WO2015100913A1 - 图像缩略图的生成方法、装置和终端 - Google Patents

图像缩略图的生成方法、装置和终端 Download PDF

Info

Publication number
WO2015100913A1
WO2015100913A1 PCT/CN2014/077610 CN2014077610W WO2015100913A1 WO 2015100913 A1 WO2015100913 A1 WO 2015100913A1 CN 2014077610 W CN2014077610 W CN 2014077610W WO 2015100913 A1 WO2015100913 A1 WO 2015100913A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
rectangular frame
pixel
value
information
Prior art date
Application number
PCT/CN2014/077610
Other languages
English (en)
French (fr)
Inventor
王琳
张祺深
张波
Original Assignee
小米科技有限责任公司
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 小米科技有限责任公司 filed Critical 小米科技有限责任公司
Priority to KR1020147021528A priority Critical patent/KR20150088970A/ko
Priority to MX2014009069A priority patent/MX347781B/es
Priority to RU2015123209A priority patent/RU2615679C2/ru
Priority to JP2015555587A priority patent/JP2016511875A/ja
Priority to BR112014018578A priority patent/BR112014018578A8/pt
Priority to US14/479,531 priority patent/US9727972B2/en
Publication of WO2015100913A1 publication Critical patent/WO2015100913A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00Geometric image transformations in the plane of the image
    • G06T3/04Context-preserving transformations, e.g. by using an importance map
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/13Edge detection
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/85Assembly of content; Generation of multimedia applications
    • H04N21/854Content authoring
    • H04N21/8549Creating video summaries, e.g. movie trailer
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/85Assembly of content; Generation of multimedia applications
    • H04N21/854Content authoring
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/387Composing, repositioning or otherwise geometrically modifying originals
    • H04N1/3872Repositioning or masking
    • H04N1/3873Repositioning or masking defined only by a limited number of coordinate points or parameters, e.g. corners, centre; for trimming
    • H04N1/3875Repositioning or masking defined only by a limited number of coordinate points or parameters, e.g. corners, centre; for trimming combined with enlarging or reducing

Definitions

  • the present disclosure relates to the field of communications, and in particular, to a method, an apparatus, and a terminal for generating an image thumbnail. Background technique
  • a personal album will generate a thumbnail for the stored image and place the thumbnail in preview mode for the user to preview and review.
  • a more common image thumbnail generation method is to directly intercept the middle portion of the image, and perform the corresponding scaling of the intercepted intermediate portion to obtain a thumbnail of the image. The advantage of this method is that it is very efficient.
  • the above method only considers the spatial position information of the image, and does not consider the content information of the image at all, which may cause the generated thumbnail to sometimes not fully express the content information of the original image. For example, when the position of a person in an image is relatively biased, if only the middle portion of the image is clipped, only a part of the character may be included in the generated thumbnail, resulting in poor accuracy in expressing the original image content information.
  • the present disclosure provides a method and apparatus for generating image thumbnails to improve the accuracy of thumbnail representation of original image content information.
  • the technical solution is as follows:
  • a method for generating an image thumbnail including:
  • a rectangular frame having the largest distribution value of the information is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • the rectangular frame at each position of the sliding search calculates the information amount distribution value of the rectangular frame according to the edge intensity value of the inner pixel point, including:
  • the edge intensity values of all the pixels in the same are summed to obtain the information distribution value of the rectangular frame.
  • the sliding frame is searched on the image by a preset rectangular frame, and the rectangular frame at each position of the sliding search is calculated according to the edge intensity value of the inner pixel point, including the information distribution value of the rectangular frame, including :
  • the information quantity distribution value of each pixel in the rectangular frame is added to obtain an information quantity distribution value of the rectangular frame, including:
  • the information quantity distribution value of each pixel in the rectangular frame is multiplied by the corresponding weight value, and then summed to obtain an information quantity distribution value of the rectangular frame; wherein the kernel function is closer to the center point of the image The weight value taken by the pixel is larger.
  • the calculating by using the attention model established in advance according to the center point of the image and the coordinates of each pixel point, calculating a spatial position attention value of each pixel in the image, including:
  • the calculating the information quantity distribution value of each pixel in the image by using the information quantity distribution model established in advance according to the edge intensity value and the spatial position attention value includes:
  • the information amount distribution value of each pixel in the image is calculated by using the following information quantity distribution model:
  • (i, j) represents any pixel in the image
  • /(i, j) represents the information distribution value of the pixel
  • £(i, j) represents the edge intensity value of the pixel
  • P (i, j) represents the spatial position attention value of the pixel.
  • the rectangular frame is a square, and the side length is equal to the length of the short side of the image.
  • the image is filtered to obtain an edge intensity value of each pixel in the image, including:
  • the image content is truncated to obtain a thumbnail of the original image.
  • an apparatus for generating an image thumbnail including:
  • a filtering module configured to filter an image to obtain an edge intensity value of each pixel in the image
  • a search module configured to slide a search on the image by a preset rectangular frame, and calculate a distribution amount of the information of the rectangular frame according to an edge intensity value of the pixel point in the rectangular frame at each position of the sliding search;
  • the intercepting module is configured to select a rectangular frame with the largest distribution value of the information, and intercept the selected image content of the rectangular frame to obtain a thumbnail of the image.
  • the search module includes:
  • a search unit configured to slide a search on the image with a preset rectangular frame
  • the calculating unit is configured to sum the edge intensity values of all the pixels in the sliding position of the sliding search to obtain the information quantity distribution value of the rectangular frame.
  • the search module includes:
  • a search unit configured to slide a search on the image with a preset rectangular frame
  • a calculation unit configured to calculate a spatial position attention value of each pixel in the image by using a attention model established in advance according to a center point of the image and coordinates of each pixel; using a pre-determined value according to the edge intensity value and the spatial position Establishing an information quantity distribution model, calculating an information quantity distribution value of each pixel in the image; and sliding a search on the rectangular frame at each position of the search unit, and distributing the information quantity distribution value of each pixel in the rectangular frame Add the information distribution value of the rectangular frame.
  • the computing unit includes:
  • An information quantity distribution value calculation sub-unit configured to calculate a weight value corresponding to each pixel point in the rectangular frame by using a pre-selected kernel function; and performing an information quantity distribution value of each pixel point in the rectangular frame and a corresponding weight value After the product is summed, the information distribution value of the rectangular frame is obtained;
  • the weight value of the pixel function closer to the pixel point from the center point of the image is larger.
  • the computing unit includes:
  • the computing unit includes:
  • the information quantity distribution value calculation sub-unit is configured to calculate the information quantity distribution value of each pixel in the image by using the following information quantity distribution model:
  • (i, j) represents any pixel in the image
  • /(i, j) represents the information distribution value of the pixel
  • £(i, j) represents the edge intensity value of the pixel
  • P (i, j) represents the spatial position attention value of the pixel.
  • the rectangular frame is a square, and the side length is equal to the length of the short side of the image.
  • the device further includes:
  • the filtering module is configured to: filter the compressed image of the compression module to obtain an edge intensity value of each pixel in the image;
  • the intercepting module is configured to: the selected rectangular frame is corresponding to a rectangular frame in the original image, and the image content in the rectangular frame in the original image is intercepted to obtain a thumbnail of the original image.
  • a terminal comprising a memory, and one or more programs, wherein one or more programs are stored in a memory and configured to execute the one or more by one or more processors
  • the above program contains instructions for doing the following:
  • a rectangular frame having the largest distribution value of the information is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • Some beneficial effects brought by the technical solutions provided by the present disclosure may include: obtaining an edge intensity value of each pixel in the image by filtering an image, sliding a search on the image with a preset rectangular frame, and sliding Searching the rectangular frame at each position, calculating the information quantity distribution value of the rectangular frame according to the edge intensity value of the pixel point therein, selecting the rectangular frame with the largest information distribution value, and selecting the image content corresponding to the selected rectangular frame
  • the thumbnail image of the image is obtained, and the thumbnail image is generated based on the image content information, which improves the accuracy of the thumbnail image representing the original image content information, and is more in line with people's cognitive habits.
  • FIG. 2 is an exemplary flowchart of a method for generating an image thumbnail provided by Embodiment 2 of the present disclosure
  • FIG. 3a is an exemplary flowchart of a method for generating an image thumbnail provided by Embodiment 3 of the present disclosure
  • FIG. 3b is an exemplary schematic diagram of a process of generating a thumbnail provided by Embodiment 3 of the present disclosure
  • FIG. 4a is an exemplary flowchart of a method for generating an image thumbnail provided by Embodiment 4 of the present disclosure
  • FIG. 4b is an exemplary schematic diagram of a kernel function provided by Embodiment 4 of the present disclosure.
  • FIG. 5a is one exemplary structural diagram of a device for generating an image thumbnail provided by Embodiment 5 of the present disclosure
  • FIG. 5b is a second structural diagram of an apparatus for generating an image thumbnail provided by Embodiment 5 of the present disclosure.
  • FIG. 6 is an exemplary structural diagram of a terminal provided in Embodiment 6 of the present disclosure.
  • this embodiment provides a method for generating an image thumbnail, and the method includes the following steps.
  • step 101 the image is filtered to obtain an edge intensity value for each pixel within the image.
  • filtering the image to obtain edge intensity values of each pixel in the image may include: using a Laplacian edge filter operator, a Sobel edge filter operator, a Robert edge operator, and a Prewitt edge operator.
  • the LOG edge operator filters the image to obtain the edge intensity value of each pixel in the image.
  • the pixel points whose edge intensity values are relatively close may be considered to have little difference in color; the pixel points having a large difference in edge intensity values may be considered to have a large difference in color, and therefore, the edge intensity value may be considered to be To some extent, it reflects the content information of the image.
  • step 102 the search is performed on the image by a preset rectangular frame, and the rectangular frame at each position of the sliding search is calculated based on the edge intensity value of the pixel in the rectangle.
  • the preset rectangular frame may be a rectangular frame of any size smaller than the image.
  • the short side of the rectangle is equal to the short side of the image
  • the long side of the rectangle is smaller than the long side of the image.
  • the short side of the rectangle is smaller than the short side of the image
  • the long side of the rectangle is equal to the long side of the image.
  • the short side of the rectangular frame is smaller than The short side of the image
  • the long side of the rectangular frame is smaller than the long side of the image, etc., which is not specifically limited in this embodiment.
  • the sliding search of the rectangular frame on the image may be a sliding search in any direction, which is not specifically limited in this embodiment, such as sliding the search only in the horizontal direction, or sliding the search only in the vertical direction, or along the 45° direction search.
  • step 103 a rectangular frame with the largest amount of information distribution is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • the size of the generated thumbnail is not limited, such as an image of 1600 x 1200 or the like.
  • the image that is truncated can also be compressed first, and then the compressed image is used as a thumbnail. This embodiment does not specifically limit this.
  • the rectangular frame at each position of the sliding search may calculate the information amount distribution value of the rectangular frame according to the edge intensity value of the pixel point, and may include:
  • the edge intensity values of all the pixels in the same are summed to obtain the information distribution value of the rectangular frame.
  • the above-mentioned rectangular frame is slid and searched on the image, and the rectangular frame at each position of the sliding search is calculated according to the edge intensity value of the inner pixel point.
  • the rectangular frame at each position of the sliding search is calculated according to the edge intensity value of the inner pixel point.
  • the sum of the information amount distribution values of the pixels in the rectangular frame is obtained by obtaining the information quantity distribution value of the rectangular frame, which may include:
  • the spatial position attention value of each pixel in the image is calculated by using the attention model established in advance according to the center point of the image and the coordinates of each pixel, and may include:
  • the information distribution value of each pixel in the rectangular frame is calculated by using the information quantity distribution model established in advance according to the edge intensity value and the spatial position attention value, and may include:
  • the information distribution value of each pixel in the rectangular frame is calculated by using the following information quantity distribution model:
  • (i, j) represents any pixel in the image
  • /(i, j) represents the information distribution value of the pixel
  • £(i, j) represents the edge intensity value of the pixel
  • P( i, j) represents the spatial position attention value of the pixel.
  • the rectangular frame may be a square, and the side length is equal to the length of the short side of the image.
  • the image may be subjected to compression processing before filtering to obtain an image with a smaller resolution, and then a subsequent step such as filtering is performed, after selecting a rectangular frame having the largest distribution value of the information amount, Then, the rectangular frame is converted into a position corresponding to the original image for interception.
  • the filtering the image to obtain the edge intensity value of each pixel in the image may include: compressing the original image, and filtering the compressed image to obtain an edge intensity value of each pixel in the image;
  • the intercepting the image content corresponding to the selected rectangular frame to obtain the thumbnail of the image comprises: matching the selected rectangular frame to a rectangular frame in the original image, and the original The image content within the rectangular frame in the image is truncated to obtain a thumbnail of the original image.
  • a 1600x 1200 image is first compressed into a 400x400 image, and then a rectangular frame is selected on the 400x400 image. After the selection is completed, the corresponding area of the rectangular frame is converted into a corresponding area on the 1600x 1200 image, and then Capture and compress to get thumbnails.
  • This method greatly improves the processing speed, saves time, and fully meets the requirements of real-time.
  • an edge intensity value of each pixel in the image is obtained by filtering an image, and a search is performed on the image by using a preset rectangular frame, and a rectangle on each position of the sliding search is performed.
  • a frame calculating an information quantity distribution value of the rectangular frame according to an edge intensity value of the inner pixel point, selecting a rectangular frame having the largest information amount distribution value, and intercepting the selected image content corresponding to the rectangular frame to obtain the image
  • the thumbnail image realizes the generation of thumbnails based on the image content information, which improves the accuracy of the thumbnail image expressing the original image content information, and is more in line with people's cognitive habits.
  • this embodiment provides a method for generating an image thumbnail, and the method includes the following steps.
  • step 201 the image is filtered to obtain an edge intensity value for each pixel in the image.
  • the filtering of the image can be implemented by using various filtering operators. For details, refer to the description in Embodiment 1, which is not here. Said.
  • step 202 the search is performed on the image by a preset rectangular frame, and the rectangular frame at each position of the sliding search is summed with the edge intensity values of all the pixels in the search to obtain the information distribution of the rectangular frame. value.
  • the size of the rectangular frame may be set as needed, as long as it is smaller than the size of the image.
  • the sliding search on the image may be a sliding search in any direction, which is not limited in this embodiment. For details, refer to the description in Embodiment 1, and details are not described herein again.
  • a rectangular box at each position of the sliding search can be calculated using the following formula:
  • the information quantity distribution value of each pixel in the rectangular frame is equal to the edge intensity value of the point, and therefore, the sum of the edge intensity values of the respective pixel points in the rectangular frame is the pixel of each pixel in the rectangular frame.
  • the information amount distribution values are summed, so that the information amount distribution value of the rectangular frame can be obtained.
  • step 203 a rectangular frame with the largest amount of information distribution is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • an edge intensity value of each pixel in the image is obtained by filtering an image, and a search is performed on the image by using a preset rectangular frame, and a rectangle on each position of the sliding search is performed. a frame, summing the edge intensity values of all the pixels in the rectangle to obtain the information distribution value of the rectangular frame, selecting a rectangular frame with the largest distribution value of the information, and intercepting the selected image content of the rectangular frame to obtain the
  • the thumbnail of the image since the thumbnail is generated based on the edge intensity value, the image-based content information is generated to generate a thumbnail image, so that the thumbnail image can include important and significant content in the image, and the thumbnail image representation of the original image content information is improved. Sex, more in line with people's cognitive habits.
  • the embodiment provides a method for generating an image thumbnail, and the method includes the following steps.
  • step 301 the image is filtered to obtain an edge intensity value for each pixel in the image.
  • the filtering of the image can be implemented by using various filtering operators. For details, refer to the description in Embodiment 1, which is not described here.
  • the edge intensity value can be normalized to obtain a value in the range of 0 to 255, and then calculated.
  • a spatial position attention value of each pixel in the image is calculated using a attention model established in advance based on the center point of the image and the coordinates of each pixel.
  • the step may include the following steps:
  • the value of the coefficient ⁇ can be set in advance according to requirements, for example, the minimum value of the length and the width of the image can be selected, and then 1/4 of the minimum value is taken as the value of the coefficient, etc., this embodiment is This is not specifically limited.
  • step 303 the information amount distribution value of each pixel in the image is calculated using an information amount distribution model established in advance based on the edge intensity value and the spatial position attention value.
  • the step may include the following steps:
  • the following information amount distribution model is used to calculate the information distribution value of each pixel in the image:
  • (i, j) represents any pixel in the image
  • /(i,j) represents the information distribution value of the pixel
  • £(i,j) represents the edge intensity value of the pixel
  • P( i, j) represents the spatial position attention value of the pixel.
  • step 304 a search is performed on the image by a preset rectangular frame, and a rectangular frame at each position of the sliding search is added, and the information distribution values of the pixels in the rectangular frame are added to obtain the rectangular frame. The amount of information distributed.
  • the size of the rectangular frame may be set as needed, as long as it is smaller than the size of the image.
  • the rectangular frame may be square and its side length is equal to the length of the short side of the image, so that the thumbnail obtained after the interception contains as much content information as possible.
  • other embodiments may be used, which are not specifically limited in this embodiment.
  • the sliding search of the rectangular frame on the image may be a sliding search in any direction, which is not limited in this embodiment.
  • any direction which is not limited in this embodiment.
  • step 305 a rectangular frame with the largest amount of information distribution is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • FIG. 3b a schematic diagram of a process for generating a thumbnail image provided by this embodiment is shown.
  • (1) is the original image.
  • (2) The result of filtering the original image by using the Laplacian edge filter operator, wherein the edge intensity value of each pixel in the image is also normalized, and the processed value range is 0 ⁇ 255.
  • (3) a result of calculating a spatial position attention value of each pixel in the image using a pre-established attention model, wherein the brighter portion indicates that the user's attention is higher, that is, the more the user is interested, the darker The part indicates that the user's attention is lower.
  • the edge intensity value of each pixel in the image is obtained by filtering the image, and the spatial position attention value of each pixel in the image is calculated by using a pre-established attention model, and is preset.
  • a rectangular frame is slidably searched on the image, and a rectangular frame at each position of the sliding search is used to calculate an information amount distribution value of each pixel in the rectangular frame by using a pre-established information amount distribution model, and the rectangular frame is obtained by summing
  • the information quantity distribution value is selected, and the rectangular frame with the largest distribution value of the information quantity is selected, and the image content corresponding to the selected rectangular frame is intercepted to obtain a thumbnail of the image, and the edge intensity value and the spatial position attention value are generated.
  • the thumbnail image makes the thumbnail not only include important and significant content in the image, but also considers the position of the image content information, which greatly improves the accuracy of the thumbnail image expressing the original image content information, and is more in line with people's cognitive habits.
  • the above algorithm also ensures the real-time performance of the method, and can achieve the effect of efficiently generating thumbnails.
  • an image with a size of 1600x 1200 can generate thumbnails in a time of 40 to 50 ms.
  • thumbnails can be generated in about 10ms, which fully meets the real-time requirements of mobile devices.
  • the embodiment provides a method for generating an image thumbnail, and the method includes the following steps.
  • step 401 the image is filtered to obtain an edge intensity value for each pixel in the image.
  • the filtering of the image can be implemented by using various filtering operators. For details, refer to the description in Embodiment 1, which is not described here.
  • a spatial position attention value of each pixel in the image is calculated using a attention model previously established based on the center point of the image and the coordinates of each pixel.
  • step 403 the information amount distribution value of each pixel in the image is calculated using an information amount distribution model established in advance based on the edge intensity value and the spatial position attention value.
  • step 404 a search is slid over the image with a predetermined rectangular frame.
  • the size of the rectangular frame may be set as needed, as long as it is smaller than the size of the image.
  • the rectangular frame may be square, and its side length is equal to the length of the short side of the image, so that the thumbnail obtained after the interception may contain as much content information as possible.
  • this embodiment does not do this. Specifically limited.
  • the sliding search of the rectangular frame on the image may be a sliding search in any direction, which is not limited in this embodiment.
  • any direction which is not limited in this embodiment.
  • step 405 for the rectangular frame at each position of the sliding search, the weight value corresponding to each pixel in the rectangular frame is calculated by using a pre-selected kernel function; the information distribution value of each pixel in the rectangular frame is corresponding to The weight value is multiplied and then summed to obtain the information distribution value of the rectangular frame.
  • the weight value of the kernel function that is closer to the pixel point from the center point of the image is larger.
  • the kernel function takes a smaller weight value for the pixel point farther from the center point of the image.
  • the kernel function may have multiple implementation manners, and is usually a functional form with low convex sides on both sides.
  • the kernel function may be set to a maximum weight value of 2 to 3 times the minimum weight value, and the minimum weight value is not 0, etc., of course, other methods may also be used.
  • a function form such as a sine function or a two-line line that rises first and then falls can be selected as a kernel function, which is not specifically limited in this embodiment.
  • FIG. 4b a schematic diagram of a kernel function provided by this embodiment is shown.
  • the rectangular frame is a square, and the square side has the same length as the short side of the original image.
  • the original image is shown as the long side in the horizontal direction and the short side in the vertical direction.
  • the rectangular frame performs a sliding search in the horizontal direction, and there is no sliding in the vertical direction.
  • the abscissa of the kernel function in the figure represents the abscissa of each pixel on the image, and the ordinate of the kernel function represents the weight value taken for each pixel.
  • the kernel function takes a larger weight value for the pixel points near the center of the image, and the weight value of the pixel points at both sides of the image is smaller, thereby calculating the information distribution of the rectangular frame. The value is then intercepted to ensure that the amount of image information is maximized as much as possible, and the most significant area is placed in the center of the thumbnail.
  • step 406 a rectangular frame having the largest amount of information distribution is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • the edge intensity value of each pixel in the image is obtained by filtering the image, and the spatial position attention value of each pixel in the image is calculated by using the attention model, and the preset rectangular frame is Performing a sliding search on the image, calculating a distribution amount of information of each pixel in the rectangular frame by using an information amount distribution model for a rectangular frame at each position of the sliding search, and calculating a weight corresponding to each pixel in the image by using a kernel function
  • the value is obtained by multiplying the information distribution value of each pixel in the rectangular frame with the corresponding weight value, and then obtaining the information distribution value of the rectangular frame, and selecting the rectangular frame with the largest information distribution value, and selecting
  • the image content corresponding to the rectangular frame is truncated to obtain a thumbnail of the image, and the thumbnail is generated based on the edge intensity value and the spatial position attention value, so that the thumbnail not only includes important and significant content in the image, but also considers The position of the image content information greatly improves the accuracy of the thumbnail image representing
  • the kernel function is used to calculate the weight value corresponding to each pixel point and is combined with the weight value for calculation, because the kernel function pairs the distance map The closer the pixel point closer to the center point is, the larger the weight value is. Therefore, the calculated information distribution value of the rectangular frame is more in line with the user's higher attention to the image center, and the amount of information in the image is made as much as possible.
  • the largest and most significant area is placed in the center of the thumbnail so that the thumbnails best reflect the focus of the image, fully meeting the needs of the user.
  • the above algorithm also ensures the real-time performance of the method, can achieve the effect of efficiently generating thumbnails, and meets the real-time requirements of devices such as mobile devices.
  • an image thumbnail generating apparatus including:
  • a filtering module 501 configured to filter an image to obtain an edge intensity value of each pixel in the image
  • a search module 502 configured to slide a search on the image by using a preset rectangular frame, and calculate a distribution amount of the information of the rectangular frame according to an edge intensity value of the pixel point in the rectangular frame at each position of the sliding search;
  • the intercepting module 503 is configured to select a rectangular frame with the largest distribution value of the information, and intercept the selected image content of the rectangular frame to obtain a thumbnail of the image.
  • the search module 502 can include a search unit for sliding a search on the image with a preset rectangular frame.
  • the calculating unit is configured to sum the edge intensity values of all the pixels in the sliding position of the sliding search to obtain the information quantity distribution value of the rectangular frame.
  • the search module 502 can include:
  • a searching unit 502a configured to slide a search on the image with a preset rectangular frame
  • the calculating unit 502b is configured to calculate a spatial position attention value of each pixel in the image by using a attention model established in advance according to a center point of the image and coordinates of each pixel point; using a pre-determined value according to the edge intensity value and the spatial position attention value
  • the information amount distribution model calculates the information amount distribution value of each pixel in the image; and scans the rectangular frame at each position of the search unit 502a, and adds the information quantity distribution values of the pixels in the rectangular frame to obtain The information distribution value of the rectangular frame.
  • the calculating unit 503b may include:
  • the information quantity distribution value calculation sub-unit is configured to calculate a weight value corresponding to each pixel point in the rectangular frame by using a pre-selected kernel function; and multiplying the information quantity distribution value of each pixel point in the rectangular frame with the corresponding weight value And summing, obtaining the information quantity distribution value of the rectangular frame; wherein, the kernel function has a larger weight value for the pixel point closer to the center point of the image.
  • the calculating unit 503b may include:
  • a spatial position attention value calculation sub-unit for calculating a spatial position attention value of each pixel in the image using the following attention model - (- (D ) 2 ) ;
  • the calculating unit 503b may include:
  • the information quantity distribution value calculation sub-unit is configured to calculate the information quantity distribution value of each pixel in the image by using the following information quantity distribution model:
  • (i, j) represents any pixel in the image
  • /(i,j) represents the information distribution value of the pixel
  • £(i,j) represents the edge intensity value of the pixel
  • P( i, j) represents the spatial position attention value of the pixel.
  • the rectangular frame may be a square, and the side length is equal to the length of the short side of the image.
  • the foregoing apparatus may further include:
  • the filtering module is configured to: filter the compressed image of the compression module to obtain an edge intensity value of each pixel in the image;
  • the intercepting module is configured to: the selected rectangular frame is corresponding to a rectangular frame in the original image, and the image content in the rectangular frame in the original image is intercepted to obtain a thumbnail of the original image.
  • the foregoing apparatus provided in this embodiment may be applied to a terminal, and the terminal includes but is not limited to: a mobile phone, a tablet computer, and the like.
  • the foregoing apparatus may perform the method in any one of the foregoing method embodiments. For details, refer to the description in the method embodiment, and details are not described herein again.
  • the apparatus provided in this embodiment obtains an edge intensity value of each pixel in the image by filtering an image, and scans the image on the image by a preset rectangular frame, and searches for a rectangle on each position of the sliding search. a frame, calculating an information quantity distribution value of the rectangular frame according to an edge intensity value of the inner pixel point, selecting a rectangular frame having the largest information amount distribution value, and intercepting the selected image content corresponding to the rectangular frame to obtain the image
  • the thumbnail image realizes the generation of thumbnails based on the image content information, which improves the accuracy of the thumbnail image expressing the original image content information, and is more in line with people's cognitive habits.
  • the embodiment provides a terminal 600, which may include a communication unit 610, a memory 620 including one or more non-volatile readable storage media, an input unit 630, a display unit 640, a sensor 650, and audio.
  • the circuit 660 a WiFi (wireless fidelity) module 670, a processor 680 including one or more processing cores, and a power supply 690 and the like.
  • WiFi wireless fidelity
  • the terminal structure shown in FIG. 6 does not constitute a limitation to the terminal, and may include more or less components than those illustrated, or a combination of certain components, or different component arrangements. among them:
  • the communication unit 610 can be used for transmitting and receiving information and receiving and transmitting signals during a call.
  • the communication unit 610 can be a network communication device such as an RF (Radio Frequency) circuit, a router, a modem, or the like. Specifically, when the communication unit 610 is an RF circuit, the downlink information of the base station is received, and then processed by one or more processors 680; in addition, data related to the uplink is transmitted to the base station.
  • RF Radio Frequency
  • RF circuitry as a communication unit includes, but is not limited to, an antenna, at least one amplifier, a tuner, one or more oscillators, a Subscriber Identity Module (SIM) card, a transceiver, a coupler, LNA (Low Noise Amplifier, low) Noise amplifiers, duplexers, etc.
  • communication unit 610 can also communicate with the network and other devices via wireless communication.
  • the wireless communication may use any communication standard or protocol, including but not limited to GSM (Global System of Mobile communication), GPRS (General Packet Radio Service), CDMA (Code Division Multiple Access).
  • the memory 620 can be used to store software programs and modules, and the processor 680 executes various functional applications and data processing by running software programs and modules stored in the memory 620.
  • the memory 620 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be stored according to The data created by the use of the terminal 600 (such as audio data, phone book, etc.) and the like.
  • memory 620 can include high speed random access memory, and can also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, memory 620 can also include a memory controller to provide access to memory 620 by processor 680 and input unit 630.
  • Input unit 630 can be used to receive input numeric or character information, as well as to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function controls.
  • input unit 630 can include touch-sensitive surface 630a as well as other input devices 630b.
  • Touch-sensitive surface 630a also referred to as a touch display or trackpad, can collect touch operations on or near the user (eg, the user uses a finger, stylus, etc., on any suitable object or accessory on touch-sensitive surface 630a or The operation near the touch-sensitive surface 630a) and driving the corresponding connecting device according to a preset program.
  • the touch-sensitive surface 630a may include two parts of a touch detection device and a touch controller.
  • the touch detection device detects the touch orientation of the user, and detects a signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts the touch information into contact coordinates, and sends the touch information
  • the processor 680 is provided and can receive commands from the processor 680 and execute them.
  • the touch-sensitive surface 630a can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves.
  • the input unit 630 can also include other input devices 630b.
  • other input devices 630b may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons) Etc., one or more of a trackball, a mouse, a joystick, and the like.
  • Display unit 640 can be used to display information entered by the user or information provided to the user as well as various graphical user interfaces of terminal 600, which can be comprised of graphics, text, icons, video, and any combination thereof.
  • the display unit 640 can include a display panel 640a.
  • the display panel 640a can be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like.
  • the touch-sensitive surface 630a may cover the display panel 640a, and when the touch-sensitive surface 630a detects a touch operation thereon or nearby, it is transmitted to the processor 680 to determine the type of the touch event, and then the processor 680 is based on the touch event.
  • touch-sensitive surface 630a and display panel 640a are implemented as two separate components to implement input and input functions, in some embodiments, touch-sensitive surface 630a can be integrated with display panel 640a for input. And output function.
  • Terminal 600 can also include at least one type of sensor 650, such as a light sensor, motion sensor, and other sensors.
  • the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 640a according to the brightness of the ambient light, and the proximity sensor may close the display panel 640a when the terminal 600 moves to the ear. And / or backlight.
  • the gravity acceleration sensor can detect the acceleration of each direction (usually three axes). When it is stationary, it can detect the magnitude and direction of gravity.
  • the terminal 600 can also be configured with gyroscopes, barometers, hygrometers, thermometers, infrared sensors and other sensors, not here Let me repeat.
  • the audio circuit 660, the speaker 660a, and the microphone 660b provide an audio interface between the user and the terminal 600.
  • the audio circuit 660 can transmit the converted electrical data of the received audio data to the speaker 660a for conversion to the sound signal output by the speaker 660a.
  • the microphone 660b converts the collected sound signal into an electrical signal by the audio circuit 660. After receiving, it is converted into audio data, and then processed by the audio data output processor 680, sent to the terminal, for example, by the RF circuit 610, or outputted to the memory 620 for further processing.
  • the audio circuit 660 may also include an earphone jack to provide communication of the peripheral earphones with the terminal 600.
  • the terminal may be configured with a wireless communication unit 670, which may be a WiFi module.
  • WiFi is a short-range wireless transmission technology, and the terminal 600 can help users to send and receive emails, browse web pages, and access streaming media through the wireless communication unit 670, which provides users with wireless broadband Internet access.
  • FIG. 6 shows the wireless communication unit 670, it can be understood that it does not belong to the essential configuration of the terminal 600, and may be omitted as needed within the scope of not changing the essence of the invention.
  • Processor 680 is the control center of terminal 600, which connects various portions of the entire handset using various interfaces and lines, by running or executing software programs and/or modules stored in memory 620, and recalling the number stored in memory 620. According to the various functions and processing data of the terminal 600, the overall monitoring of the mobile phone is performed.
  • the processor 680 may include one or more processing cores; in one embodiment, the processor 680 may integrate an application processor and a modem processor, where the application processor mainly processes an operating system, a user interface, and For applications, etc., the modem processor primarily handles wireless communications. It will be appreciated that the above described modem processor may also not be integrated into the processor 680.
  • the terminal 600 also includes a power source 690 (such as a battery) that supplies power to the various components.
  • the power source can be logically coupled to the processor 680 through a power management system to manage charging, discharging, and power management through the power management system. And other functions.
  • Power supply 690 may also include any one or more of a DC or AC power source, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.
  • the terminal 600 may further include a camera, a Bluetooth module, and the like, and details are not described herein.
  • terminal 600 An alternative structure of terminal 600 is presented above in connection with FIG. 6, wherein one or more modules are stored in the memory and configured to be executed by the one or more processors, the one or more modules having the following Features:
  • a rectangular frame having the largest distribution value of the information is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • the rectangular frame at each position of the sliding search calculates the information amount distribution value of the rectangular frame according to the edge intensity value of the inner pixel point, including:
  • the edge intensity values of all the pixels in the same are summed to obtain the information distribution value of the rectangular frame.
  • the sliding frame is searched on the image by a preset rectangular frame, and the rectangular frame at each position of the sliding search is calculated according to the edge intensity value of the inner pixel point, including the information distribution value of the rectangular frame, including :
  • the information quantity distribution value of each pixel in the rectangular frame is added to obtain an information quantity distribution value of the rectangular frame, including:
  • the weight value of the pixel function closer to the pixel point from the center point of the image is larger.
  • the calculating by using the attention model established in advance according to the center point of the image and the coordinates of each pixel point, calculating a spatial position attention value of each pixel in the image, including:
  • the spatial position attention value of each pixel in the image is calculated using the following model of interest:
  • the calculating the information distribution value of each pixel in the image by using the information quantity distribution model established in advance according to the edge intensity value and the spatial position attention value includes:
  • the following information amount distribution model is used to calculate the information distribution value of each pixel in the image:
  • (i, j) represents any pixel in the image
  • /(i, j) represents the information distribution value of the pixel
  • £(i, j) represents the edge intensity value of the pixel
  • P (i, j) represents the spatial position attention value of the pixel.
  • the rectangular frame is a square, and the side length is equal to the length of the short side of the image.
  • the image is filtered to obtain an edge intensity value of each pixel in the image, including:
  • the foregoing terminal provided in this embodiment may perform the method provided in any one of the foregoing method embodiments.
  • the terminal obtained in this embodiment obtains an edge intensity value of each pixel in the image by filtering an image, and scans the image on the image with a preset rectangular frame, and searches for a rectangle on each position of the sliding search. a frame, calculating an information quantity distribution value of the rectangular frame according to an edge intensity value of the inner pixel point, selecting a rectangular frame having the largest information amount distribution value, and intercepting the selected image content corresponding to the rectangular frame to obtain the image
  • the thumbnail image realizes the generation of thumbnails based on the image content information, which improves the accuracy of the thumbnail image expressing the original image content information, and is more in line with people's cognitive habits.
  • Example 7 This embodiment provides a non-volatile readable storage medium having one or more modules stored therein
  • programs when the one or more modules are applied to the device, the device can be caused to execute the instructions of the following steps:
  • a rectangular frame having the largest distribution value of the information is selected, and the selected image content of the rectangular frame is intercepted to obtain a thumbnail of the image.
  • the rectangular frame at each position of the sliding search calculates the information amount distribution value of the rectangular frame according to the edge intensity value of the inner pixel point, including:
  • the edge intensity values of all the pixels in the same are summed to obtain the information distribution value of the rectangular frame.
  • the sliding frame is searched on the image by a preset rectangular frame, and the rectangular frame at each position of the sliding search is calculated according to the edge intensity value of the inner pixel point, including the information distribution value of the rectangular frame, including :
  • the information quantity distribution value of each pixel in the rectangular frame is added to obtain an information quantity distribution value of the rectangular frame, including:
  • the information distribution value of each pixel in the rectangular frame is multiplied by the corresponding weight value, and then summed to obtain an information quantity distribution value of the rectangular frame;
  • the weight value of the pixel function closer to the pixel point from the center point of the image is larger.
  • the calculating by using the attention model established in advance according to the center point of the image and the coordinates of each pixel point, calculating a spatial position attention value of each pixel in the image, including:
  • the spatial position attention value of each pixel in the image is calculated using the following model of interest: Where (i, j) represents any pixel point in the image, and P(iJ) represents a spatial position attention value of the pixel point,
  • the calculating the information distribution value of each pixel in the image by using the information quantity distribution model established in advance according to the edge intensity value and the spatial position attention value includes:
  • the following information amount distribution model is used to calculate the information distribution value of each pixel in the image:
  • (i, j) represents any pixel in the image
  • /(i, j) represents the information distribution value of the pixel
  • £(i, j) represents the edge intensity value of the pixel
  • P (i, j) represents the spatial position attention value of the pixel.
  • the rectangular frame is a square, and the side length is equal to the length of the short side of the image.
  • the image is filtered to obtain an edge intensity value of each pixel in the image, including:
  • the non-volatile readable storage medium obtaineds an edge intensity value of each pixel in the image by filtering an image, and scans the image on the image by a preset rectangular frame, and slides Searching the rectangular frame at each position, calculating the information quantity distribution value of the rectangular frame according to the edge intensity value of the pixel point therein, selecting the rectangular frame with the largest information distribution value, and selecting the image content corresponding to the selected rectangular frame
  • the thumbnail image of the image is obtained, and the thumbnail image is generated based on the image content information, which improves the accuracy of the thumbnail image representing the original image content information, and is more in line with people's cognitive habits.
  • a person skilled in the art can understand that all or part of the steps of implementing the above embodiments may be completed by hardware, or may be instructed by a program to execute related hardware, and the program may be stored in a non-volatile readable storage medium.
  • the storage medium mentioned above may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Security & Cryptography (AREA)
  • Databases & Information Systems (AREA)
  • Library & Information Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)
  • Compression Of Band Width Or Redundancy In Fax (AREA)
  • Processing Or Creating Images (AREA)

Abstract

本公开提供了一种图像缩略图的生成方法、装置和终端,属于通信领域。所述方法包括:对图像进行滤波得到所述图像内各像素点的边缘强度值;以预先设定的矩形框在所述图像上滑动搜索,对滑动搜索的各个位置上的矩形框,根据其内像素点的边缘强度值计算该矩形框的信息量分布值;选取信息量分布值最大的矩形框,并将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图。所述装置包括:滤波模块、搜索模块和截取模块。本公开实现了基于图像的内容信息生成缩略图,提高了缩略图表达原图像内容信息的准确性,更加符合人们的认知习惯。

Description

图像缩略图的生成方法、 装置和终端 本申请基于申请号为 201310743545.7、申请日为 2013年 12月 30日的中国专利申请提出, 并要求该中国专利申请的优先权, 该中国专利申请的全部内容在此引入本申请作为参考。 技术领域
本公开涉及通信领域, 特别涉及一种图像缩略图的生成方法、 装置和终端。 背景技术
近年来随着移动设备自拍、 连拍等技术的实现, 移动设备中个人相册的容量也在飞速的 增长。 通常, 个人相册会为存储的图像生成缩略图, 并将缩略图置于预览模式中供用户预览 和査阅。 目前, 比较常见的图像缩略图生成方法是直接截取图像的中间部分, 将截取的中间 部分进行相应的缩放从而得到该图像的缩略图。 这种方法的优点就是非常高效。
但是, 上述方法仅考虑图像的空间位置信息, 完全没有考虑图像的内容信息, 会导致生 成的缩略图有时候不能充分表达原图像的内容信息。例如, 当一个图像中的人物位置较偏时, 如果只截取图像的中间部分, 则生成的缩略图中可能仅仅包括该人物的一部分, 导致表达原 图像内容信息的准确性较差。 发明内容
有鉴于此, 本公开提供了一种图像缩略图的生成方法和装置, 以提高缩略图表达原图像 内容信息的准确性。 所述技术方案如下:
第一方面, 提供了一种图像缩略图的生成方法, 包括:
对图像进行滤波得到所述图像内各像素点的边缘强度值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据 其内像素点的边缘强度值计算该矩形框的信息量分布值;
选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到 所述图像的缩略图。
其中, 所述对滑动搜索的各个位置上的矩形框, 根据其内像素点的边缘强度值计算该矩 形框的信息量分布值, 包括:
对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求和得到该矩形框 的信息量分布值。 其中, 所述以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩 形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 包括:
使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所述图像内各 像素点的空间位置关注值;
使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算所述图像内各 像素点的信息量分布值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 将该 矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值。
其中, 所述将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值, 包括:
使用预先选取的核函数计算所述矩形框内各像素点对应的权重值;
将所述矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得到该矩 形框的信息量分布值; 其中, 所述核函数对距离图像中心点越近的像素点所取的权重值越大。
其中, 所述使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所 述图像内各像素点的空间位置关注值, 包括:
使用如下关注模型计算所述图像内各像素点的空间位置关注值: (- (D )2 ) ;
2 * σ
其中, (i,j)表示所述图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值, (Χε, Υε)表示所述图像的中心点, σ为预设的系数。
其中, 所述使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算所 述图像内各像素点的信息量分布值, 包括:
使用如下信息量分布模型计算所述图像内各像素点的信息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示所述图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j) 表示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
其中, 所述矩形框为正方形, 且边长与所述图像短边的长度相等。
其中, 所述对图像进行滤波得到所述图像内各像素点的边缘强度值, 包括:
对原始图像进行压缩,对压缩后的图像进行滤波得到所述图像内各像素点的边缘强度值; 所述将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 包括: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始图像中的矩形框内的 图像内容截取下来得到所述原始图像的缩略图。
第二方面, 提供了一种图像缩略图的生成装置, 包括:
滤波模块, 用于对图像进行滤波得到所述图像内每个像素点的边缘强度值;
搜索模块, 用于以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上 的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值;
截取模块, 用于选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内 容截取下来得到所述图像的缩略图。
其中, 所述搜索模块包括:
搜索单元, 用于以预先设定的矩形框在所述图像上滑动搜索;
计算单元, 用于对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求 和得到该矩形框的信息量分布值。
其中, 所述搜索模块包括:
搜索单元, 用于以预先设定的矩形框在所述图像上滑动搜索;
计算单元, 用于使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计 算所述图像内各像素点的空间位置关注值; 使用预先根据边缘强度值和空间位置关注值建立 的信息量分布模型, 计算所述图像内各像素点的信息量分布值; 对所述搜索单元滑动搜索的 各个位置上的矩形框, 将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分 布值。
其中, 所述计算单元包括:
信息量分布值计算子单元, 用于使用预先选取的核函数计算所述矩形框内各像素点对应 的权重值; 将所述矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得 到该矩形框的信息量分布值;
其中, 所述核函数对距离图像中心点越近的像素点所取的权重值越大。
其中, 所述计算单元包括:
空间位置关注值计算子单元, 用于使用如下关注模型计算所述图像内各像素点的空间位 置关注值:
Figure imgf000005_0001
其中, (i,j)表示所述图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值, (Χε, Υε)表示所述图像的中心点, σ为预设的系数。
其中, 所述计算单元包括: 信息量分布值计算子单元, 用于使用如下信息量分布模型计算所述图像内各像素点的信 息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示所述图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j) 表示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
其中, 所述矩形框为正方形, 且边长与所述图像短边的长度相等。
其中, 所述装置还包括:
压缩模块, 用于对原始图像进行压缩;
所述滤波模块用于: 对所述压缩模块压缩后的图像进行滤波得到所述图像内各像素点的 边缘强度值;
所述截取模块用于: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始 图像中的矩形框内的图像内容截取下来得到所述原始图像的缩略图。
第三方面, 提供了一种终端, 包括有存储器, 以及一个或者一个以上的程序, 其中一个 或者一个以上程序存储于存储器中, 且经配置以由一个或者一个以上处理器执行所述一个或 者一个以上程序包含用于进行以下操作的指令:
对图像进行滤波得到所述图像内各像素点的边缘强度值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据 其内像素点的边缘强度值计算该矩形框的信息量分布值;
选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到 所述图像的缩略图。
本公开提供的技术方案带来的一些有益效果可以包括: 通过对图像进行滤波得到所述图 像内各像素点的边缘强度值, 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各 个位置上的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 选取信息 量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩 略图, 实现了基于图像的内容信息生成缩略图, 提高了缩略图表达原图像内容信息的准确性, 更加符合人们的认知习惯。
应当理解的是, 以上的一般描述和后文的细节描述仅是示例性的, 并不能限制本公开。 附图说明
为了更清楚地说明本公开的实施例中的技术方案, 下面将对实施例描述中所需要使用的 附图作简单地介绍, 显而易见地, 下面描述中的附图仅仅是本公开的一些实施例, 对于本领 域普通技术人员来讲, 在不付出创造性劳动的前提下, 还可以根据这些附图获得其他的附图。 图 1是本公开实施例 1提供的图像缩略图的生成方法示例性流程图;
图 2是本公开实施例 2提供的图像缩略图的生成方法示例性流程图;
图 3a是本公开实施例 3提供的图像缩略图的生成方法示例性流程图;
图 3b是本公开实施例 3提供的生成缩略图的过程示例性示意图;
图 4a是本公开实施例 4提供的图像缩略图的生成方法示例性流程图;
图 4b是本公开实施例 4提供的核函数示例性示意图;
图 5a是本公开实施例 5提供的图像缩略图的生成装置示例性结构图之一;
图 5b是本公开实施例 5提供的图像缩略图的生成装置示例性结构图之二;
图 6是本公开实施例 6提供的终端示例性结构图。
通过上述附图, 已示出本公开明确的实施例, 后文中将有更详细的描述。 这些附图和文 字描述并不是为了通过任何方式限制本公开构思的范围, 而是通过参考特定实施例为本领域 技术人员说明本公开的概念。 具体实施方式
为使本公开的目的、 技术方案和优点更加清楚, 下面将结合附图对本公开实施方式作进 一步地详细描述。
实施例 1
参见图 1, 本实施例提供了一种图像缩略图的生成方法, 该方法包括如下步骤。
在步骤 101中, 对图像进行滤波得到该图像内各像素点的边缘强度值。
本实施例中, 对图像进行滤波得到该图像内各像素点的边缘强度值, 可以包括: 采用拉 普拉斯边缘滤波算子、 索贝尔边缘滤波算子、 Robert边缘算子、 Prewitt边缘算子或 LOG边 缘算子对图像进行滤波, 得到该图像内各像素点的边缘强度值。
本实施例中, 边缘强度值比较接近的像素点, 可以认为颜色的差异性不大; 边缘强度值 相差较多的像素点, 可以认为颜色的差异性较大, 因此, 可以认为边缘强度值在一定程度上 反映了图像的内容信息。
在步骤 102中, 以预先设定的矩形框在该图像上滑动搜索, 对滑动搜索的各个位置上的 矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值。
其中, 预先设定的矩形框可以是比该图像小的任何一种尺寸的矩形框。 例如, 一种情况 下, 矩形框的短边等于图像的短边, 矩形框的长边小于图像的长边。 另一种情况下, 矩形框 的短边小于图像的短边, 矩形框的长边等于图像的长边。 又一种情况下, 矩形框的短边小于 图像的短边, 且矩形框的长边小于图像的长边等等, 本实施例对此不做具体限定。
所述矩形框在图像上进行滑动搜索可以是任意方向上的滑动搜索, 本实施例对此不做具 体限定, 如可以为仅在横向上滑动搜索, 或者仅在纵向上滑动搜索, 或者沿着 45°的方向搜索 。
在步骤 103 中, 选取信息量分布值最大的矩形框, 并将选取的该矩形框对应的图像内容 截取下来得到该图像的缩略图。
本实施例中, 生成的缩略图的大小不限定, 如可以为 1600x 1200的图像等等。 其中, 截 取下来的图像还可以先进行压缩, 然后将压缩后的图像作为缩略图, 本实施例对此不做具体 限定。
本实施例中, 上述对滑动搜索的各个位置上的矩形框, 根据其内像素点的边缘强度值计 算该矩形框的信息量分布值, 可以包括:
对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求和得到该矩形框 的信息量分布值。
本实施例中, 上述以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置 上的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 可以包括:
使用预先根据该图像的中心点和各像素点的坐标建立的关注模型, 计算该图像内各像素 点的空间位置关注值; 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置 上的矩形框, 使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算该矩 形框内各像素点的信息量分布值, 对该矩形框内各像素点的信息量分布值求和得到该矩形框 的信息量分布值。
本实施例中, 上述对该矩形框内各像素点的信息量分布值求和得到该矩形框的信息量分 布值, 可以包括:
使用预先选取的核函数计算该图像内各像素点对应的权重值; 将该矩形框内各像素点的 信息量分布值与对应的权重值进行乘积后再求和, 得到该矩形框的信息量分布值; 其中, 该 核函数对距离图像中心点越近的像素点所取的权重值越大。
本实施例中, 上述使用预先根据该图像的中心点和各像素点的坐标建立的关注模型, 计 算该图像内各像素点的空间位置关注值, 可以包括:
使用如下关注模型计算该图像内各像素点的空间位置关注值: (- (D )2 ) ;
2 * σ
其中, (i,j)表示该图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值,(Χεε) 表示该图像的中心点, σ为预设的系数。
本实施例中, 上述使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算该矩形框内各像素点的信息量分布值, 可以包括:
使用如下信息量分布模型计算该矩形框内各像素点的信息量分布值:
/(iJ) = E(iJ) * P(ij) ;
其中, (i,j)表示该图像内的任一个像素点, /(i, j)表示该像素点的信息量分布值, £(i, j)表 示该像素点的边缘强度值, P(i, j)表示该像素点的空间位置关注值。
本实施例中, 上述矩形框可以为正方形, 且边长与该图像短边的长度相等。 从而可以使 得截取后得到的缩略图包含尽可能多的内容信息。
为了提高运算效率, 上述方法中, 还可以在滤波前先对上述图像进行压缩处理, 得到分 辨率更小的图像, 然后再执行滤波等后续步骤, 在选取信息量分布值最大的矩形框之后, 再 将该矩形框转换为原始图像对应的位置进行截取即可。 其中, 所述对图像进行滤波得到所述 图像内各像素点的边缘强度值, 可以包括: 对原始图像进行压缩, 对压缩后的图像进行滤波 得到所述图像内各像素点的边缘强度值;
相应地, 所述将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 包 括: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始图像中的矩形框内的 图像内容截取下来得到所述原始图像的缩略图。
例如, 一张 1600x 1200的图像, 先压缩为 400x400的图像, 然后在该 400x400的图像上 选取矩形框, 选取完成后再将该矩形框对应的区域转换为 1600x 1200图像上对应的区域, 然 后进行截取及压缩就可以得到缩略图。 这种方式极大地提高了处理速度, 节省了时间, 充分 满足了实时性的要求。
本实施例提供的上述方法,通过对图像进行滤波得到所述图像内各像素点的边缘强度值, 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据其内 像素点的边缘强度值计算该矩形框的信息量分布值, 选取信息量分布值最大的矩形框, 并将 选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 实现了基于图像的内容 信息生成缩略图, 提高了缩略图表达原图像内容信息的准确性, 更加符合人们的认知习惯。 实施例 2
参见图 2, 本实施例提供了一种图像缩略图的生成方法, 该方法包括如下步骤。
在步骤 201中, 对图像进行滤波得到该图像内各像素点的边缘强度值。
其中, 对图像进行滤波可以采用多种滤波算子实现, 详见实施例 1 中的描述, 此处不赘 述。
在步骤 202中, 以预先设定的矩形框在该图像上滑动搜索, 对滑动搜索的各个位置上的 矩形框, 对其内所有像素点的边缘强度值求和得到该矩形框的信息量分布值。
其中, 所述矩形框的大小可以根据需要设定, 只要比该图像的尺寸小即可。 所述矩形框 在图像上进行滑动搜索可以是任意方向上的滑动搜索, 本实施例对此不做具体限定, 可以参 见实施例 1中的描述, 此处不再赘述。
例如, 对滑动搜索的各个位置上的矩形框可以采用如下公式进行计算:
/ = );
其中, (i,j)表示该图像内的任一个像素点, £(i, j)表示该像素点的边缘强度值, /表示该 矩形框的信息量分布值。
这里可以认为将该矩形框内的各个像素点的信息量分布值等于该点的边缘强度值,因此, 该矩形框内各个像素点的边缘强度值的求和就是该矩形框内各个像素点的信息量分布值求 和, 从而可以得到该矩形框的信息量分布值。
在步骤 203 中, 选取信息量分布值最大的矩形框, 并将选取的该矩形框对应的图像内容 截取下来得到该图像的缩略图。
本实施例提供的上述方法,通过对图像进行滤波得到所述图像内各像素点的边缘强度值, 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 对其内所 有像素点的边缘强度值求和得到该矩形框的信息量分布值,选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 由于基于边缘强度 值生成缩略图, 实现了基于图像的内容信息生成缩略图, 使得缩略图能够包括图像中重要以 及显著的内容, 提高了缩略图表达原图像内容信息的准确性, 更加符合人们的认知习惯。 实施例 3
参见图 3a, 本实施例提供了一种图像缩略图的生成方法, 该方法包括如下步骤。
在步骤 301中, 对图像进行滤波得到该图像内各像素点的边缘强度值。
其中, 对图像进行滤波可以采用多种滤波算子实现, 详见实施例 1 中的描述, 此处不赘 述。
另外, 为了计算方便还可以对边缘强度值进行归一化处理, 得到 0〜255范围内的数值, 然后再进行计算。
在步骤 302中, 使用预先根据该图像的中心点和各像素点的坐标建立的关注模型, 计算 该图像内各像素点的空间位置关注值。 其中, 本步骤可以包括如下步骤:
使用如下关注模型计算该图像内各像素点的空间位置关注值:
Figure imgf000011_0001
其中, (i,j)表示该图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值, (XC,YC) 表示该图像的中心点, σ为预设的系数。
本实施例中, 系数 σ的数值可以根据需要预先设置, 如可以选取图像的长和宽中的最小 值, 再取该最小值的 1/4作为该系数的取值等等, 本实施例对此不做具体限定。
在步骤 303 中, 使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计 算该图像内各像素点的信息量分布值。
其中, 本步骤可以包括如下步骤:
使用如下信息量分布模型计算该图像内各像素点的信息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示该图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j)表 示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
在步骤 304中, 以预先设定的矩形框在该图像上滑动搜索, 对滑动搜索的各个位置上的 矩形框, 将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值。
其中, 所述矩形框的大小可以根据需要设定, 只要比该图像的尺寸小即可。 在一种实施 方式下, 该矩形框可以为正方形, 且其边长与该图像短边的长度相等, 从而可以使得截取后 得到的缩略图包含尽可能多的内容信息。 当然也可以采用其它实施方式, 本实施例对此不做 具体限定。
所述矩形框在图像上进行滑动搜索可以是任意方向上的滑动搜索, 本实施例对此不做具 体限定, 可以参见实施例 1中的描述, 此处不再赘述。
在步骤 305中, 选取信息量分布值最大的矩形框, 并将选取的该矩形框对应的图像内容 截取下来得到该图像的缩略图。
参见图 3b, 为本实施例提供的生成缩略图的过程示意图。 其中, (1 ) 中为原始图像。 (2) 中为采用拉普拉斯边缘滤波算子对原始图像进行滤波后的结果, 其中, 图像中每个像素点的 边缘强度值还进行了归一化处理, 处理后的取值范围为 0〜255。 (3 ) 为使用预先建立的关注 模型计算该图像内各像素点的空间位置关注值后的结果, 其中, 越亮的部分表示用户的关注 度越高, 即用户越感兴趣的区域, 越暗的部分表示用户的关注度越低。 (4) 为使用预先建立 的信息量分布模型计算该图像内各像素点的信息量分布值后结果, 其中, 各信息量分布值也 进行了归一化处理, 处理后的取值范围为 0〜255。 从该结果可以看出, 结合了边缘强度值以 及空间位置关注值, 信息量分布值较高的像素点均已经在图像中显示出来, 因此, 按照信息 量分布值最大来选取矩形框能够生成更准确的缩略图。 与仅仅选取图像的中央部分相比, 生 成的缩略图反映原始图像的内容信息更准确。
本实施例提供的上述方法,通过对图像进行滤波得到所述图像内各像素点的边缘强度值, 使用预先建立的关注模型计算该图像内各像素点的空间位置关注值, 以预先设定的矩形框在 所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 使用预先建立的信息量分布模型 计算该矩形框内各像素点的信息量分布值, 并求和得到该矩形框的信息量分布值, 选取信息 量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩 略图, 实现了基于边缘强度值和空间位置关注值生成缩略图, 使得缩略图不仅包括了图像中 重要以及显著的内容, 而且考虑了图像内容信息的位置, 极大地提高了缩略图表达原图像内 容信息的准确性, 更加符合人们的认知习惯。 另外, 上述算法也保证了所述方法的实时性, 能够达到高效地生成缩略图的效果, 一般一张大小为 1600x 1200的图像大概在 40〜50ms的 时间内就可以生成缩略图, 一张大小为 100x 100的图像大概在 10ms左右就可以生成缩略图, 完全能够满足移动设备等对实时性的要求。 实施例 4
参见图 4a, 本实施例提供了一种图像缩略图的生成方法, 该方法包括如下步骤。
在步骤 401中, 对图像进行滤波得到该图像内各像素点的边缘强度值。
其中, 对图像进行滤波可以采用多种滤波算子实现, 详见实施例 1 中的描述, 此处不赘 述。
在步骤 402中, 使用预先根据该图像的中心点和各像素点的坐标建立的关注模型, 计算 该图像内各像素点的空间位置关注值。
其中, 该关注模型可以参见实施例 3中的描述, 此处不再赘述。
在步骤 403 中, 使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计 算该图像内各像素点的信息量分布值。
其中, 该信息量分布模型可以参见实施例 3中的描述, 此处不再赘述。
在步骤 404中, 以预先设定的矩形框在该图像上滑动搜索。
其中, 所述矩形框的大小可以根据需要设定, 只要比该图像的尺寸小即可。 在一种实施 方式下, 该矩形框可以为正方形, 且其边长与该图像短边的长度相等, 从而可以使得截取后 得到的缩略图包含尽可能多的内容信息。 当然也可以采用其它实施方式, 本实施例对此不做 具体限定。
所述矩形框在图像上进行滑动搜索可以是任意方向上的滑动搜索, 本实施例对此不做具 体限定, 可以参见实施例 1中的描述, 此处不再赘述。
在步骤 405中, 对滑动搜索的各个位置上的矩形框, 使用预先选取的核函数计算该矩形 框内各像素点对应的权重值; 将该矩形框内各像素点的信息量分布值与对应的权重值进行乘 积后再求和, 得到该矩形框的信息量分布值。
其中, 该核函数对距离图像中心点越近的像素点所取的权重值越大。 相应的, 该核函数 对距离图像中心点越远的像素点所取的权重值越小。
本实施例中, 核函数可以有多种实现方式, 通常为中间凸两边低的函数形式。 其中, 可 以将核函数设置为最大权重值为最小权重值的 2〜3倍, 且最小权重值不为 0等等, 当然也可 以采用其它方式。 例如, 可以选取正弦函数、 或者先上升后下降的两条直线等函数形式作为 核函数, 本实施例对此不做具体限定。
参见图 4b, 为本实施例提供的一种核函数示意图。 这里以矩形框是正方形为例, 且该正 方形的边长与原始图像短边的长度相等。 以原始图像横向为长边, 纵向为短边进行说明。 滑 动搜索时矩形框在水平方向上进行滑动搜索, 在垂直方向上无滑动。 图中核函数的横坐标表 示图像上各个像素点的横坐标, 核函数的纵坐标表示为各个像素点所取的权重值。 从图中可 以看出, 该核函数对图像中心附近的像素点所取的权重值较大, 对图像两边处的像素点所取 的权重值较小, 以此计算出矩形框的信息量分布值, 再进行截取, 从而保证了尽可能的将图 像信息量最大, 最显著的区域放置在缩略图的中央。
在步骤 406中, 选取信息量分布值最大的矩形框, 并将选取的该矩形框对应的图像内容 截取下来得到该图像的缩略图。
本实施例提供的上述方法,通过对图像进行滤波得到所述图像内各像素点的边缘强度值, 使用关注模型计算该图像内各像素点的空间位置关注值, 以预先设定的矩形框在所述图像上 滑动搜索, 对滑动搜索的各个位置上的矩形框, 使用信息量分布模型计算该矩形框内各像素 点的信息量分布值, 使用核函数计算该图像内各像素点对应的权重值; 将该矩形框内各像素 点的信息量分布值与对应的权重值进行乘积后再求和, 得到该矩形框的信息量分布值, 选取 信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到所述图像 的缩略图, 实现了基于边缘强度值和空间位置关注值生成缩略图, 使得缩略图不仅包括了图 像中重要以及显著的内容, 而且考虑了图像内容信息的位置, 极大地提高了缩略图表达原图 像内容信息的准确性, 更加符合人们的认知习惯。 进一步地, 在计算矩形框的信息量分布值 时, 使用核函数计算各像素点对应的权重值并结合权重值进行计算, 由于该核函数对距离图 像中心点越近的像素点所取的权重值越大, 因此, 计算出的矩形框的信息量分布值更加符合 用户对图像中心关注度更高的特点, 且尽可能的使图像中信息量最大、 最显著的区域放置在 缩略图的中央, 使得缩略图最能反映图像的重点部分, 充分满足了用户的需求。 另外, 上述 算法也保证了所述方法的实时性, 能够达到高效地生成缩略图的效果, 符合移动设备等设备 对实时性的要求。 实施例 5
参见图 5a, 本实施例提供了一种图像缩略图的生成装置, 包括:
滤波模块 501, 用于对图像进行滤波得到该图像内每个像素点的边缘强度值;
搜索模块 502, 用于以预先设定的矩形框在该图像上滑动搜索, 对滑动搜索的各个位置 上的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值;
截取模块 503, 用于选取信息量分布值最大的矩形框, 并将选取的该矩形框对应的图像 内容截取下来得到该图像的缩略图。
其中, 搜索模块 502可以包括- 搜索单元, 用于以预先设定的矩形框在该图像上滑动搜索;
计算单元, 用于对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求 和得到该矩形框的信息量分布值。
或者, 参见图 5b, 搜索模块 502可以包括:
搜索单元 502a, 用于以预先设定的矩形框在该图像上滑动搜索;
计算单元 502b, 用于使用预先根据该图像的中心点和各像素点的坐标建立的关注模型, 计算该图像内各像素点的空间位置关注值; 使用预先根据边缘强度值和空间位置关注值建立 的信息量分布模型, 计算所述图像内各像素点的信息量分布值; 对搜索单元 502a滑动搜索的 各个位置上的矩形框, 将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分 布值。
其中, 上述计算单元 503b可以包括:
信息量分布值计算子单元, 用于使用预先选取的核函数计算该矩形框内各像素点对应的 权重值; 将该矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得到该 矩形框的信息量分布值; 其中, 该核函数对距离图像中心点越近的像素点所取的权重值越大。
其中, 上述计算单元 503b可以包括:
空间位置关注值计算子单元, 用于使用如下关注模型计算该图像内各像素点的空间位置 关注值- (- (D )2 ) ;
2 * σ
其中, (i,j)表示该图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值, (XC,YC) 表示该图像的中心点, σ为预设的系数。
其中, 上述计算单元 503b可以包括:
信息量分布值计算子单元, 用于使用如下信息量分布模型计算该图像内各像素点的信息 量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示该图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j)表 示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
本实施例中, 上述矩形框可以为正方形, 且边长与该图像短边的长度相等。
本实施例中, 上述装置还可以包括:
压缩模块, 用于对原始图像进行压缩;
所述滤波模块用于: 对所述压缩模块压缩后的图像进行滤波得到所述图像内各像素点的 边缘强度值;
所述截取模块用于: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始 图像中的矩形框内的图像内容截取下来得到所述原始图像的缩略图。
本实施例提供的上述装置可以应用于终端中, 该终端包括但不限于: 手机、 平板电脑等 等。 上述装置可以执行上述任一方法实施例中的方法, 详细过程详见方法实施例中的描述, 此处不再赘述。
本实施例提供的上述装置,通过对图像进行滤波得到所述图像内各像素点的边缘强度值, 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据其内 像素点的边缘强度值计算该矩形框的信息量分布值, 选取信息量分布值最大的矩形框, 并将 选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 实现了基于图像的内容 信息生成缩略图, 提高了缩略图表达原图像内容信息的准确性, 更加符合人们的认知习惯。 实施例 6
参见图 6, 本实施例提供了一种终端 600, 可以包括通信单元 610、 包括有一个或一个以 上非易失性可读存储介质的存储器 620、 输入单元 630、 显示单元 640、 传感器 650、 音频电 路 660、 WiFi(wireless fidelity, 无线保真)模块 670、包括有一个或者一个以上处理核心的处理 器 680、 以及电源 690等部件。 本领域技术人员可以理解, 图 6中示出的终端结构并不构成对终端的限定, 可以包括比 图示更多或更少的部件, 或者组合某些部件, 或者不同的部件布置。 其中:
通信单元 610可用于收发信息或通话过程中, 信号的接收和发送, 该通信单元 610可以 为 RF (Radio Frequency, 射频) 电路、 路由器、 调制解调器、 等网络通信设备。 特别地, 当 通信单元 610为 RF 电路时, 将基站的下行信息接收后, 交由一个或者一个以上处理器 680 处理; 另外, 将涉及上行的数据发送给基站。 通常, 作为通信单元的 RF 电路包括但不限于 天线、 至少一个放大器、 调谐器、 一个或多个振荡器、 用户身份模块 (SIM) 卡、 收发信机、 耦合器、 LNA (Low Noise Amplifier, 低噪声放大器)、 双工器等。 此外, 通信单元 610还可 以通过无线通信与网络和其他设备通信。 所述无线通信可以使用任一通信标准或协议, 包括 但不限于 GSM(Global System of Mobile communication, 全球移动通讯系统)、 GPRS(General Packet Radio Service, 通用分组无线服务)、 CDMA(Code Division Multiple Access, 码分多址)、 WCDMA(Wideband Code Division Multiple Access, 宽带码分多址)、 LTE(Long Term Evolution, 长期演进)、 电子邮件、 SMS(Short Messaging Service, 短消息服务)等。 存储器 620可用于存 储软件程序以及模块, 处理器 680通过运行存储在存储器 620的软件程序以及模块, 从而执 行各种功能应用以及数据处理。 存储器 620可主要包括存储程序区和存储数据区, 其中, 存 储程序区可存储操作系统、 至少一个功能所需的应用程序 (比如声音播放功能、 图像播放功 能等)等; 存储数据区可存储根据终端 600的使用所创建的数据(比如音频数据、 电话本等) 等。 此外, 存储器 620可以包括高速随机存取存储器, 还可以包括非易失性存储器, 例如至 少一个磁盘存储器件、 闪存器件、 或其他易失性固态存储器件。 相应地, 存储器 620还可以 包括存储器控制器, 以提供处理器 680和输入单元 630对存储器 620的访问。
输入单元 630可用于接收输入的数字或字符信息, 以及产生与用户设置以及功能控制有 关的键盘、 鼠标、 操作杆、 光学或者轨迹球信号输入。 可选地, 输入单元 630可包括触敏表 面 630a以及其他输入设备 630b。 触敏表面 630a, 也称为触摸显示屏或者触控板, 可收集用 户在其上或附近的触摸操作 (比如用户使用手指、 触笔等任何适合的物体或附件在触敏表面 630a上或在触敏表面 630a附近的操作), 并根据预先设定的程式驱动相应的连接装置。 可选 的, 触敏表面 630a可包括触摸检测装置和触摸控制器两个部分。 其中, 触摸检测装置检测用 户的触摸方位, 并检测触摸操作带来的信号, 将信号传送给触摸控制器; 触摸控制器从触摸 检测装置上接收触摸信息, 并将它转换成触点坐标, 再送给处理器 680, 并能接收处理器 680 发来的命令并加以执行。 此外, 可以采用电阻式、 电容式、 红外线以及表面声波等多种类型 实现触敏表面 630a。 除了触敏表面 630a, 输入单元 630还可以包括其他输入设备 630b。可选 地, 其他输入设备 630b可以包括但不限于物理键盘、 功能键(比如音量控制按键、 开关按键 等)、 轨迹球、 鼠标、 操作杆等中的一种或多种。
显示单元 640可用于显示由用户输入的信息或提供给用户的信息以及终端 600的各种图 形用户接口, 这些图形用户接口可以由图形、 文本、 图标、 视频和其任意组合来构成。 显示 单元 640可包括显示面板 640a,可选的, 可以采用 LCD(Liquid Crystal Display,液晶显示器)、 OLED(Organic Light-Emitting Diode,有机发光二极管)等形式来配置显示面板 640a。进一步的, 触敏表面 630a可覆盖显示面板 640a, 当触敏表面 630a检测到在其上或附近的触摸操作后, 传送给处理器 680以确定触摸事件的类型, 随后处理器 680根据触摸事件的类型在显示面板 640a上提供相应的视觉输出。 虽然在图 6中, 触敏表面 630a与显示面板 640a是作为两个独 立的部件来实现输入和输入功能, 但是在某些实施例中, 可以将触敏表面 630a与显示面板 640a集成而实现输入和输出功能。
终端 600还可包括至少一种传感器 650, 比如光传感器、 运动传感器以及其他传感器。 可选地, 光传感器可包括环境光传感器及接近传感器, 其中, 环境光传感器可根据环境光线 的明暗来调节显示面板 640a的亮度, 接近传感器可在终端 600移动到耳边时, 关闭显示面板 640a和 /或背光。作为运动传感器的一种,重力加速度传感器可检测各个方向上(一般为三轴) 加速度的大小, 静止时可检测出重力的大小及方向, 可用于识别手机姿态的应用 (比如横竖 屏切换、 相关游戏、 磁力计姿态校准)、 振动识别相关功能 (比如计步器、 敲击) 等; 至于终 端 600还可配置的陀螺仪、 气压计、 湿度计、 温度计、 红外线传感器等其他传感器, 在此不 再赘述。
音频电路 660、 扬声器 660a, 传声器 660b可提供用户与终端 600之间的音频接口。 音频 电路 660可将接收到的音频数据转换后的电信号,传输到扬声器 660a, 由扬声器 660a转换为 声音信号输出; 另一方面, 传声器 660b将收集的声音信号转换为电信号, 由音频电路 660接 收后转换为音频数据, 再将音频数据输出处理器 680处理后, 经 RF电路 610以发送给比如 另一终端, 或者将音频数据输出至存储器 620以便进一步处理。 音频电路 660还可能包括耳 塞插孔, 以提供外设耳机与终端 600的通信。
为了实现无线通信, 该终端上可以配置有无线通信单元 670, 该无线通信单元 670可以 为 WiFi模块。 WiFi属于短距离无线传输技术, 终端 600通过无线通信单元 670可以帮助用 户收发电子邮件、 浏览网页和访问流式媒体等, 它为用户提供了无线的宽带互联网访问。 虽 然图 6示出了无线通信单元 670, 但是可以理解的是, 其并不属于终端 600的必须构成, 完 全可以根据需要在不改变发明的本质的范围内而省略。
处理器 680是终端 600的控制中心, 利用各种接口和线路连接整个手机的各个部分, 通 过运行或执行存储在存储器 620内的软件程序和 /或模块, 以及调用存储在存储器 620内的数 据, 执行终端 600的各种功能和处理数据, 从而对手机进行整体监控。 可选的, 处理器 680 可包括一个或多个处理核心; 在一个实施例中, 处理器 680可集成应用处理器和调制解调处 理器, 其中, 应用处理器主要处理操作系统、 用户界面和应用程序等, 调制解调处理器主要 处理无线通信。 可以理解的是, 上述调制解调处理器也可以不集成到处理器 680中。
终端 600还包括给各个部件供电的电源 690 (比如电池), 在一个实施例中, 电源可以通 过电源管理系统与处理器 680逻辑相连, 从而通过电源管理系统实现管理充电、 放电、 以及 功耗管理等功能。 电源 690还可以包括一个或一个以上的直流或交流电源、 再充电系统、 电 源故障检测电路、 电源转换器或者逆变器、 电源状态指示器等任意组件。
尽管未示出, 终端 600还可以包括摄像头、 蓝牙模块等, 在此不再赘述。
以上结合图 6给出了终端 600的可选结构, 其中一个或多个模块存储于所述存储器中 并被配置成由所述一个或多个处理器执行, 所述一个或多个模块具有如下功能:
对图像进行滤波得到所述图像内各像素点的边缘强度值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据 其内像素点的边缘强度值计算该矩形框的信息量分布值;
选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到 所述图像的缩略图。
其中, 所述对滑动搜索的各个位置上的矩形框, 根据其内像素点的边缘强度值计算该矩 形框的信息量分布值, 包括:
对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求和得到该矩形框 的信息量分布值。
其中, 所述以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩 形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 包括:
使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所述图像内各 像素点的空间位置关注值;
使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算所述图像内各 像素点的信息量分布值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 将该 矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值。
其中, 所述将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值, 包括:
使用预先选取的核函数计算所述矩形框内各像素点对应的权重值; 将所述矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得到该矩 形框的信息量分布值;
其中, 所述核函数对距离图像中心点越近的像素点所取的权重值越大。
其中, 所述使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所 述图像内各像素点的空间位置关注值, 包括:
使用如下关注模型计算所述图像内各像素点的空间位置关注值:
Figure imgf000019_0001
其中, (i,j)表示所述图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值, (Χε, Υε)表示所述图像的中心点, σ为预设的系数。
其中, 所述使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算该 图像内各像素点的信息量分布值, 包括:
使用如下信息量分布模型计算该图像内各像素点的信息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示所述图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j) 表示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
其中, 所述矩形框为正方形, 且边长与所述图像短边的长度相等。
其中, 所述对图像进行滤波得到所述图像内各像素点的边缘强度值, 包括:
对原始图像进行压缩,对压缩后的图像进行滤波得到所述图像内各像素点的边缘强度值; 所述将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 包括: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始图像中的矩形框内的 图像内容截取下来得到所述原始图像的缩略图。
本实施例提供的上述终端可以执行上述任一方法实施例中提供的方法, 过程详见方法实 施例中的描述, 此处不再赘述。
本实施例提供的上述终端,通过对图像进行滤波得到所述图像内各像素点的边缘强度值, 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据其内 像素点的边缘强度值计算该矩形框的信息量分布值, 选取信息量分布值最大的矩形框, 并将 选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 实现了基于图像的内容 信息生成缩略图, 提高了缩略图表达原图像内容信息的准确性, 更加符合人们的认知习惯。 实施例 7 本实施例提供了一种非易失性可读存储介质, 该存储介质中存储有一个或多个模块
( programs) , 该一个或多个模块被应用在设备中时, 可以使得该设备执行如下步骤的指 令 ( instructions) :
对图像进行滤波得到所述图像内各像素点的边缘强度值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据 其内像素点的边缘强度值计算该矩形框的信息量分布值;
选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到 所述图像的缩略图。
其中, 所述对滑动搜索的各个位置上的矩形框, 根据其内像素点的边缘强度值计算该矩 形框的信息量分布值, 包括:
对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求和得到该矩形框 的信息量分布值。
其中, 所述以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩 形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 包括:
使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所述图像内各 像素点的空间位置关注值;
使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算所述图像内各 像素点的信息量分布值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 将该 矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值。
其中, 所述将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值, 包括:
使用预先选取的核函数计算所述矩形框内各像素点对应的权重值;
将所述矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得到该矩 形框的信息量分布值;
其中, 所述核函数对距离图像中心点越近的像素点所取的权重值越大。
其中, 所述使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所 述图像内各像素点的空间位置关注值, 包括:
使用如下关注模型计算所述图像内各像素点的空间位置关注值:
Figure imgf000020_0001
其中, (i,j)表示所述图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值,
ε, Υε)表示所述图像的中心点, σ为预设的系数。
其中, 所述使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算该 图像内各像素点的信息量分布值, 包括:
使用如下信息量分布模型计算该图像内各像素点的信息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示所述图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j) 表示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
其中, 所述矩形框为正方形, 且边长与所述图像短边的长度相等。
其中, 所述对图像进行滤波得到所述图像内各像素点的边缘强度值, 包括:
对原始图像进行压缩,对压缩后的图像进行滤波得到所述图像内各像素点的边缘强度值; 所述将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 包括: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始图像中的矩形框内的 图像内容截取下来得到所述原始图像的缩略图。
本实施例提供的上述非易失性可读存储介质,通过对图像进行滤波得到所述图像内各像 素点的边缘强度值, 以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上 的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 选取信息量分布值 最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 实 现了基于图像的内容信息生成缩略图, 提高了缩略图表达原图像内容信息的准确性, 更加符 合人们的认知习惯。 本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成, 也可以通过程序来指令相关的硬件完成,所述的程序可以存储于一种非易失性可读存储介质 中, 上述提到的存储介质可以是只读存储器, 磁盘或光盘等。
以上所述仅为本公开的较佳实施例, 并不用以限制本公开, 凡在本公开的精神和原则之 内, 所作的任何修改、 等同替换、 改进等, 均应包含在本公开的保护范围之内。

Claims

权利要求
1、 一种图像缩略图的生成方法, 其特征在于, 所述方法包括:
对图像进行滤波得到所述图像内各像素点的边缘强度值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据 其内像素点的边缘强度值计算该矩形框的信息量分布值;
选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到 所述图像的缩略图。
2、 根据权利要求 1所述的方法, 其特征在于, 所述对滑动搜索的各个位置上的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值, 包括:
对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求和得到该矩形框 的信息量分布值。
3、 根据权利要求 1所述的方法, 其特征在于, 所述以预先设定的矩形框在所述图像上滑 动搜索, 对滑动搜索的各个位置上的矩形框, 根据其内像素点的边缘强度值计算该矩形框的 信息量分布值, 包括- 使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计算所述图像内各 像素点的空间位置关注值;
使用预先根据边缘强度值和空间位置关注值建立的信息量分布模型, 计算所述图像内各 像素点的信息量分布值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 将该 矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分布值。
4、 根据权利要求 3所述的方法, 其特征在于, 所述将该矩形框内各像素点的信息量分布 值相加得到该矩形框的信息量分布值, 包括:
使用预先选取的核函数计算所述矩形框内各像素点对应的权重值;
将所述矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得到该矩 形框的信息量分布值;
其中, 所述核函数对距离图像中心点越近的像素点所取的权重值越大。
5、 根据权利要求 3所述的方法, 其特征在于, 所述使用预先根据所述图像的中心点和各 像素点的坐标建立的关注模型, 计算所述图像内各像素点的空间位置关注值, 包括:
使用如下关注模型计算所述图像内各像素点的空间位置关注值: (- (D )2 ) ; 其中, (i,j)表示所述图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值,
ε, Υε)表示所述图像的中心点, σ为预设的系数。
6、 根据权利要求 3所述的方法, 其特征在于, 所述使用预先根据边缘强度值和空间位置 关注值建立的信息量分布模型, 计算所述图像内各像素点的信息量分布值, 包括:
使用如下信息量分布模型计算所述图像内各像素点的信息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示所述图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j) 表示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
7、 根据权利要求 1至 6中任一项所述的方法, 其特征在于, 所述矩形框为正方形, 且边 长与所述图像短边的长度相等。
8、 根据权利要求 1至 6中任一项所述的方法, 其特征在于, 所述对图像进行滤波得到所 述图像内各像素点的边缘强度值, 包括:
对原始图像进行压缩,对压缩后的图像进行滤波得到所述图像内各像素点的边缘强度值; 所述将选取的所述矩形框对应的图像内容截取下来得到所述图像的缩略图, 包括: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始图像中的矩形框内的 图像内容截取下来得到所述原始图像的缩略图。
9、 一种图像缩略图的生成装置, 其特征在于, 所述装置包括:
滤波模块, 用于对图像进行滤波得到所述图像内每个像素点的边缘强度值;
搜索模块, 用于以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上 的矩形框, 根据其内像素点的边缘强度值计算该矩形框的信息量分布值;
截取模块, 用于选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内 容截取下来得到所述图像的缩略图。
10、 根据权利要求 9所述的装置, 其特征在于, 所述搜索模块包括:
搜索单元, 用于以预先设定的矩形框在所述图像上滑动搜索;
计算单元, 用于对滑动搜索的各个位置上的矩形框, 对其内所有像素点的边缘强度值求 和得到该矩形框的信息量分布值。
11、 根据权利要求 9所述的装置, 其特征在于, 所述搜索模块包括:
搜索单元, 用于以预先设定的矩形框在所述图像上滑动搜索;
计算单元, 用于使用预先根据所述图像的中心点和各像素点的坐标建立的关注模型, 计 算所述图像内各像素点的空间位置关注值; 使用预先根据边缘强度值和空间位置关注值建立 的信息量分布模型, 计算所述图像内各像素点的信息量分布值; 对所述搜索单元滑动搜索的 各个位置上的矩形框, 将该矩形框内各像素点的信息量分布值相加得到该矩形框的信息量分 布值。
12、 根据权利要求 11所述的装置, 其特征在于, 所述计算单元包括:
信息量分布值计算子单元, 用于使用预先选取的核函数计算所述矩形框内各像素点对应 的权重值; 将所述矩形框内各像素点的信息量分布值与对应的权重值进行乘积后再求和, 得 到该矩形框的信息量分布值;
其中, 所述核函数对距离图像中心点越近的像素点所取的权重值越大。
13、 根据权利要求 11所述的装置, 其特征在于, 所述计算单元包括:
空间位置关注值计算子单元, 用于使用如下关注模型计算所述图像内各像素点的空间位 置关注值:
Figure imgf000024_0001
其中, (i,j)表示所述图像内的任一个像素点, P(iJ)表示该像素点的空间位置关注值, (Χε, Υε)表示所述图像的中心点, σ为预设的系数。
14、 根据权利要求 11所述的装置, 其特征在于, 所述计算单元包括:
信息量分布值计算子单元, 用于使用如下信息量分布模型计算所述图像内各像素点的信 息量分布值:
7(iJ) = E(iJ) *P(iJ) ;
其中, (i,j)表示所述图像内的任一个像素点, /(i,j)表示该像素点的信息量分布值, £(i,j) 表示该像素点的边缘强度值, P(i,j)表示该像素点的空间位置关注值。
15、 根据权利要求 9至 14中任一项所述的装置, 其特征在于, 所述矩形框为正方形, 且 边长与所述图像短边的长度相等。
16、 根据权利要求 9至 14中任一项所述的装置, 其特征在于, 所述装置还包括: 压缩模块, 用于对原始图像进行压缩;
所述滤波模块用于: 对所述压缩模块压缩后的图像进行滤波得到所述图像内各像素点的 边缘强度值;
所述截取模块用于: 将选取的所述矩形框对应至所述原始图像中的矩形框, 将所述原始 图像中的矩形框内的图像内容截取下来得到所述原始图像的缩略图。
17、 一种终端, 其特征在于, 所述终端包括有存储器, 以及一个或者一个以上的程序, 其中一个或者一个以上程序存储于存储器中, 且经配置以由一个或者一个以上处理器执行所 述一个或者一个以上程序包含用于进行以下操作的指令:
对图像进行滤波得到所述图像内各像素点的边缘强度值;
以预先设定的矩形框在所述图像上滑动搜索, 对滑动搜索的各个位置上的矩形框, 根据 其内像素点的边缘强度值计算该矩形框的信息量分布值;
选取信息量分布值最大的矩形框, 并将选取的所述矩形框对应的图像内容截取下来得到 所述图像的缩略图。
PCT/CN2014/077610 2013-12-30 2014-05-15 图像缩略图的生成方法、装置和终端 WO2015100913A1 (zh)

Priority Applications (6)

Application Number Priority Date Filing Date Title
KR1020147021528A KR20150088970A (ko) 2013-12-30 2014-05-15 이미지 썸네일의 생성방법, 장치, 단말기, 프로그램 및 기록매체
MX2014009069A MX347781B (es) 2013-12-30 2014-05-15 Metodo, aparato y terminal para generar miniatura de imagen.
RU2015123209A RU2615679C2 (ru) 2013-12-30 2014-05-15 Способ, устройство и терминал для формирования миниатюры изображения
JP2015555587A JP2016511875A (ja) 2013-12-30 2014-05-15 画像サムネイルの生成方法、装置、端末、プログラム、及び記録媒体
BR112014018578A BR112014018578A8 (pt) 2013-12-30 2014-05-15 Método, aparelho e terminal para a geração de miniatura de imagem
US14/479,531 US9727972B2 (en) 2013-12-30 2014-09-08 Method and terminal for generating thumbnail of image

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310743545.7 2013-12-30
CN201310743545.7A CN103714161B (zh) 2013-12-30 2013-12-30 图像缩略图的生成方法、装置和终端

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/479,531 Continuation US9727972B2 (en) 2013-12-30 2014-09-08 Method and terminal for generating thumbnail of image

Publications (1)

Publication Number Publication Date
WO2015100913A1 true WO2015100913A1 (zh) 2015-07-09

Family

ID=50407136

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/077610 WO2015100913A1 (zh) 2013-12-30 2014-05-15 图像缩略图的生成方法、装置和终端

Country Status (8)

Country Link
EP (1) EP2890105B1 (zh)
JP (1) JP2016511875A (zh)
KR (1) KR20150088970A (zh)
CN (1) CN103714161B (zh)
BR (1) BR112014018578A8 (zh)
MX (1) MX347781B (zh)
RU (1) RU2615679C2 (zh)
WO (1) WO2015100913A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111563495A (zh) * 2020-05-09 2020-08-21 北京奇艺世纪科技有限公司 一种图像中字符的识别方法、装置及电子设备
CN112037127A (zh) * 2020-07-27 2020-12-04 浙江大华技术股份有限公司 视频监控的隐私遮挡方法及装置、存储介质、电子装置

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9727972B2 (en) 2013-12-30 2017-08-08 Xiaomi Inc. Method and terminal for generating thumbnail of image
CN103714161B (zh) * 2013-12-30 2017-06-16 小米科技有限责任公司 图像缩略图的生成方法、装置和终端
CN103903223B (zh) * 2014-04-24 2017-03-01 厦门美图之家科技有限公司 一种基于图像显著性检测的获取缩略图的方法
CN104461280A (zh) * 2014-11-03 2015-03-25 联想(北京)有限公司 一种信息处理方法及电子设备
CN104346772B (zh) * 2014-11-06 2018-06-05 杭州华为数字技术有限公司 缩略图制作方法和装置
CN105426047B (zh) * 2015-11-17 2018-09-18 小米科技有限责任公司 缩略图展示方法和装置
CN106126108B (zh) * 2016-06-30 2019-04-12 维沃移动通信有限公司 一种缩略图的生成方法及移动终端
US10284861B2 (en) * 2016-12-09 2019-05-07 Advanced Micro Devices, Inc. Concurrent image compression and thumbnail generation
CN110134651B (zh) * 2019-05-09 2021-10-26 北京金山安全软件有限公司 信息文件的处理方法、装置、电子设备和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007207055A (ja) * 2006-02-03 2007-08-16 Seiko Epson Corp 画像処理装置、プログラム、情報記憶媒体及び画像処理方法
US20100158409A1 (en) * 2008-12-19 2010-06-24 Prabindh Sundareson Method to Generate Object Focused Thumbnails for Digital Images
WO2013069341A1 (ja) * 2011-11-09 2013-05-16 楽天株式会社 画像処理装置、画像処理装置の制御方法、プログラム、及び情報記憶媒体
JP2013168902A (ja) * 2012-02-17 2013-08-29 Seiko Epson Corp 画像処理装置、表示装置および画像処理方法
CN103714161A (zh) * 2013-12-30 2014-04-09 小米科技有限责任公司 图像缩略图的生成方法、装置和终端

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004295776A (ja) * 2003-03-28 2004-10-21 Minolta Co Ltd 画像認識装置および画像認識プログラム
RU2400815C2 (ru) * 2006-10-09 2010-09-27 Самсунг Электроникс Ко., Лтд. Способ повышения качества цифрового фотоизображения
JP5042251B2 (ja) * 2009-02-18 2012-10-03 三菱電機株式会社 画像処理装置および画像処理方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007207055A (ja) * 2006-02-03 2007-08-16 Seiko Epson Corp 画像処理装置、プログラム、情報記憶媒体及び画像処理方法
US20100158409A1 (en) * 2008-12-19 2010-06-24 Prabindh Sundareson Method to Generate Object Focused Thumbnails for Digital Images
WO2013069341A1 (ja) * 2011-11-09 2013-05-16 楽天株式会社 画像処理装置、画像処理装置の制御方法、プログラム、及び情報記憶媒体
JP2013168902A (ja) * 2012-02-17 2013-08-29 Seiko Epson Corp 画像処理装置、表示装置および画像処理方法
CN103714161A (zh) * 2013-12-30 2014-04-09 小米科技有限责任公司 图像缩略图的生成方法、装置和终端

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ARAI, K. ET AL.: "Visual-Attention-based Thumbnail Using Two-Stage GrabCut", INTERNATIONAL CONFERENCE ON MULTIMEDIA COMPUTING AND SYSTEMS (ICMCS, 12 May 2012 (2012-05-12), pages 96 - 101, XP032259013, DOI: doi:10.1109/ICMCS.2012.6320153 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111563495A (zh) * 2020-05-09 2020-08-21 北京奇艺世纪科技有限公司 一种图像中字符的识别方法、装置及电子设备
CN111563495B (zh) * 2020-05-09 2023-10-27 北京奇艺世纪科技有限公司 一种图像中字符的识别方法、装置及电子设备
CN112037127A (zh) * 2020-07-27 2020-12-04 浙江大华技术股份有限公司 视频监控的隐私遮挡方法及装置、存储介质、电子装置

Also Published As

Publication number Publication date
BR112014018578A2 (zh) 2017-06-20
KR20150088970A (ko) 2015-08-04
JP2016511875A (ja) 2016-04-21
RU2615679C2 (ru) 2017-04-06
BR112014018578A8 (pt) 2017-07-11
EP2890105A1 (en) 2015-07-01
EP2890105B1 (en) 2017-03-15
MX2014009069A (es) 2015-10-30
RU2015123209A (ru) 2017-01-10
CN103714161B (zh) 2017-06-16
MX347781B (es) 2017-05-12
CN103714161A (zh) 2014-04-09

Similar Documents

Publication Publication Date Title
WO2015100913A1 (zh) 图像缩略图的生成方法、装置和终端
WO2018103525A1 (zh) 人脸关键点跟踪方法和装置、存储介质
WO2017206916A1 (zh) 处理器中内核运行配置的确定方法以及相关产品
CN111464716B (zh) 一种证件扫描方法、装置、设备及存储介质
US20160112632A1 (en) Method and terminal for acquiring panoramic image
US11843715B2 (en) Photographing method and terminal
CN107241552B (zh) 一种图像获取方法、装置、存储介质和终端
CN107784089B (zh) 一种多媒体数据的存储方法、处理方法及移动终端
CN108256853B (zh) 一种支付方法和移动终端
CN108415641B (zh) 一种图标的处理方法及移动终端
CN109857297B (zh) 信息处理方法及终端设备
CN107749046B (zh) 一种图像处理方法及移动终端
WO2017050090A1 (zh) 生成gif文件的方法、设备及计算机可读存储介质
CN109753202B (zh) 一种截屏方法和移动终端
CN106791916B (zh) 一种推荐音频数据的方法、装置和系统
CN108718389B (zh) 一种拍摄模式选择方法及移动终端
CN105635553B (zh) 一种图像拍摄方法和装置
JP6862564B2 (ja) 画像合成のための方法、装置および不揮発性コンピュータ可読媒体
WO2019120190A1 (zh) 拨号方法及移动终端
CN107728877B (zh) 一种应用推荐方法及移动终端
CN108881721A (zh) 一种显示方法及终端
CN110007821B (zh) 一种操作方法及终端设备
CN109491964B (zh) 一种文件分享方法及终端
CN105513098B (zh) 一种图像处理的方法和装置
WO2021104162A1 (zh) 显示方法及电子设备

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: MX/A/2014/009069

Country of ref document: MX

WWE Wipo information: entry into national phase

Ref document number: 1020147021528

Country of ref document: KR

ENP Entry into the national phase

Ref document number: 2015555587

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 2015123209

Country of ref document: RU

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112014018578

Country of ref document: BR

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

Ref document number: 14876340

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14876340

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 112014018578

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20140728