邻的两个图像块的亮度值差值绝对值的平 均值 本领域技术人员可以理解,根据行和列确定任意相邻的两个图像块之间的亮度关系 的方式类似, 因此, 本实施例在此以行为例, 同时, 沿行的方向时, 沿待检测图像的每 行从第一个图像块至最后一个图像块的顺序与沿待检测图像的每行从最后一个图像块 至第一个图像块的顺序确定任意相邻的两个图像块之间的亮度关系的方式类似, 因此, 本发明实施例在此以沿待检测图像的每行从第一个图像块至最后一个图像块的顺序为 例对确定任意相邻的两个图像块之间的亮度关系进行详细说明。 在沿待检测图像的每行从第一个图像块至最后一个图像块的顺序确定任意相邻的 两个图像块之间的亮度关系时, 第一预设亮度阈值 等于^ £ | ϊ+1ΑϊΙ, 根据图 2所 示的各图像块的亮度值,可以计算得到第一预设亮度阈值 等于 34。根据图像块 Lij+1 与图像块 Lij的差值与 A L的大小关系, 确定图像块 Lij+1与图像块 Lij的亮度关系。 本领域技术人员可以理解, 不同的待检测图像, 其第一预设亮度阈值 的取值不同。 在沿待检测图像的每列从第一个图像块至最后一个图像块的顺序确定任意相邻的两个 图像块之间的亮度关系时, 还需重新计算第一预设亮度阈值 的取值。 若 Lij+1— Lij > A L, 则 Lij <Lij+l, 即图像块 Li j+1的亮度大于图像块 Li j的亮 度; 若 I A L, 则 Lij=Lij+l, 即图像块 Lij+1的亮度等于图像块 Lij的亮度; 若 Lij+1— Lij <—A L,则 Lij >Lij+l,即图像块 Lij+1的亮度小于图像块 Lij的亮度。 此后, 再根据相邻图像块之间的亮度关系确定待检测图像中的暗区域与亮区域, 此 处同样以行的顺序为例。 对于列的顺序, 与行类似, 本实施例在此不再赘述。 本实施例在此以图 2中的第 4行前 9列的图像块为例。 表一示出了相应的 Lij与 Lij+1— Lij的取值。
Figure imgf000013_0001
在本实施例中,可以避免将亮度值很大的图像块确定为暗区域的第二预设亮度阈值
Ldark可以取 120, 可以避免将亮度值很小的图像块确定为亮区域的第三预设亮度阈值 Llight可以取 130。 在本实施例中, 在确定待检测图像中的暗区域与亮区域后, 可对暗 区域和亮区域进行标识, 对于亮暗过渡区域, 或无法确定亮暗区域的图像块, 不进行标 识。 在对待检测图像的暗区域与亮区域标识过程中, 将暗区域标识为白色圆点, 将亮区 域标识为黑色叉形。 根据表一可知, L42-L41=-68<-34, 则且 L42小于 Ldark, 则表明 L42为暗区域, 对图像块 L42进行标识。 对于图像块 L41, 由于其为第 4行第 1个图像块, 因此, 没有 可比较的图像块, 则图像块 L41亮暗无法确定, 因此, 对图像块 L41不进行标识。 以下为确定各图像块为暗区域获取亮区域的简单示意, 对于具体的标识, 可参见图 3A所示, 图 3A为本发明逆光检测方法实施例二的沿行方向亮暗区域示意图一。
L43-L42 | =18<34, 则 L43与 L42的亮度相同, L43为暗区域; L44-L43=61 >34, 且 L44大于 Llight, 则 L44为亮区域; L45-L44=-101 <-34, 且 L45小于 Ldark, 则 L45为暗区域;
L46-L45 | =5<34, 则 L46与 L45的亮度相同, L46为暗区域;
L47-L46 | =13<34, 则 L47与 L46的亮度相同, L47为暗区域; L48-L47=81 >34, 但 L48小于 Llight, 则 L48为亮暗过渡区域; L49-L48=121 >34, 且 L49大于 Ll ight , 则 L49为亮区域。 图 3B为本发明逆光检测方法实施例二的沿行方向亮暗区域示意图二。 图 3B沿行方 向与图 3A沿行方向相反。 图 3C为本发明逆光检测方法实施例二的沿列方向亮暗区域示 意图一, 即沿列从上至下的顺序。 图 3D为本发明逆光检测方法实施例二的沿列方向亮 暗区域示意图二, 即沿列从下至上的顺序。 其中, 图 3B至图 3D的获取方法, 与图 3A 类似, 本实施例在此不再赘述。 在得到图 3A至图 3D之后, 对图 3A至图 3D进行融合, 融合成图 4, 得到待检测图 像的最终暗区域和亮区域。 图 4为本发明逆光检测方法实施例二的亮暗区域示意图一。 在具体实现过程中, 首先根据图 3A和图 3B确定沿行方向的待检测图像的暗区域和亮区 域, 根据图 3C和图 3D确定沿列方向的待检测图像的暗区域和亮区域。 再根据图 3A和 图 3B的确定结果以及图 3C和图 3D的确定结果确定最终待检测图像的暗区域和亮区域。 优选地,根据图 3A和图 3B,或图 3C和图 3D或图 3A和图 3B的确定结果以及图 3C和图 3D的确定结果确定待检测图像的暗区域和亮区域的确定原则为: 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 该图像块确定为亮区域或暗区域; 或者, 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则该图像块确定为亮暗过渡区域; 或者, 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则该图像块确定为暗区域或亮区域。 根据上述原则确定待检测图像的亮区域和暗区域, 能够综和考虑所有图像块的亮暗 关系, 并对待检测图像的亮区域和暗区域进行修正, 以提高逆光检测的准确性。 根据上述方法得到的图 4为本发明实施例待检测图像的最终暗区域与亮区域。可根 据图 4所示的暗区域和亮区域确定待检测图像是否为逆光场景。当待检测图像中的暗区 域和亮区域满足预设关系时, 则确定待检测图像为逆光场景。 其中, 亮暗预设关系满足 如下中的任一条件或多种条件的组合: 暗区域的总面积占待检测图像的面积的比例大于 预设比值; 暗区域的亮度值平均值小于预设平均值; 亮区域的亮度值平均值与暗区域的 亮度值平均值的差值大于预设差值。 在确定待检测图像为逆光场景之后, 对待检测图像进行验证, 排除将黑色物体误认 为是逆光场景的情况。 在具体实现过程中, 若待检测图像中的暗区域所包含的像素点的 平均梯度大于预设梯度阈值, 则确定待检测图像为逆光场景, 若待检测图像中的暗区域 所包含的像素点的平均梯度小于预设梯度阈值, 则确定被摄主体为黑色物体, 该待检测 图像为非逆光场景。其中, 预设梯度阈值的具体取值, 可根据经验设定。在本实施例中, 设定预设梯度阈值为 4, 本实施例图 4中的暗区域所包含的像素点的平均梯度为 6. 13, 大于预设梯度阈值, 则说明待检测图像中的被摄主体纹理丰富, 不是黑色物体, 该待检 测图像为逆光场景。 图 5A为本发明逆光检测方法实施例二的非逆光场景示意图一, 图 5B为本发明逆光 检测方法实施例二的非逆光场景示意图二。 如图 5A所示, 待检测图像的被摄主体为黑 色物体, 图 5B中示出待检测图像的亮区域和暗区域, 计算图 5B中的暗区域所包含的像 素点的平均梯度为 0. 988, 小于预设梯度阈值, 则说明待检测图像的被摄主体纹理不丰 富, 为黑色物体, 根据图 5B得出的结论, 与图 5A所示一致。 本发明实施例通过对逆光场景进行验证,通过计算暗区域所包含的像素点的平均梯 度, 能够排除将有较大面积黑色物体的场景错误的识别为逆光场景的情况, 提高逆光检 测的准确性。 综上, 本发明实施例提供的逆光检测方法, 通过根据相邻图像块之间的亮度关系确 定待检测图像中的暗区域和亮区域, 不需要设定精确的判断暗区域的亮度阈值, 能够提 高拍摄主体的逆光程度发生较大变化时的逆光场景检测的准确性; 通过根据暗区域和亮 区域确定待检测图像是否为逆光场景, 能够提高拍摄主体的位置、 面积、 形状等发生较 大变化时逆光场景的准确性; 通过对逆光场景的验证, 能够排除将有较大面积黑色物体 的场景错误的识别为逆光场景的情况, 提高逆光检测的准确性。 图 6为本发明逆光检测设备实施一的结构示意图。本实施例提供的逆光检测设备可 以为包括成像镜头的手机、 照相机、 摄像机、 电脑等终端设备。 其中, 通过镜头可以获 取待检测图像, 并通过本发明实施例提供的图 6所示的逆光检测设备中的各模块对检测 待检测图像是否为逆光场景。 如图 6所示, 本发明实施例提供的逆光检测设备包括获取 模块 601、 确定模块 602、 处理模块 603。 其中, 获取模块 601用于获取待检测图像中的各图像块的亮度值; 确定模块 602用于根据所述各图像块的亮度值,确定相邻所述图像块之间的亮度关 系; 处理模块 603用于根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗 区域和亮区域, 并根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。 本实施例的逆光检测设备, 可以用于执行图 1所示方法实施例的技术方案, 其实现 原理和技术效果类似, 此处不再赘述。 进一步地, 所述待检测图像中包括 M行 N列图像块, M和 N均为正整数; 所述确定 模块 602具体用于- 根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每行中任意相邻的两个图像块之间的亮度关 系; 和 /或, 根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮度关 系。 进一步地, 所述确定模块 602具体用于: 若所述任意相邻的两个图像块的亮度值差值大于所述第一预设亮度阈值, 则确定所 述任意相邻的两个图像块中的一个图像块的亮度大于另一个图像块的亮度; 或者, 若所 述任意相邻的两个图像块的亮度值差值的绝对值小于或等于第一预设亮度阈值, 则确定 所述任意相邻的两个图像块的亮度相同。 进一步地,所述第一预设亮度阈值为所述待检测图像中所有所述任意相邻的两个图 像块的亮度值差值绝对值的平均值。 进一步地, 所述处理模块 603具体用于: 若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值小于第二预设亮度阈值, 则确定其中亮度小的图像块为暗 区域; 若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值大于等于所述第二预设亮度阈值, 则确定其中亮度小的图 像块为亮暗过渡区域; 若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值大于第三预设亮度阈值, 则确定其中亮度大的图像块为亮 区域; 若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值小于等于所述第三预设亮度阈值, 则确定其中亮度大的图 像块为亮暗过渡区域。 进一步地, 所述确定模块 602还用于: 沿所述待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获取每行中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每行从最后一个图像块至第一 个图像块的顺序, 获取每行中任意相邻的两个图像块的亮度值差值; 沿所述待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获取每列中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每列从最后一个图像块至第一 个图像块的顺序, 获取每列中任意相邻的两个图像块的亮度值差值。 进一步地, 所述处理模块 603还用于: 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 确定所述图像块为亮区域或暗区域; 或者, 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则确定所述图像块为亮暗过渡区域; 或者, 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则确定所述图像块为暗区域或亮区域。 进一步地, 所述处理模块 603具体用于: 若所述待检测图像中的所述暗区域和亮区域满足亮暗预设关系, 则确定所述待检测 图像为逆光场景, 所述亮暗预设关系满足如下中的任一条件或多种条件的组合: 所述暗 区域的总面积占所述待检测图像的面积的比例大于预设比值; 所述暗区域的亮度值平均 值小于预设平均值; 所述亮区域的亮度值平均值与所述暗区域的亮度值平均值的差值大 于预设差值。 进一步地, 所述处理模块 603还用于: 对所述逆光场景进行验证,若所述待检测图像中的暗区域所包含的像素点的平均梯 度大于预设梯度阈值, 则确定所述待检测图像为逆光场景, 若所述待检测图像中的暗区 域所包含的象素点的平均梯度小于预设梯度阈值, 则确定所述待检测图像为非逆光场
本实施例的逆光检测设备, 可以用于执行图 2所示方法实施例的技术方案, 其实现 原理和技术效果类似, 此处不再赘述。 图 7为本发明逆光检测设备实施二的结构示意图。本实施例提供的逆光检测设备可 以为包括成像镜头的手机、 照相机、 摄像机、 电脑等终端设备。 通过镜头可以获取待检 测图像, 并通过本发明实施例提供的图 7所示的逆光检测设备中的各部件来检测待检测 图像是否为逆光场景。 如图 7所示, 本实施例提供的逆光检测设备 70包括处理器 701 和存储器 702。 逆光检测设备 70还包括镜头 703, 镜头 703可以和处理器 701相连。 其 中, 处理器 701可以通过镜头 703获取待检测图像, 并将待检测图像划分为图像块, 并 计算得到各图像块的亮度值。 存储器 702存储执行指令, 当逆光检测设备 70运行时, 处理器 701与存储器 702之间通信, 处理器 701调用存储器 702中的执行指令, 用于执 行以下操作- 获取待检测图像中的各图像块的亮度值; 根据所述各图像块的亮度值, 确定相邻所述图像块之间的亮度关系; 根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区域和亮区域, 并 根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。 进一步地, 所述待检测图像中包括 M行 N列图像块, M和 N均为正整数; 所述根据 所述各图像块的亮度值, 确定相邻所述图像块之间的亮度关系, 包括- 根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每行中任意相邻的两个图像块之间的亮度关 系; 和 /或, 根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮度关 系。 进一步地,所述根据所述任意相邻的两个图像块的亮度值差值与第一预设亮度阈值 的大小关系, 确定所述任意相邻的两个图像块之间的亮度关系, 包括: 若所述任意相邻的两个图像块的亮度值差值大于所述第一预设亮度阈值, 则所述任 意相邻的两个图像块中的一个图像块的亮度大于另一个图像块的亮度; 或者, 若所述任 意相邻的两个图像块的亮度值差值的绝对值小于或等于第一预设亮度阈值, 则所述任意 相邻的两个图像块的亮度相同。 进一步地,所述第一预设亮度阈值为所述待检测图像中所有所述任意相邻的两个图 像块的亮度值差值绝对值的平均值。 进一步地,所述根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区 域和亮区域, 包括: 若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值小于第二预设亮度阈值, 则其中亮度小的图像块为暗区 域; 若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值大于等于所述第二预设亮度阈值, 则其中亮度小的图像块 为亮暗过渡区域; 若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值大于第三预设亮度阈值, 则其中亮度大的图像块为亮区 域; 若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值小于等于所述第三预设亮度阈值, 则其中亮度大的图像块 为亮暗过渡区域。 进一步地,所述根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值 与第一预设亮度阈值的大小关系之前, 还包括- 沿所述待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获取每行中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每行从最后一个图像块至第一 个图像块的顺序, 获取每行中任意相邻的两个图像块的亮度值差值; 所述根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预 设亮度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮 度关系之前, 还包括: 沿所述待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获取每列中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每列从最后一个图像块至第一 个图像块的顺序, 获取每列中任意相邻的两个图像块的亮度值差值。 进一步地,所述根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区 域和亮区域, 还包括: 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 所述图像块确定为亮区域或暗区域; 或者, 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则所述图像块确定为亮暗过渡区域; 或者, 对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则所述图像块确定为暗区域或亮区域。 进一步地, 所述根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景, 包 括- 若所述待检测图像中的所述暗区域和亮区域满足亮暗预设关系, 则确定所述待检测 图像为逆光场景, 所述亮暗预设关系满足如下中的任一条件或多种条件的组合: 所述暗 区域的总面积占所述待检测图像的面积的比例大于预设比值; 所述暗区域的亮度值平均 值小于预设平均值; 所述亮区域的亮度值平均值与所述暗区域的亮度值平均值的差值大 于预设差值。 进一步地, 所述根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景之 后, 还包括: 对所述逆光场景进行验证,若所述待检测图像中的暗区域所包含的像素点的平均梯 度大于预设梯度阈值, 则确定所述待检测图像为逆光场景, 若所述待检测图像中的暗区 域所包含的象素点的平均梯度小于预设梯度阈值, 则确定所述待检测图像为非逆光场
本实施例的逆光检测设备, 可以用于执行上述方法实施例的技术方案, 其实现原理 和技术效果类似, 此处不再赘述。
在本申请所提供的几个实施例中, 应该理解到, 所揭露的设备和方法, 可以通过其 它的方式实现。 例如, 以上所描述的设备实施例仅仅是示意性的, 例如, 所述单元或模 块的划分, 仅仅为一种逻辑功能划分, 实际实现时可以有另外的划分方式, 例如多个单 元或模块可以结合或者可以集成到另一个系统, 或一些特征可以忽略, 或不执行。 另一 点, 所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口, 设备 或模块的间接耦合或通信连接, 可以是电性, 机械或其它的形式。 所述作为分离部件说明的模块可以是或者也可以不是物理上分开的, 作为模块显示 的部件可以是或者也可以不是物理模块, 即可以位于一个地方, 或者也可以分布到多个 网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的 目的。 本领域普通技术人员可以理解: 实现上述各方法实施例的全部或部分步骤可以通过 程序指令相关的硬件来完成。 前述的程序可以存储于一计算机可读取存储介质中。 该程 序在执行时, 执行包括上述各方法实施例的步骤; 而前述的存储介质包括: 醒、 醒、 磁碟或者光盘等各种可以存储程序代码的介质。 最后应说明的是: 以上各实施例仅用以说明本发明的技术方案, 而非对其限制; 尽 管参照前述各实施例对本发明进行了详细的说明, 本领域的普通技术人员应当理解: 其 依然可以对前述各实施例所记载的技术方案进行修改, 或者对其中部分或者全部技术特 征进行等同替换; 而这些修改或者替换, 并不使相应技术方案的本质脱离本发明各实施 例技术方案的范围。

Claims

权利要求
1、 一种逆光检测方法, 其特征在于, 包括:
获取待检测图像中的各图像块的亮度值;
根据所述各图像块的亮度值, 确定相邻所述图像块之间的亮度关系;
根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区域和亮区域, 并 根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。
2、 根据权利要求 1所述的方法, 其特征在于, 所述待检测图像中包括 M行 N列图 像块, M和 N均为正整数; 所述根据所述各图像块的亮度值, 确定相邻所述图像块之间 的亮度关系, 包括:
根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每行中任意相邻的两个图像块之间的亮度关 系; 和 /或,
根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮度关 系。
3、 根据权利要求 2所述的方法, 其特征在于, 所述根据所述任意相邻的两个图像 块的亮度值差值与第一预设亮度阈值的大小关系,确定所述任意相邻的两个图像块之间 的亮度关系, 包括:
若所述任意相邻的两个图像块的亮度值差值大于所述第一预设亮度阈值, 则所述任 意相邻的两个图像块中的一个图像块的亮度大于另一个图像块的亮度; 或者, 若所述任 意相邻的两个图像块的亮度值差值的绝对值小于或等于第一预设亮度阈值, 则所述任意 相邻的两个图像块的亮度相同。
4、 根据权利要求 2或 3所述的方法, 其特征在于, 所述第一预设亮度阈值为所述 待检测图像中所有所述任意相邻的两个图像块的亮度值差值绝对值的平均值。
5、 根据权利要求 2-4任一项所述的方法, 其特征在于, 所述根据相邻所述图像块 之间的亮度关系确定所述待检测图像中的暗区域和亮区域, 包括:
若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值小于第二预设亮度阈值, 则其中亮度小的图像块为暗区 域;
若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值大于等于所述第二预设亮度阈值, 则其中亮度小的图像块 为亮暗过渡区域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值大于第三预设亮度阈值, 则其中亮度大的图像块为亮区 域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值小于等于所述第三预设亮度阈值, 则其中亮度大的图像块 为亮暗过渡区域。
6、 根据权利要求 2-5任一项所述的方法, 其特征在于, 所述根据所述待检测图像 的每行中任意相邻的两个图像块的亮度值差值与第一预设亮度阈值的大小关系之前,还 包括:
沿所述待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获取每行中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每行从最后一个图像块至第一 个图像块的顺序, 获取每行中任意相邻的两个图像块的亮度值差值;
所述根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预 设亮度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮 度关系之前, 还包括:
沿所述待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获取每列中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每列从最后一个图像块至第一 个图像块的顺序, 获取每列中任意相邻的两个图像块的亮度值差值。
7、 根据权利要求 6所述的方法, 其特征在于, 所述根据相邻所述图像块之间的亮 度关系确定所述待检测图像中的暗区域和亮区域, 还包括:
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 所述图像块确定为亮区域或暗区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则所述图像块确定为亮暗过渡区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则所述图像块确定为暗区域或亮区域。
8、 根据权利要求 6所述的方法, 其特征在于, 所述根据所述暗区域和亮区域确定 所述待检测图像是否为逆光场景, 包括- 若所述待检测图像中的所述暗区域和亮区域满足亮暗预设关系, 则确定所述待检测 图像为逆光场景, 所述亮暗预设关系满足如下中的任一条件或多种条件的组合: 所述暗 区域的总面积占所述待检测图像的面积的比例大于预设比值; 所述暗区域的亮度值平均 值小于预设平均值; 所述亮区域的亮度值平均值与所述暗区域的亮度值平均值的差值大 于预设差值。
9、 根据权利要求 1-8任一项所述的方法, 其特征在于, 所述根据所述暗区域和亮 区域确定所述待检测图像是否为逆光场景之后, 还包括:
对所述逆光场景进行验证,若所述待检测图像中的暗区域所包含的像素点的平均梯 度大于预设梯度阈值, 则确定所述待检测图像为逆光场景, 若所述待检测图像中的暗区 域所包含的象素点的平均梯度小于预设梯度阈值, 则确定所述待检测图像为非逆光场 牙、。
10、 一种逆光检测设备, 其特征在于, 包括:
获取模块: 用于获取待检测图像中的各图像块的亮度值;
确定模块:用于根据所述各图像块的亮度值,确定相邻所述图像块之间的亮度关系; 处理模块: 用于根据相邻所述图像块之间的亮度关系确定所述待检测图像中的暗区 域和亮区域, 并根据所述暗区域和亮区域确定所述待检测图像是否为逆光场景。
11、 根据权利要求 10所述的设备, 其特征在于, 所述待检测图像中包括 M行 N列 图像块, M和 N均为正整数; 所述确定模块具体用于:
根据所述待检测图像的每行中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每行中任意相邻的两个图像块之间的亮度关 系; 和 /或,
根据所述待检测图像的每列中任意相邻的两个图像块的亮度值差值与第一预设亮 度阈值的大小关系,确定所述待检测图像的每列中任意相邻的两个图像块之间的亮度关 系。
12、 根据权利要求 11所述的设备, 其特征在于, 所述确定模块具体用于: 若所述任意相邻的两个图像块的亮度值差值大于所述第一预设亮度阈值, 则确定所 述任意相邻的两个图像块中的一个图像块的亮度大于另一个图像块的亮度; 或者, 若所 述任意相邻的两个图像块的亮度值差值的绝对值小于或等于第一预设亮度阈值, 则确定 所述任意相邻的两个图像块的亮度相同。
13、 根据权利要求 11或 12所述的设备, 其特征在于, 所述第一预设亮度阈值为所 述待检测图像中所有所述任意相邻的两个图像块的亮度值差值绝对值的平均值。
14、根据权利要求 11-13任一项所述的设备,其特征在于,所述处理模块具体用于: 若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值小于第二预设亮度阈值, 则确定其中亮度小的图像块为暗 区域;
若所述任意相邻的两个图像块中一个图像块的亮度小于另一个图像块的亮度, 且其 中亮度值小的图像块的亮度值大于等于所述第二预设亮度阈值, 则确定其中亮度小的图 像块为亮暗过渡区域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值大于第三预设亮度阈值, 则确定其中亮度大的图像块为亮 区域;
若所述任意相邻的两个图像块中一个图像块的亮度大于另一个图像块的亮度, 且其 中亮度值大的图像块的亮度值小于等于所述第三预设亮度阈值, 则确定其中亮度大的图 像块为亮暗过渡区域。
15、 根据权利要求 11-14任一项所述的设备, 其特征在于, 所述确定模块还用于: 沿所述待检测图像的每行从第一个图像块至最后一个图像块的顺序, 获取每行中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每行从最后一个图像块至第一 个图像块的顺序, 获取每行中任意相邻的两个图像块的亮度值差值;
沿所述待检测图像的每列从第一个图像块至最后一个图像块的顺序, 获取每列中任 意相邻的两个图像块的亮度值差值,沿所述待检测图像的每列从最后一个图像块至第一 个图像块的顺序, 获取每列中任意相邻的两个图像块的亮度值差值。
16、 根据权利要求 15所述的设备, 其特征在于, 所述处理模块还用于:
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序均确定为亮区域或暗区域, 则 确定所述图像块为亮区域或暗区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为亮区域, 另一个顺序确定为暗区域, 则确定所述图像块为亮暗过渡区域; 或者,
对于每行或每列中的任意一个图像块,若沿每行或每列的第一个图像块至最后一个 图像块的顺序和沿最后一个图像块至第一个图像块的顺序中的一个顺序确定为暗区域 或亮区域, 另一个顺序确定为亮暗过渡区域, 则确定所述图像块为暗区域或亮区域。
17、 根据权利要求 15所述的设备, 其特征在于, 所述处理模块具体用于: 若所述待检测图像中的所述暗区域和亮区域满足亮暗预设关系, 则确定所述待检测 图像为逆光场景, 所述亮暗预设关系满足如下中的任一条件或多种条件的组合: 所述暗 区域的总面积占所述待检测图像的面积的比例大于预设比值; 所述暗区域的亮度值平均 值小于预设平均值; 所述亮区域的亮度值平均值与所述暗区域的亮度值平均值的差值大 于预设差值。
18、 根据权利要求 10-17任一项所述的设备, 其特征在于, 所述处理模块还用于: 对所述逆光场景进行验证,若所述待检测图像中的暗区域所包含的像素点的平均梯 度大于预设梯度阈值, 则确定所述待检测图像为逆光场景, 若所述待检测图像中的暗区 域所包含的象素点的平均梯度小于预设梯度阈值, 则确定所述待检测图像为非逆光场 牙、。
19、 一种逆光检测设备, 其特征在于, 包括: 处理器和存储器, 所述存储器存储执 行指令, 当所述逆光检测设备运行时, 所述处理器与所述存储器之间通信, 所述处理器 执行所述执行指令使得所述逆光检测设备执行如权利要求 1至 9任一项所述的方法。
PCT/CN2013/090806 2013-01-07 2013-12-28 逆光检测方法及设备 WO2014106445A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2015527788A JP5988119B2 (ja) 2013-01-07 2013-12-28 逆光検出方法及びデバイス
EP13870104.0A EP2849431B1 (en) 2013-01-07 2013-12-28 Method and apparatus for detecting backlight
US14/566,020 US9390475B2 (en) 2013-01-07 2014-12-10 Backlight detection method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310005069.9A CN103916603B (zh) 2013-01-07 2013-01-07 逆光检测方法及设备
CN201310005069.9 2013-01-07

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/566,020 Continuation US9390475B2 (en) 2013-01-07 2014-12-10 Backlight detection method and device

Publications (1)

Publication Number Publication Date
WO2014106445A1 true WO2014106445A1 (zh) 2014-07-10

Family

ID=51041961

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/090806 WO2014106445A1 (zh) 2013-01-07 2013-12-28 逆光检测方法及设备

Country Status (5)

Country Link
US (1) US9390475B2 (zh)
EP (1) EP2849431B1 (zh)
JP (1) JP5988119B2 (zh)
CN (1) CN103916603B (zh)
WO (1) WO2014106445A1 (zh)

Families Citing this family (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104700426B (zh) * 2015-04-02 2017-11-03 厦门美图之家科技有限公司 一种判断图像是否偏暗或偏亮的方法和系统
CN105791709B (zh) * 2015-12-29 2019-01-25 福建星网锐捷通讯股份有限公司 带有逆光补偿的自动曝光处理方法及装置
CN105467604B (zh) * 2016-02-16 2018-01-12 京东方科技集团股份有限公司 一种3d显示装置及其驱动方法
CN105959773B (zh) * 2016-04-29 2019-06-18 魔方天空科技(北京)有限公司 多媒体文件的处理方法和装置
CN105959585B (zh) * 2016-05-12 2019-08-16 南昌黑鲨科技有限公司 多级逆光检测方法及装置
CN106067177B (zh) * 2016-06-15 2020-06-26 深圳市万普拉斯科技有限公司 Hdr场景侦测方法和装置
CN106125767B (zh) * 2016-08-31 2020-03-17 北京小米移动软件有限公司 飞行器的控制方法、装置及飞行器
US10902244B2 (en) 2017-03-27 2021-01-26 Samsung Electronics Co., Ltd. Apparatus and method for image processing
CN107295270B (zh) * 2017-07-31 2020-01-24 爱博康电子(深圳)有限公司 一种图像亮度值的确定方法、装置、终端和计算机可读存储介质
CN109788205B (zh) * 2017-11-15 2021-03-09 浙江宇视科技有限公司 图像调整方法、装置及电子设备
CN108564537B (zh) * 2017-12-29 2022-08-26 北京乐蜜科技有限责任公司 图像处理的方法、装置、电子设备及介质
CN109246399A (zh) * 2018-08-17 2019-01-18 优视科技新加坡有限公司 一种识别逆光区域方法、装置和设备/终端/服务器
CN109543581A (zh) * 2018-11-15 2019-03-29 北京旷视科技有限公司 图像处理方法、图像处理装置以及非易失性存储介质
CN110189312B (zh) * 2019-05-24 2021-06-29 北京康夫子健康技术有限公司 眼底图像的亮度评估方法、装置、电子设备及存储介质
CN111142528B (zh) * 2019-12-31 2023-10-24 天津职业技术师范大学(中国职业培训指导教师进修中心) 车用危险场景感知方法、装置和系统
CN112261306B (zh) * 2020-09-27 2021-10-08 厦门亿联网络技术股份有限公司 一种逆光判断方法、装置、设备及存储介质
CN112291548B (zh) * 2020-10-28 2023-01-31 Oppo广东移动通信有限公司 白平衡统计方法、装置、移动终端及存储介质
CN112526534B (zh) * 2020-11-03 2024-03-08 上海炬佑智能科技有限公司 ToF传感装置及其距离检测方法
CN112543286A (zh) * 2020-11-27 2021-03-23 展讯通信(上海)有限公司 一种用于终端的图像生成方法及装置、存储介质、终端
CN112733827B (zh) * 2020-12-29 2023-08-04 深圳市天视通视觉有限公司 一种车牌识别方法、装置、电子设备及存储介质
CN115086566B (zh) * 2021-03-16 2024-03-29 广州视源电子科技股份有限公司 图片场景检测方法、装置

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101008762A (zh) * 2007-01-30 2007-08-01 北京中星微电子有限公司 一种逆光检测方法、逆光补偿终止检测方法及相应装置
CN102006422A (zh) * 2009-09-01 2011-04-06 华晶科技股份有限公司 一种逆光拍摄方法

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3412174B2 (ja) * 1992-05-21 2003-06-03 松下電器産業株式会社 自動露光制御装置
JP4057147B2 (ja) * 1998-06-16 2008-03-05 コニカミノルタビジネステクノロジーズ株式会社 逆光シーン判定方法、逆光シーン判定方法プログラムを記憶したコンピュータにより読み取り可能な記憶媒体及び逆光シーン判定機能を有する画像処理装置
JP3615462B2 (ja) * 2000-05-12 2005-02-02 三洋電機株式会社 自動露光調節カメラ
JP2002196389A (ja) * 2000-12-26 2002-07-12 Canon Inc 測光装置およびカメラ
US6853806B2 (en) * 2002-09-13 2005-02-08 Olympus Optical Co., Ltd. Camera with an exposure control function
TWI285047B (en) * 2005-11-24 2007-08-01 Sunplus Technology Co Ltd Method of automatic exposure control and automatic exposure compensated apparatus
JP2007166028A (ja) * 2005-12-09 2007-06-28 Matsushita Electric Ind Co Ltd 露出量制御装置および露出量制御方法
JP5319078B2 (ja) * 2006-07-07 2013-10-16 オリンパスイメージング株式会社 カメラ、カメラの画像処理方法、プログラム、記録媒体
CN101282425B (zh) * 2008-04-30 2010-06-23 北京中星微电子有限公司 一种背光补偿的方法和装置
JP5157753B2 (ja) * 2008-08-27 2013-03-06 カシオ計算機株式会社 画像処理装置、画像処理方法、画像処理プログラム
JP5138510B2 (ja) * 2008-08-28 2013-02-06 エイオーエフ イメージング テクノロジー リミテッド 画像データ処理装置
TWI387332B (zh) * 2009-08-26 2013-02-21 Altek Corp A method of backlighting
CN102196182A (zh) * 2010-03-09 2011-09-21 株式会社理光 逆光检测设备及逆光检测方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101008762A (zh) * 2007-01-30 2007-08-01 北京中星微电子有限公司 一种逆光检测方法、逆光补偿终止检测方法及相应装置
CN102006422A (zh) * 2009-09-01 2011-04-06 华晶科技股份有限公司 一种逆光拍摄方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2849431A4 *

Also Published As

Publication number Publication date
US9390475B2 (en) 2016-07-12
EP2849431A4 (en) 2015-08-19
JP5988119B2 (ja) 2016-09-07
JP2015534301A (ja) 2015-11-26
US20150093040A1 (en) 2015-04-02
EP2849431A1 (en) 2015-03-18
EP2849431B1 (en) 2016-10-12
CN103916603A (zh) 2014-07-09
CN103916603B (zh) 2019-01-15

Similar Documents

Publication Publication Date Title
WO2014106445A1 (zh) 逆光检测方法及设备
JP6615917B2 (ja) 実時間ビデオエンハンスメント方法、端末及び非一時的コンピュータ可読記憶媒体
KR102149187B1 (ko) 전자 장치와, 그의 제어 방법
US10620826B2 (en) Object selection based on region of interest fusion
WO2017107700A1 (zh) 一种实现图像配准的方法及终端
WO2019100675A1 (zh) 一种分区背光控制方法及其装置
WO2018054054A1 (zh) 一种人脸识别的方法、装置、移动终端及计算机存储介质
CN105243371A (zh) 一种人脸美颜程度的检测方法、系统及拍摄终端
WO2019042419A1 (zh) 图像跟踪点获取方法、设备及存储介质
CN106981078B (zh) 视线校正方法、装置、智能会议终端及存储介质
US9666161B2 (en) Method and system for replacing theme of a mobile terminal on the basis of a user&#39;s clothes color
CN106412458A (zh) 一种图像处理方法和装置
CN108009997B (zh) 一种调整图像对比度的方法及装置
US9613404B2 (en) Image processing method, image processing apparatus and electronic device
US8693783B2 (en) Processing method for image interpolation
WO2016197639A1 (zh) 屏幕画面显示方法及装置
US20220329729A1 (en) Photographing method, storage medium and electronic device
CN111311481A (zh) 背景虚化方法、装置、终端设备及存储介质
US20150320311A1 (en) Method and apparatus for iris recognition using natural light
EP3076386A1 (en) Image display method and user terminal
CN108564537B (zh) 图像处理的方法、装置、电子设备及介质
WO2019196240A1 (zh) 拍摄照片的方法、装置、计算机设备及存储介质
CN105893578B (zh) 一种照片选择的方法及装置
CN111885371A (zh) 图像遮挡检测方法、装置、电子设备和计算机可读介质
US20150215602A1 (en) Method for ajdusting stereo image and image processing device using the same

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: 13870104

Country of ref document: EP

Kind code of ref document: A1

REEP Request for entry into the european phase

Ref document number: 2013870104

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2013870104

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2015527788

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE