KR101785982B1 - Method and apparatus for generating mosaic image - Google Patents

Method and apparatus for generating mosaic image Download PDF

Info

Publication number
KR101785982B1
KR101785982B1 KR1020167027049A KR20167027049A KR101785982B1 KR 101785982 B1 KR101785982 B1 KR 101785982B1 KR 1020167027049 A KR1020167027049 A KR 1020167027049A KR 20167027049 A KR20167027049 A KR 20167027049A KR 101785982 B1 KR101785982 B1 KR 101785982B1
Authority
KR
South Korea
Prior art keywords
image
feature vector
storage
vector table
tile
Prior art date
Application number
KR1020167027049A
Other languages
Korean (ko)
Other versions
KR20160130792A (en
Inventor
롱준 지안
시안펭 랑
리 리우
Original Assignee
바이두 온라인 네트웍 테크놀러지 (베이징) 캄파니 리미티드
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 바이두 온라인 네트웍 테크놀러지 (베이징) 캄파니 리미티드 filed Critical 바이두 온라인 네트웍 테크놀러지 (베이징) 캄파니 리미티드
Publication of KR20160130792A publication Critical patent/KR20160130792A/en
Application granted granted Critical
Publication of KR101785982B1 publication Critical patent/KR101785982B1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06T3/40—Scaling the whole image or part thereof
    • G06T3/4038—Scaling the whole image or part thereof for image mosaicing, i.e. plane images composed of plane sub-images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformation in the plane of the image
    • G06F17/30247—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20021—Dividing image into blocks, subimages or windows

Abstract

The present application discloses a method for generating a mosaic image, wherein one specific embodiment of the method comprises the steps of: dividing a target image into image tiles of a plurality of predetermined sizes; extracting image tile feature vectors for each image tile; Acquiring a feature vector table corresponding to a material library; acquiring a storage position in a feature vector table of the material feature vector matched for each image tile feature vector; Obtaining a storage address of a corresponding image material based on the storage location based on a correspondence relationship between the material feature vectors and acquiring a corresponding image material and generating a mosaic image based on each storage address . The above embodiment improves the generation efficiency of the mosaic image.

Description

[0001] METHOD AND APPARATUS FOR GENERATING MOSAIC IMAGE [0002]

The present invention relates to the field of computer technology, and more specifically to the field of electronic image processing technology, and in particular to a method and apparatus for generating mosaic images.

The present application claims priority of Chinese patent application filed on March 31, 2015 and having the filing number "201510149873.3 ", all contents of said Chinese patent application being incorporated herein by reference in its entirety.

As Internet technology developed, digital entertainment products began appearing around. Digital entertainment products refer to entertainment products based on digital technologies such as animation, cartoons, network games, and mosaic images. Here, a mosaic image refers to an image formed by mosaicing a plurality of small image tiles. There are many ways of mosaicking existing mosaic images in the existing technology. One type of image can be mosaicked into another type of image, for example, You can mosaic a wide range of landscape images.

In the process of generating an existing mosaic image, the target image is usually divided into a plurality of image tiles, and these image tiles must be matched to the image materials in the material library, respectively. There is a problem that the generation efficiency of the mosaic image is deteriorated due to overlapping matching in the generation process.

SUMMARY OF THE INVENTION The present invention has been made to solve one or a plurality of technical problems mentioned in the background section, and it is an object of the present invention to provide an improved mosaic image generating method and apparatus.

In a first aspect, the present application provides a method of generating a mosaic image, the method comprising: dividing a target image into image tiles of a plurality of predetermined sizes; extracting image tile feature vectors for each image tile; Obtaining a feature vector table for storing a material feature vector of each image material in the material library corresponding to a material library; calculating, for each image tile feature vector, a feature vector Acquiring a storage location of a corresponding image material based on the storage location based on a correspondence relationship between the image material and the material feature vector; And acquiring a corresponding image material and generating a mosaic image.

In some embodiments, obtaining the storage location in the feature vector table of the matched texture feature vector for each image tile feature vector comprises obtaining an index structure of the feature vector table, And to provide a storage location in the feature vector table of search order and material feature vectors when searching for each material feature vector; and for each image tile feature vector, Retrieving a matching feature vector from the index structure, and obtaining a storage location in the feature vector table of each matched texture feature vector from the index structure.

In some embodiments, the index structure may include a step of dividing the feature vector table according to a neighborhood relationship of each material feature vector and creating a tree structure, and a step of dividing the feature vector table into a storage location in the feature vector table of each material feature vector And generating an index structure of the feature vector table by replacing each corresponding feature feature vector in the tree structure.

In some embodiments, obtaining a storage address of a corresponding image material based on the storage location based on a correspondence relationship between the image material and the material feature vector may include: Acquiring a mapping relationship between a storage address in the feature vector table and a storage location in a feature vector table of a material feature vector corresponding to the image material based on the mapping relationship; And obtaining a storage address in the storage file of the material library.

In some embodiments, the step of acquiring a corresponding image material and generating a mosaic image based on each of the storage addresses comprises the steps of: reading the storage file; storing the corresponding image in the storage file, And a step of generating a mosaic image by replacing each corresponding image tile in the target image with each acquired image material.

In some embodiments, prior to dividing the target image into a plurality of predetermined sized image tiles, obtaining an image includes determining whether a pixel size of the image falls within a predetermined pixel size range, And adjusting the pixel size of the image to be within the predetermined pixel size range as the target image if the pixel size of the image does not fall within the predetermined pixel size range.

In some embodiments, the method further comprises: obtaining an original image; and modifying the original image to an image of a predetermined size as an image material, wherein the modification includes at least one of image region selection and pixel compression .

In some embodiments, the step of modifying the original image into an image of a predetermined size and treating it as an image material may include matching the original image to a spherical model first, and then backprojecting the same to a plane And modifying the original image that was projected back to the plane into an image of a predetermined size as an image material.

In some embodiments, the method further comprises categorizing and storing the image material into a plurality of material libraries such that the user selects the material library.

In a second aspect, the present application provides a mosaic image generating device, comprising: a partitioning module configured to partition a target image into a plurality of image tiles of predetermined size; An extraction module configured to extract an image tile feature vector for each image tile; A vector table acquisition module configured to acquire a feature vector table for storing a material feature vector of each image material in the material library corresponding to the material library; A location acquiring module configured to acquire, for each image tile feature vector, a storage location in the feature vector table of the matching feature feature vector; An address acquisition module configured to acquire a storage address of a corresponding image material based on the storage location, based on a correspondence relationship between the image material and the material feature vector; And a generation module configured to acquire a corresponding image material and generate a mosaic image based on each storage address.

In some embodiments, the location acquisition module is configured to obtain an index structure of the feature vector table, wherein when the search is performed for each material feature vector, the search sequence and the feature vector of the material feature vector An index structure acquiring unit configured to provide a storage location in a table; A search unit configured to search, for each image tile feature vector, a feature feature vector matching in the feature vector table according to the index structure; And a storage location acquiring unit configured to acquire a storage location in the feature vector table of each matched texture feature vector from the index structure.

In some embodiments, the position acquiring module divides the feature vector table according to a neighborhood relationship of each material feature vector, generates a tree structure, and stores the position of the material feature vector in the feature vector table And an index structure generation unit configured to generate an index structure of the feature vector table by replacing each corresponding material feature vector in the tree structure.

In some embodiments, the address acquisition module acquires a mapping relationship between a storage address of the image material in the storage file of the material library and a storage location in the feature vector table of the material feature vector corresponding to the image material A configured mapping relationship acquisition unit; And a storage address acquisition unit configured to acquire a storage address in a storage file of the material library of the corresponding image material based on the storage location based on the mapping relationship.

In some embodiments, the generating module comprises: a reading unit configured to read the storage file; A material acquiring unit configured to acquire corresponding image material in the storage file according to each storage address; And an alternative unit configured to generate a mosaic image by replacing each corresponding image tile in the target image with each acquired image material.

In some embodiments, further comprising a selection module, the selection module comprising: an image acquisition unit configured to acquire an image; A pixel size determination unit configured to determine whether a pixel size of the image falls within a predetermined pixel size range; And a pixel size adjustment unit configured to adjust the pixel size of the image to within a predetermined pixel size range as a target image when the pixel size of the image is not within the predetermined pixel size range.

In some embodiments, the apparatus further comprises an image material generation module, wherein the image material generation module comprises: an original image acquisition unit configured to acquire an original image; And an image modification unit configured to modify the original image to a predetermined size as an image material, the modification including at least one of image region selection and pixel compression.

In some embodiments, the image material creation module comprises: an image conversion unit configured to map the original image first to a spherical model and then to project back onto the plane again; And a conversion image modification unit configured to modify the original image that is projected back to the plane to a predetermined size as an image material.

The mosaic image generating method and apparatus provided in the present application can match feature vectors of image tiles divided from a target image to material feature vectors in a feature vector table of a material library, Acquires the storage location of the corresponding image material based on the storage location based on the correspondence relationship between the image material and the material feature vector, , A corresponding image material is acquired and a mosaic image is generated. The method and apparatus for generating a mosaic image provided in the present application improves the efficiency of generating a mosaic image.

Other features, objects, and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, which proceeds with reference to the following drawings.
1 is a flow diagram of one embodiment of a method for generating a mosaic image according to the present application.
Figure 2 is a flow diagram of one embodiment for obtaining a storage location of a material feature vector matching an image tile feature vector in a feature vector table according to an index structure according to the present application.
Figure 3 is an exemplary screen shot of geographic information points considered as image material according to the present application.
4 is a flowchart of another embodiment of a mosaic image generating method according to the present application.
5A and 5B are schematic diagrams of an original image and a mosaic image of an embodiment of a mosaic image generating method according to the present application, respectively.
6 is a structural schematic diagram of an embodiment of a mosaic image generating device according to the present application.
7 is a structural schematic diagram of a computer system for generating a mosaic image in accordance with the present application.

In the following, a more detailed description of the present application will be made in connection with the accompanying drawings and embodiments. It will be understood that the specific embodiments described herein are for interpretation of the relevant invention only and are not intended to limit the invention in any way. For convenience of explanation, only the parts related to the invention are shown in the attached drawings.

It is to be understood that the features of the embodiments and the examples of the present application can be combined with each other unless they are contradictory. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the present embodiments of the invention, examples of which are illustrated in the accompanying drawings.

Referring to FIG. 1, FIG. 1 illustrates a flow 100 of a mosaic image generation method. This embodiment will mainly be described by taking as an example that the method is applied to an electronic apparatus including a display screen, and the electronic apparatus is a smart phone, a tablet computer, an electronic book terminal, an MP3 player (Moving Picture Experts Group Audio Layer III; MPEG audio layer 3), MP4 (Moving Picture Experts Group Audio Layer IV) player, notebook type computer, and desktop computer. The method for generating a mosaic image according to the present application includes the following steps.

In step S101, the target image is divided into a plurality of predetermined-sized image tiles.

In this embodiment, an electronic device (e.g., a smartphone on which an application of the type generating a mosaic image is run) may divide the target image to be produced as a mosaic image into a plurality of predetermined sized image tiles . Here, the electronic device may acquire a target image to be produced as a mosaic image locally or remotely, or may take a single picture as a target image. The target image may be divided into image tiles of a predetermined size in units of pixels, and may be divided into image tiles of, for example, 32x32 pixels. Here, the predetermined size indicates that the image tiles acquired by dividing the target image have the same pixel size as the image material used for generating the mosaic image.

In an alternative implementation of this embodiment, the electronic device may send the destination image to a remote server (e.g., a back-end server that provides support for an application of the type generating a mosaic image on an electronic device). The remote server may partition the target image into a number of image tiles of predetermined size. The remote server may be a single server or a server cluster connected through a network. At this time, the electronic device can receive such information from a remote server through a wired or wireless connection scheme. The wireless connection method may include a 3G / 4G connection, a WiFi connection, a Bluetooth connection, a WiMAX connection, a Zigbee connection, an UWB (ultra wideband) connection and other already known or future developed wireless connection methods. It does not.

In step S102, an image tile feature vector is extracted for each image tile.

In this embodiment, the electronic device can extract the image tile feature vectors for each of the divided image tiles. Here, the image tile feature vector may be a vector representing one or more features (e.g., color features, etc.) of the image tile and may be, for example, a vector of color values of a plurality of feature points .

As an example of an image tile feature vector, the image tile feature vector may represent only the color feature of the image tile. For example, in a conventional display device, colors are implemented by using an RGB color standard, that is, by emitting an electron gun to three color light emitting electrodes of red (R), green (G), and blue (B) . Here, the RGB color standard is to acquire various colors through a change of three color channels of red (R), green (G), and blue (B) and superimposition among them. Accordingly, the color value of the image tile can be expressed by the numerical value (color value) of the characteristic point on the three color channels of R, G, B. Specifically, for each image tile, the electronic device first selects a number of points, e. G., 25 points, as feature points on the image, and then, on each of the three color channels R, G, Finally, 25 × 3 numerical values are arranged in order to obtain a 25 × 3 dimensional vector, ie, an image tile feature vector. Here, as feature points, 25 points may be selected in accordance with the default rule. For example, if the image tile has 32 x 32 pixel points, a total of 5 rows can be selected by 5 points for each row to form a single 5 x 5 point array. Twenty-five feature points may be selected in the form of " x " at 32 x 32 pixel points, or may be selected in other forms, and the present invention is not limited to this. Alternatively, in the 25x3 dimensional vector, the order of the numerical values of each feature point on the R, G, B three color channels may vary. For example, a vector of 25 x 3 dimensions is first a number of each feature point on the R color channel, then a number of each feature point on the G color channel, and finally a number of each feature point on the B color channel Lt; / RTI > A vector of 25 x 3 dimensions may be arranged in the order of one feature point. That is, it can be a value on three color channels of R, G, B of one feature point, and then a value on three color channels of R, G, B of the next feature point. In practice, the 25x3 vector can set the order of the values according to the actual application, and the present application does not limit it.

Optionally, extracting the image tile feature vector for each image tile may be performed by a remote server.

In step S103, a feature vector table corresponding to the material library is acquired.

In this embodiment, the electronic device obtains a feature vector table corresponding to a default or selected material library from a local or remote server. Here, the material library may be configured to store relevant data of the material image (e.g., a storage file of the material image, etc.). Here, the material library may be a default material library or a single material library selected from a plurality of selectable material libraries. Alternatively, when there are a plurality of material libraries, they can be classified and stored according to different classification methods. For example, many of the material libraries may be libraries of Street View image creatives stored according to a geographic area (e.g., Beijing, Shanghai, etc.), or a portrait image creative library stored by gender (e.g., Or a library of landscape image material stored according to different seasons (for example, spring, winter, etc.). Alternatively, a plurality of material libraries may be simultaneously categorized and stored according to a plurality of classification methods, for example, a landscape image library (hereinafter referred to as " landscape image library ") classified and stored according to different seasons And so on. The present application does not proceed to the limitation thereof.

In this embodiment, the feature vector table is configured to store the material feature vectors of each image material in the corresponding material library. The feature vector table may be pre-stored in a local or remote server of the electronic device. Here, the material feature vector may be a vector representing one or more features (e.g., color features, etc.) of the image material, and may be, for example, a vector of color values of a plurality of feature points. Those skilled in the art will understand that for convenience of matching, the material feature vector of each image material has the same display method as the feature vector of the image tile above. For example, select 25 points as feature points on the image material, then obtain the values on the three color channels R, G, B for each feature point, and finally, To obtain a vector of 25 × 3 dimensions, that is, a feature vector of the image material. Here, the feature vector representing the image material and the feature vector representing the image tile may have the same arrangement order for the values of the feature points on the R, G, B color channels.

Optionally, the step of acquiring the feature vector table corresponding to the material library may be performed by the remote server.

In step S104, for each image tile feature vector, a storage location in the feature vector table of the matched material feature vector is obtained.

In this embodiment, for each image tile feature vector, the electronic device proceeds vector matching based on the material feature vector in the feature vector table obtained in step S103, and the feature vector table of the matched feature material vector Lt; / RTI > Here, the electronic device can determine whether the material feature vector and the image tile feature vector match each other according to the matching degree of the feature vector. For example, the degree of matching may be expressed as a variance of a numerical value corresponding to a material feature vector and an image tile feature vector. As the variance increases, the degree of matching is lower and the degree of matching is higher as the variance decreases. The degree of matching may be expressed as a weight percentage. For example, the weight percentage can be obtained by the following method. That is, assuming that the number of dimensions of the vector is 75, the weight of each dimension is 1/75, the weight obtained by dividing the value of each dimension in the material feature vector by the value of the dimension corresponding to the image tile feature vector, 75 as a weighting factor for the dimension, and adding the weighted fractions of the respective dimensions to each other, the material feature vector obtains a weighted percentage value for the image tile feature vector. The larger the value of the weighting percentage, the higher the degree of matching, and the smaller the weighting value, the lower the degree of matching. The degree of matching may be displayed by other methods. The present application does not proceed to the limitation thereof. Those skilled in the art will recognize that a material feature vector whose electronic device is matched to an image tile feature vector selects a feature feature vector having the highest degree of matching with an image tile feature vector, Selecting a feature feature vector that is larger than the threshold value (e.g., having a weight percentage of 80% or more), and the like.

In the present embodiment, the electronic device sequentially performs matching according to the order of the material feature vectors stored in the vector table, or performs the matching according to the search order provided in the index structure (e.g., a tree-shaped index structure) Matching can be performed, or vector matching can be performed by performing matching in other order. The present application does not proceed to the limitation thereof. As an example, FIG. 2 shows a flow 200 of one embodiment for obtaining a storage location of a material feature vector that matches an image tile feature vector in a feature vector table according to an index structure. The specific steps of the flow 200 are as follows.

In step S201, the index structure of the feature vector table is obtained.

In this embodiment, the electronic device can obtain the index structure of the feature vector table locally or remotely. Here, the index structure may be a tree structure or a table structure. When searching for each material feature vector, a search sequence and a material feature vector storage location are provided in the feature vector table. When the index structure is a tree structure, if the design of the tree structure is reasonable, after matching is completed at one node of the tree, matching is succeeded, or matching is performed by discarding one of left and right dependent trees of the node The speed can be remarkably improved.

In step S202, for each image tile feature vector, the feature feature vector matching in the feature vector table is retrieved according to the index structure.

In this embodiment, for each image tile feature vector, the electronic device or remote server retrieves a material feature vector matching the image tile feature vector in the feature vector table according to the search order provided in the index structure.

In step S203, a storage location in the feature vector table of each material feature vector matched from the index structure is obtained.

In this embodiment, the electronic device or the remote server obtains the storage location in the feature vector table of each material feature vector matched from the index structure according to the matching result in step S202. Here, the storage position may be a logical address of a material feature vector (for example, a storage order of feature vectors in the feature vector table), a physical address of a material feature vector (a physical address of a physical memory space in which a feature vector is located ). The storage address has a constant mapping relationship with the material feature vector. Alternatively, the mapping relationship may be such that the stored address is stored in a feature vector table in the form of a subscript of the index of the feature vector.

In some optional implementations of this embodiment, the index structure may be a tree structure. The index structure of the tree structure type can be generated through the following steps. Next, the tree structure is generated by dividing the feature vector table according to the neighborhood relation of each material feature vector. Subsequently, each material feature vector corresponding to the tree structure is replaced with the storage position in the feature vector table of each material feature vector, Creates the index structure of the vector table. Here, in the tree structure divided according to the neighborhood relationship of the material feature vectors in the feature vector table, the material feature vectors may be distributed according to the recent lean relationship. On the other hand, the data actually stored in the node of the tree structure may be a storage location in the feature vector table of the material feature vector so as to reduce the amount of data in the tree structure and simplify the tree structure. The electronic device or the remote server can read the corresponding material feature vector from the storage position in the tree structure node according to the relationship between the storage position and the material feature vector in the feature vector table. For example, in the matching process of the actual image tile feature vector, the electronic device or the remote server performs the matching according to the order of the tree structure. For example, first, the root node of the tree structure is matched, the electronic device or the remote server reads the storage location at the root node, reads the corresponding material feature vector according to the mapping relationship between the storage location and the material feature vector The comparison is made with the image tile feature vector, and it is determined whether the material feature vector and the image tile feature vector match according to the matching degree of the feature feature vector and the image tile feature vector. Here, the degree of matching represents the degree of matching between the material feature vector and the image tile feature vector. When the matching degree is greater than the first threshold value, it is determined that the material feature vector corresponding to the root node and the image tile feature vector are matched with each other, and the storage position at the root node is obtained. If the degree of matching is smaller than the second threshold value, matching is continued for the left dependent tree of the tree structure, and the result is inferred in this manner .

In some optional implementations of this embodiment, the index structure may be a k-dimensional tree (k-dimensional tree) structure. Here, the k-d tree is a data structure that divides the k-dimensional data space. The k-d tree is a binary tree where each node is a k-dimensional point. All non-leaf nodes of the k-d tree can be considered as dividing the space into two parts with one hyperplane. Here, a point located on the left side of the hyperplane represents a left dependent tree of the non-leaf node, and a point located on the right side of the hyperplane represents a right dependent tree of the non-leaf node. The direction of the hyperplane can be selected according to the relationship that each node is all related to a dimension perpendicular to the hyperplane in the k dimension. For example, if you choose to split along the x axis, all nodes with an x value less than the specified value will appear in the left dependent tree, and all nodes with an x value greater than the specified value will appear in the right dependent tree. the partition space of the index structure of the k-d tree does not overlap, and the speed of processing a large amount of data is relatively fast.

In step S105, based on the correspondence relationship between the image material and the material feature vector, a storage address of the corresponding image material is obtained based on the storage location.

In the present embodiment, based on the correspondence relationship between the image material and the material feature vector, the electronic device or the remote server can acquire the storage address of the corresponding image material based on the storage location. Here, the storage address of the image material may be the storage order number or physical address (physical address) of the image material in the material library. Those skilled in the art will recognize that one-to-one correspondence or mapping relationship exists between the storage location and the corresponding storage location of the image material since the material feature vector is a feature vector of the image material and there is a one- Lt; / RTI > Alternatively, the image material may have the same storage order as the corresponding material feature vector.

In some implementations of this embodiment, the electronic device or the remote server first obtains the mapping relationship between the storage address in the storage file of the material library of the image material and the storage location in the feature vector table of the material feature vector corresponding to the image material And based on the mapping relationship, obtains a storage address in the storage file of the material library of the corresponding image material based on the storage location. Here, the mapping relationship may be stored in advance in a storage device of an electronic device or a remote server. The storage file is a format in which all image materials in the material library are present on the storage device of the electronic device or the remote server. Alternatively, the storage file may be a compressed format (e.g., jpg format) of the image, a binary format of the color value of the pixel point, and the like, and the present invention does not limit the present invention. For example, the save file may be a binary file consisting of the numerical values of each pixel point of the image material on the three color channels R, G, B. At this time, when the binary file is acquired, the image material can be displayed according to the numerical value of each pixel point of the image material on the three color channels of R, G, and B colors.

In step S106, based on each storage address, a corresponding image material is acquired and a mosaic image is generated.

In this embodiment, the electronic device or the remote server obtains the corresponding image material on the basis of each storage address acquired in step S105, replaces the image tile corresponding to the acquired image material, Images can be generated.

In some implementations of this embodiment, the electronic device or remote server reads the storage file of the material library from the storage device, and then obtains the corresponding image material from the storage file, respectively, based on each storage address, A mosaic image can be generated by replacing each corresponding image tile in the target image with each image material. Alternatively, the electronic device or the remote server may detect whether the image tiles in the target image have been completely replaced and, if not completely replaced, return to step S102 to extract the image tile feature vectors of the image tiles that have not been replaced , Vector matching can be performed to obtain a corresponding material image.

In some optional implementations of this embodiment, the image material may be obtained by first acquiring the original image to an electronic device or remote server, and then modifying the original image to a predetermined size image to be considered as the image material . Here, the modification includes at least one of image region selection and pixel compression, but is not limited thereto. Here, the original image may be any image such as, for example, a portrait image, a landscape image, a street view image, or the like. The predetermined size of the image is a pixel value (for example, 32 x 32 pixels) of the preset image material. Alternatively, the pixel value may be an empirical value, a pixel value obtained by sample set training based on the pixel value and the effect of generating the generated mosaic image, and the like, but the present application does not limit it. If the size of the original image is larger than the predetermined size, the electronic device or the remote server can correct the original image and obtain the image material of the predetermined size. Alternatively, the electronic device or the remote server may select a certain region of the original image, for example, a head image portion of the portrait image, or the like. The electronic device or remote server may proceed with pixel compression of the original image or the image of the selected region portion to obtain the image material of the predetermined size. In practice, if the original image is a Street View image, the geographic information points in the Street View image (one geographic information point is one house, one store, one land Mark, etc.). ≪ / RTI > As shown in FIG. 3, FIG. 3 shows compressed image materials after selecting geographic information point regions as original images from some sightseeing image images of Beijing.

In some optional implementations of this embodiment, the electronic device or the remote server may proceed with the following processing of the original image. First, the original image is mapped to the spherical model and then back-projected to the plane. Next, the original image that is projected back to the plane is corrected to the image of the predetermined size to be regarded as the image material. Those skilled in the art will recognize that if the original image is a street view image on an electronic map, then the original image may be an image including a fisheye view, and the size of the fisheye is very large (For example, 8192 × 4096 pixels) and directly compressed to a predetermined size (for example, 32 × 32 pixels), the details of the image can be omitted. In addition, Time) on a plane, and if the image is directly compressed, edge distortion occurs and there is a possibility of affecting the generation effect of the mosaic image. Therefore, for an image including the fisheye, the image is first mapped to the three-dimensional spherical model to restore the image at the fisheye time, and then the image is further projected back to the plane to form an image of 180 degrees. Can be removed.

In the present embodiment, the application can be selected by a user by selecting an image (by inputting a search condition to acquire an image such as a portrait image from a remote server, for example) and selecting a material library City), then the electronic device or remote server may acquire the image, match the image material in the material library, and replace the image tile to form and display the mosaic image. The electronic device or the remote server may divide the image into image tiles and obtain the storage address of each image material whose feature vector matches the image tile based on the feature vector table of the material library and the index structure of the feature vector table After reading the storage file of the material library, the corresponding image material is sequentially acquired to generate a mosaic image. The method provided in the above embodiment of the present application aids in improving the efficiency of generating a mosaic image.

The method of generating a mosaic image provided in this embodiment is merely exemplary, and the steps in the flow 100 do not include a temporal relationship in time. Some of the steps can be tailored to demand in real applications. The change of the step sequence does not affect the execution result of the mosaic image generating method of the present application. For example, step 103 and step 102 may be executed simultaneously or in reverse order. The present application does not proceed to the limitation thereof.

Referring to FIG. 4, FIG. 4 is a flow 400 of another embodiment of a method for generating a mosaic image according to the present application. The flow 400 includes the following steps.

In step S401, an image is acquired.

In this embodiment, the electronic device can first acquire one image locally or remotely, or take one picture as an image. Wherein the image may be an image having any content and size.

In step S402, it is determined whether the pixel size of the image falls within the predetermined pixel size range.

In this embodiment, the electronic device determines whether the pixel size of the image obtained in step S401 falls within the predetermined pixel size range. The electronic device may upload the image obtained in step S401 to the remote server. After obtaining the image, the remote server determines whether the pixel size of the image falls within the predetermined pixel size range. Those skilled in the art will appreciate that if the pixel size of the image acquired by the electronic device or the remote server is too small, the effect of generating the mosaic image will be greatly affected. For example, assuming that the pixel size of the image is 96 × 96 pixels and that the predetermined size of the image tile is 32 × 32 pixels when dividing the image, it is inevitably divided into 3 × 3 image tiles. When you create a mosaic image of a human face using Street View image mosaic, you may not notice that it is a human face at all. Thus, the electronic device or the remote server can first determine the pixel size of the image.

In the present embodiment, the predetermined pixel size range may be in a relatively preferable pixel size range for generating a mosaic image, for example, 512 × 512 pixels to 8192 × 4096 pixels. The predetermined pixel size range may be obtained by experience or by training the sample set. The present application does not proceed to the limitation thereof.

In step S403, if the pixel size of the image does not fall within the predetermined pixel size range, the pixel size of the image is adjusted to be within the predetermined pixel size range as the target image.

In this embodiment, if the pixel size of the image does not fall within the predetermined pixel size range, the electronic device or the remote server may adjust the pixel size of the image to within the predetermined pixel size range to regard the image as the target image . Here, if the pixel size of the image is smaller than the minimum pixel value in the pixel size range, the electronic device or the remote server adjusts the image, such as tension, width, or height, and then regards the image as a target image. If the maximum pixel value in the pixel size range is larger than the maximum pixel value in the pixel size range, it is regarded as the target image after adjusting the pixel compression and the segmentation of the image.

In step S404, the target image is divided into a plurality of predetermined-sized image tiles.

In this embodiment, the electronic device or the remote server may divide the target image to be produced as a mosaic image into a plurality of image tiles of predetermined sizes. Here, the predetermined size indicates that the divided image tiles have the same pixel size as the image material used for creating the mosaic image.

In step S405, an image tile feature vector is extracted for each image tile.

In this embodiment, the electronic device can extract the image tile feature vectors for each of the divided image tiles. Here, the image tile feature vector may be a vector representing one or more features (e.g., color features, etc.) of the image tile and may be, for example, a vector of color values of a plurality of feature points .

In step S406, a feature vector table corresponding to the material library is acquired.

In this embodiment, the electronic device obtains a feature vector table corresponding to a default or selected material library from a local or remote server. Here, the material library can be configured to store the relevant data of the material image (which may be, for example, a storage file of the material image, etc.). In this embodiment, the feature vector table is configured to store the material feature vectors of each image material in the corresponding material library. The feature vector table may be pre-stored on an electronic device local or remote server. Here, the material feature vector may have the same extraction method and display method as the image tile feature vector.

In step S407, for each image tile feature vector, a storage location in the feature vector table of the matched source feature vector is obtained.

In this embodiment, for each image tile feature vector, the electronic device or remote server proceeds with vector matching based on the material feature vector in the feature vector table obtained in step S407, And acquires the storage position in the feature vector table. Here, the electronic device can determine whether the material feature vector and the image tile feature vector match each other according to the matching degree of the feature vector. The electronic device or the remote server sequentially performs matching in accordance with the order of the material feature vectors stored in the vector table in performing vector matching with respect to the image tile feature vector and the feature feature vector, For example, the search may be performed in accordance with the search order provided in the index structure of the tree type, or the search may be performed in other order. The present application does not proceed to the limitation thereof.

In step S408, based on the correspondence relationship between the image material and the material feature vector, the storage address of the corresponding image material is obtained based on the storage location.

In the present embodiment, based on the correspondence relationship between the image material and the material feature vector, the electronic device or the remote server can acquire the storage address of the corresponding image material based on the storage location. Here, the storage address of the image material may be a storage sequence number or an actual address in the material library of the image material. Those skilled in the art will recognize that one-to-one correspondence or mapping relationship exists between the storage location and the corresponding storage location of the image material since the material feature vector is a feature vector of the image material and there is a one- Lt; / RTI > Alternatively, the image material may have the same storage order as the corresponding material feature vector.

In step S409, based on each storage address, a corresponding image material is obtained and a mosaic image is generated.

In the present embodiment, the electronic device or the remote server first reads the storage file of the material library based on each storage address acquired in step S408, sequentially acquires the corresponding image material from the storage file, The mosaic image can be generated by replacing the corresponding image tile with the acquired image material.

In this embodiment, steps S404, S405, S406, S407, S408, and S409 in the implementation process are performed in steps S101, Step S102, step S103, step S104, step S105, and step S106, and therefore detailed description thereof will be omitted herein.

5A and 5B illustrate an example of generating a mosaic image of a target image using the mosaic image generating method of the present application. Here, FIG. 5B is a mosaic image in which geographical information points of one city are created as an image material for a portrait image in FIG. 5A. 5A and 5B, FIG. 5A is an image acquired by an electronic device or a remote server, and a pixel is relatively large. The electronic device or the remote server proceeds to section cutting the image and then regards it as the target image to generate a mosaic image as shown in Fig. 5B. The mosaic image in Fig. 5B includes only the head portion image of the portrait image in Fig. 5A. The mosaic image in FIG. 5B may be enlarged to represent each image material having local characteristics of the corresponding city.

As shown in FIG. 4, in contrast to the embodiment corresponding to FIG. 1, the flow 400 for generating a mosaic image in this embodiment includes a step S401 of obtaining an image, a step of determining a pixel size of the image (S402) and adjusting the image size as the target image (S403). According to the added step (S401), step (S402), and step (S403), the method described in this embodiment helps to proceed with screening and adjustment for the image selected or photographed by the user, To further enhance the authenticity and effectiveness of the invention.

6, as an implementation of the method, the present application provides an embodiment of a mosaic image generating device, the device embodiment corresponding to the method embodiment shown in Fig. 1, Specifically a back-end server that provides support for an electronic device (e.g., a smartphone on which a mosaic-generating application is run) or a server (e.g., a type of application that generates a mosaic image on an electronic device) .

6, the mosaic image generating apparatus 600 according to the present embodiment includes a division module 601, an extraction module 602, a vector table acquisition module 603, a position acquisition module 604, An acquisition module 605 and a generation module 606. Here, the partitioning module 601 is configured to partition the destination image into a plurality of predetermined sized image tiles, and the extracting module 602 is configured to extract an image tile feature vector for each image tile, Module 603 obtains a feature vector table for storing the feature vector of each image material in the material library corresponding to the material library, and the position acquisition module 604 obtains, for each image tile feature vector, Based on the correspondence relationship between the image material and the material feature vector, the address acquisition module (605) is configured to acquire the storage location of the corresponding image material based on the storage location, The generation module 606 is configured to obtain a corresponding image material based on each storage address, To generate an image.

In this embodiment, the division module 601 may divide the target image into image tiles of a predetermined size according to the pixels. Here, the predetermined size indicates that the image tiles acquired by dividing the target image have the same pixel size as the image material used for generating the mosaic image.

In this embodiment, the extraction module 602 may extract image tile feature vectors for the image tiles segmented by the segmentation module 601. [ Here, the image tile feature vector may be a vector representing one or more features (e.g., color features, etc.) of the image tile and may be, for example, a vector of color values of a plurality of feature points .

In the present embodiment, the vector table acquisition module 603 can acquire a feature vector table corresponding to a default or selected material library from a local or remote server. Here, the feature vector table is configured to store a material feature vector of each image material in the corresponding material library. The material feature vector may be a vector representing one or more features (e.g., color features, etc.) of the image material and may be, for example, a vector of color values of a plurality of feature points. Those skilled in the art will appreciate that, for convenience of matching, the material feature vectors of each image material have the same extraction and display methods as the feature vectors of the image tiles above.

In this embodiment, the position acquiring module 604 acquires, for each image tile feature vector acquired by the extracting module 602, the material feature vector in the feature vector table acquired by the vector table acquiring module 603, And obtain the storage position in the feature vector table of the matched material feature vector. Alternatively, the position acquiring module 604 may perform matching sequentially in accordance with the order of the material feature vectors stored in the vector table, or may perform an index structure of the feature vector table (for example, Structure) according to the search order, or may perform matching in other order. The present application does not proceed to the limitation thereof.

In this embodiment, the address acquisition module 605 may acquire the storage address of the corresponding image material based on the storage location, based on the correspondence relationship between the image material and the material feature vector. Here, the storage address of the image material may be a storage sequence number or an actual address in the material library of the image material.

In this embodiment, the generation module 606 acquires corresponding image materials based on the respective storage addresses acquired by the address acquisition module 605, and acquires image tiles corresponding to the next acquired image material Thereby generating a mosaic image.

In an alternative implementation of this embodiment, the location acquisition module 604 includes an index structure obtaining unit (not shown) configured to obtain an index structure of the feature vector table; A search unit (not shown) configured to search, for each image tile feature vector, a feature feature vector matching in the feature vector table according to the index structure; And an acquisition unit (not shown) configured to acquire a storage position in a feature vector table of each material feature vector matched from the index structure. Here, the index structure is configured to provide a search order and a storage location in the feature vector table of the material feature vector when searching for each material feature vector.

In an alternative embodiment of the present invention, the address acquisition module 605 may determine a mapping relationship between a storage address in the storage file of the material library of the image material and a storage location in the feature vector table of the material feature vector corresponding to the image material A mapping relationship acquisition unit (not shown) configured to acquire a mapping relationship; And a storage address acquisition unit (not shown) configured to acquire a storage address in a storage file of a material library of the corresponding image material based on the storage location, based on the mapping relationship.

In an alternative implementation of this embodiment, the generation module 606 comprises: a reading unit (not shown) configured to read the storage file; A material acquisition unit (not shown) configured to acquire corresponding image material in the storage file according to each storage address; And an alternate unit (not shown) configured to generate a mosaic image by replacing each corresponding image tile in the target image with each acquired image material.

In an alternative implementation of this embodiment, the mosaic image generating device 600 further comprises a selection module (not shown), the selection module comprising: an image acquisition unit (not shown) configured to acquire an image; A pixel size determination unit (not shown) configured to determine whether a pixel size of the image falls within a predetermined pixel size range; And a pixel size adjustment unit configured to adjust the pixel size of the image to within a predetermined pixel size range as a target image if the pixel size of the image does not fall within the predetermined pixel size range.

In an alternative implementation of this embodiment, the mosaic image generation device 600 further comprises an image material generation module (not shown), the image material generation module comprising an original image acquisition unit city); And an image modification unit (not shown) configured to modify the original image to a predetermined size as an image material. Here, the modification includes at least one of image region selection and pixel compression, but is not limited thereto. Alternatively, when the original image is a street view image (for example, a street view image in a panorama image), there is a possibility that the street view image is an image including a fisheye. When direct compression is performed on the Street View image, there is a possibility that the details of the image may be missing or edge distortion may occur. At this time, the image material creation module may include an image conversion unit (not shown) configured to map the original image first to the spherical model and then to project back to the plane again; And a conversion image modification unit (not shown) configured to modify the original image that is projected back to the plane to a predetermined size as an image material.

Those skilled in the art will appreciate that the mosaic image generating device 600 further includes other known structures, such as, for example, a processor and a storage device, and is shown in FIG. 6 so as not to unnecessarily obscure the embodiments of the present disclosure. It will be understood that the structures are not shown.

The modules or units according to the embodiments of the present application may be implemented in a software manner or in a hardware manner. The described module or unit may be installed in a processor, for example, the processor may be described as including a partitioning module, an extraction module, a vector table acquisition module, a location acquisition module, an address acquisition module and a generation module. Here, the names of these modules do not constitute a limitation for the module itself in some cases, for example, the division module is described as "a module configured to divide the destination image into a plurality of image tiles of predetermined size" It is possible.

Figure 7 shows a structural schematic diagram of a computer system provided in an embodiment of the present application.

Referring to FIG. 7, FIG. 7 shows a structural schematic diagram of a computer system 700 suitable for implementing an apparatus according to an embodiment of the present application.

7, the computer system 700 includes a central processing unit 701 (CPU) and reads programs stored in a read-only memory device 702 (ROM) from a storage 708 to a random access memory device 703; RAM), which are capable of executing various appropriate operations and processes. When the program stored in the ROM 702 or the program loaded from the storage unit 708 into the RAM 703 is executed by the CPU 701, the CPU 701 executes the mosaic image generation method according to the above- . In the RAM 703, various programs and data necessary for operating the system 700 are further stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus line 704. [ An input / output (I / O) interface 705 is also connected to bus line 704.

The input unit 706 includes a keyboard, a mouse, and the like. The output unit 706 includes a cathode ray tube (CRT), a liquid crystal display (LCD) A storage unit 708 including a hard drive or the like and a communication unit 709 including a network interface card such as a LAN card or a modem. The communication unit 709 executes communication processing through a network such as the Internet. The driver 710 is also connected to the I / O interface 705 according to demand. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory device, or the like, is installed in the drive 710 according to demand to install a computer program read from these media in the storage unit 708, do.

In particular, and in accordance with the embodiments of the present disclosure, the processes described with reference to the flowcharts may be implemented in computer software programs. For example, an embodiment of the present disclosure includes a computer program product and includes a computer program tangibly embodied in a computer-readable medium, the computer program comprising computer code for executing the method shown in the flowchart . In this embodiment, the computer program may be downloaded and installed from the network via the communication unit 709 and / or installed from the removable medium 711. [

The flowcharts and block diagrams in the accompanying figures illustrate implementable system structures, functions, and operations of the systems and methods according to the embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a portion of a code, and the module, program segment, or portion of code may include one or more Executable instructions. It should be noted that, for some alternative implementations, the functions indicated in the blocks may occur in an order different from the order indicated in the accompanying drawings. For example, two blocks that are sequentially displayed may actually be executed at the same time on a basic basis, and in some cases may be executed in the opposite order, depending on the relevant function. Each block and / or block diagram in the block diagrams and / or flowchart illustrations and / or combinations of blocks in the flowchart illustrations may be embodied in a dedicated hardware-based system that performs the prescribed functions or operations, or may be implemented in a combination of dedicated hardware and computer instructions .

In another aspect, the present application further provides a computer-readable storage medium, which is a computer-readable storage medium contained in the apparatus described in the above embodiments, But may also be a computer-readable storage medium that is not installed. The computer-readable storage medium stores one or more programs, and the program is configured to execute the mosaic image generation method described in this application by one or more processors.

The foregoing description is only an explanation of the relatively preferred embodiments of the present application and the technical principles in operation. It will be understood by those skilled in the art that the scope of the present invention is not limited to the technical solutions made up of specific combinations of the technical features but can be applied to any combination of the technical features or their equivalent features without departing from the gist of the invention , And other technical measures, including, for example, technical features that have been replaced with technical features having similar features as those disclosed in this application, but which are not limited thereto.

Claims (19)

Dividing the target image into a plurality of predetermined sized image tiles;
Extracting an image tile feature vector having color values at a plurality of feature points for each image tile;
Obtaining a feature vector table for storing a material feature vector having color values at a plurality of feature points of each image material in the material library corresponding to the material library;
Obtaining, for each image tile feature vector, a storage location in the feature vector table of the material feature vector matched by vector matching with the image tile feature vector;
Obtaining a storage address of a corresponding image material based on the storage location based on a correspondence relationship between the image material and the material feature vector; And
And acquiring a corresponding image material based on each storage address and generating a mosaic image. The method according to claim 1,
Wherein for each image tile feature vector, acquiring a storage location in the feature vector table of the matched texture feature vector comprises:
Obtaining an index structure of the feature vector table, the index structure being configured to provide a storage location in the feature vector table of a search sequence and a material feature vector, when proceeding with a search for each material feature vector;
For each image tile feature vector, retrieving a material feature vector matching in the feature vector table according to the index structure; And
And obtaining a storage position in the feature vector table of each matched texture feature vector from the index structure. 3. The method of claim 2,
The index structure may include:
Dividing the feature vector table according to a neighborhood relation of each material feature vector and generating a tree structure; And
And generating an index structure of the feature vector table by replacing each corresponding material feature vector in the tree structure with a storage position in the feature vector table of each material feature vector. Way. The method according to claim 1,
Wherein the step of acquiring a storage address of a corresponding image material based on the storage location, based on a correspondence relationship between the image material and the material feature vector,
Obtaining a mapping relationship between a storage address of the image material in the storage file of the material library and a storage location in the feature vector table of the material feature vector corresponding to the image material; And
And acquiring a storage address in a storage file of a material library of the corresponding image material based on the storage location based on the mapping relationship. 5. The method of claim 4,
Wherein the step of acquiring a corresponding image material and generating a mosaic image based on the storage addresses comprises:
Reading the storage file;
Obtaining a corresponding image material from the storage file according to each storage address; And
And generating a mosaic image by replacing each corresponding image tile in the target image with each acquired image material. 6. The method according to any one of claims 1 to 5,
Before dividing the target image into a plurality of predetermined sized image tiles,
Acquiring an image;
Determining whether a pixel size of the image falls within a predetermined pixel size range; And
Further comprising adjusting the pixel size of the image to within a predetermined pixel size range as a target image if the pixel size of the image does not fall within the predetermined pixel size range. 6. The method according to any one of claims 1 to 5,
Obtaining an original image; And
Further comprising modifying the original image to an image of a predetermined size as an image material, wherein the modification includes at least one of image region selection and pixel compression. 8. The method of claim 7,
Wherein the modifying the original image into an image of a predetermined size as an image material comprises:
Mapping the original image to a spherical model first and then projecting the original image back onto a plane; And
Further comprising modifying the original image that is projected back to the plane into an image of a predetermined size as an image material. 8. The method of claim 7,
Further comprising the step of categorizing and storing the image material into a plurality of material libraries so that the user can select the material library. A partitioning module configured to partition the destination image into a plurality of image tiles of predetermined size;
An extraction module configured to extract an image tile feature vector having color values at a plurality of feature points for each image tile;
A vector table acquisition module configured to acquire a feature vector table for storing a material feature vector having color values at a plurality of feature points of each image material in the material library corresponding to the material library;
A location acquiring module configured to acquire, for each image tile feature vector, a storage location in the feature vector table of the material feature vector matched by vector matching with the image tile feature vector;
An address acquisition module configured to acquire a storage address of a corresponding image material based on the storage location, based on a correspondence relationship between the image material and the material feature vector; And
And a generation module configured to acquire the corresponding image material based on each storage address and generate a mosaic image. 11. The method of claim 10,
The position acquiring module includes:
And an index structure configured to obtain an index structure of the feature vector table, wherein when the search is performed for each material feature vector, an index structure configured to provide a storage location in the feature vector table of search order and material feature vectors, unit;
A search unit configured to search, for each image tile feature vector, a feature feature vector matching in the feature vector table according to the index structure; And
Further comprising a storage location acquiring unit configured to acquire a storage location in the feature vector table of each matched texture feature vector from the index structure. 12. The method of claim 11,
The position acquiring module includes:
A tree structure is generated by dividing the feature vector table according to a neighborhood relationship of each material feature vector, and each corresponding material feature vector in the tree structure is replaced with a storage position in the feature vector table of each of the material feature vectors And an index structure generation unit configured to generate an index structure of the feature vector table. 11. The method of claim 10,
The address acquisition module,
A mapping relation acquisition unit configured to acquire a mapping relationship between a storage address of the image material in the storage file of the material library and a storage location in the feature vector table of the material feature vector corresponding to the image material; And
Further comprising a storage address acquisition unit configured to acquire a storage address in a storage file of a material library of the corresponding image material based on the storage location based on the mapping relationship. 14. The method of claim 13,
Wherein the generation module comprises:
A reading unit configured to read the storage file;
A material acquiring unit configured to acquire the corresponding image material in the storage file according to each storage address; And
Further comprising an alternative unit configured to generate a mosaic image by replacing each corresponding image tile in the target image with each acquired image material. 15. The method according to any one of claims 10 to 14,
Further comprising a sorting module,
An image acquisition unit configured to acquire an image;
A pixel size determination unit configured to determine whether a pixel size of the image falls within a predetermined pixel size range; And
And a pixel size adjustment unit configured to adjust the pixel size of the image within a predetermined pixel size range as a target image when the pixel size of the image does not fall within the predetermined pixel size range. 15. The method according to any one of claims 10 to 14,
Further comprising an image material creation module,
An original image acquiring unit configured to acquire an original image; And
Wherein the modification includes an image modification unit including at least one of image region selection and pixel compression, the modification being configured to modify the original image to a predetermined size as an image material. 17. The method of claim 16,
The image material creation module includes:
An image switching unit configured to map the original image first to a spherical model and to project the original image back to the plane again; And
Further comprising: a conversion image modification unit configured to modify an original image that is projected back to the plane to a predetermined size as an image material. A processor; And
Memory device,
Wherein the memory device stores computer readable instructions that can be executed by the processor and wherein when the computer readable instructions are executed the processor is configured to perform the method according to any of claims 1 to 5 The device features. For non-volatile computer storage media,
Wherein the computer storage medium stores computer readable instructions that can be executed by a processor and, when the computer readable instructions are executed by a processor, the processor is adapted to perform the method according to any one of claims 1 to 5 Lt; RTI ID = 0.0 > volatile < / RTI > computer storage medium.
KR1020167027049A 2015-03-31 2015-12-29 Method and apparatus for generating mosaic image KR101785982B1 (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
CN201510149873.3 2015-03-31
CN201510149873.3A CN104715449A (en) 2015-03-31 2015-03-31 Method and device for generating mosaic image
PCT/CN2015/099457 WO2016155382A1 (en) 2015-03-31 2015-12-29 Mosaic image generation method and device

Publications (2)

Publication Number Publication Date
KR20160130792A KR20160130792A (en) 2016-11-14
KR101785982B1 true KR101785982B1 (en) 2017-10-18

Family

ID=53414744

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020167027049A KR101785982B1 (en) 2015-03-31 2015-12-29 Method and apparatus for generating mosaic image

Country Status (4)

Country Link
JP (1) JP2017520036A (en)
KR (1) KR101785982B1 (en)
CN (1) CN104715449A (en)
WO (1) WO2016155382A1 (en)

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104715449A (en) * 2015-03-31 2015-06-17 百度在线网络技术（北京）有限公司 Method and device for generating mosaic image
CN107133920B (en) * 2017-06-13 2021-07-30 华侨大学 Automatic mosaic generation method based on visual features
CN107767340A (en) * 2017-10-26 2018-03-06 厦门理工学院 The synthesis preparation method of electronic photo
CN108010102B (en) * 2017-12-19 2021-03-05 刘邵宏 Mosaic image generation method and device, terminal equipment and storage medium
WO2019216824A1 (en) * 2018-05-11 2019-11-14 Brickzle Pte. Ltd. Method and system for optimizing the fabrication of mosaic artwork and assembly thereof
CN109325170A (en) * 2018-08-06 2019-02-12 江西清华泰豪三波电机有限公司 Material method for pushing and device
KR102241486B1 (en) * 2019-02-14 2021-05-17 엔에이치엔 주식회사 Method that provides and creates mosaic image based on image tag-word
CN110390637B (en) * 2019-07-18 2022-12-13 天津塔米智能科技有限公司 Mosaic image generation method, device, equipment and storage medium
CN110688962B (en) * 2019-09-29 2022-05-20 武汉秀宝软件有限公司 Face image processing method, user equipment, storage medium and device
CN110891195B (en) * 2019-11-22 2022-07-29 腾讯科技（深圳）有限公司 Method, device and equipment for generating screen image and storage medium
CN111782849B (en) * 2019-11-27 2024-03-01 北京沃东天骏信息技术有限公司 Image retrieval method and device
CN111353532A (en) * 2020-02-26 2020-06-30 北京三快在线科技有限公司 Image generation method and device, computer-readable storage medium and electronic device
CN113868440B (en) * 2020-06-30 2023-06-27 华为技术有限公司 Feature library management method, device, equipment and medium
CN111986089A (en) * 2020-08-28 2020-11-24 计易数据科技(上海)有限公司 Image storage and comparison method, device, equipment and medium with characteristic value being integer
CN112308036A (en) * 2020-11-25 2021-02-02 杭州睿胜软件有限公司 Bill identification method and device and readable storage medium
CN114040223B (en) * 2021-11-05 2023-11-24 亿咖通(湖北)技术有限公司 Image processing method and system

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH08110912A (en) * 1994-10-07 1996-04-30 Canon Inc Device and method for retrieving moving image
JP2000090092A (en) * 1998-09-08 2000-03-31 Canon Inc System, device and method for processing image and storage medium
JP2000089991A (en) * 1998-09-09 2000-03-31 Fujitsu Ltd Document management system
JP4355535B2 (en) * 2003-08-07 2009-11-04 株式会社岩根研究所 360 degree image conversion processing device
CN101739697B (en) * 2008-11-25 2012-01-04 王源源 Synthesis method of picture mosaic pattern and system therefor
CN101706793B (en) * 2009-11-16 2012-09-26 中兴通讯股份有限公司 Method and device for searching picture
CN102609894B (en) * 2012-01-14 2014-06-25 暨南大学 Synthetic method for mosaic image and device thereof
CN102831593A (en) * 2012-07-23 2012-12-19 陈华 Digital picture splicing system and method for carrying out mosaic picture splicing by using system
CN103049755B (en) * 2012-12-28 2016-08-10 合一网络技术(北京)有限公司 A kind of method and device realizing dynamic video mosaic
US9058673B2 (en) * 2013-03-15 2015-06-16 Oracle International Corporation Image mosaicking using a virtual grid
CN103678661A (en) * 2013-12-24 2014-03-26 中国联合网络通信集团有限公司 Image searching method and terminal
CN103729430B (en) * 2013-12-26 2016-08-31 北京京东尚科信息技术有限公司 The method and apparatus generating image file
CN103927387B (en) * 2014-04-30 2017-06-16 成都理想境界科技有限公司 Image indexing system and its correlation technique and device
CN104715449A (en) * 2015-03-31 2015-06-17 百度在线网络技术（北京）有限公司 Method and device for generating mosaic image

Also Published As

Publication number Publication date
WO2016155382A1 (en) 2016-10-06
JP2017520036A (en) 2017-07-20
KR20160130792A (en) 2016-11-14
CN104715449A (en) 2015-06-17

Similar Documents

Publication Publication Date Title
KR101785982B1 (en) Method and apparatus for generating mosaic image
US20200342652A1 (en) Generating Synthetic Image Data for Machine Learning
CN106709871B (en) Method and system for image composition using active masks
US8571338B2 (en) Image file generation device, image processing device, image file generation method, and image processing method
US7965902B1 (en) Large-scale image processing using mass parallelization techniques
US8340423B1 (en) Enhancing digital image mosaics using region-statistics
US20160027159A1 (en) Low memory content aware image modification
JP2012530319A (en) Method and system for quasi-duplicate image retrieval
WO2004109596A2 (en) Image region filling by example-based inpainting
Liang et al. Embedding user-generated content into oblique airborne photogrammetry-based 3D city model
CN103413339B (en) 1000000000 pixel high dynamic range images are rebuild and the method for display
CN111325271A (en) Image classification method and device
CN111666363A (en) Method and device for slicing and quality control of mass high-resolution image data
US20230020851A1 (en) Polygon overlap assignment using medial axis
US9471959B2 (en) Water color gradients on a digital map
US7907147B2 (en) Texture filtering apparatus, texture mapping apparatus, and method and program therefor
JP6004260B2 (en) Line drawing coloring system
US8885952B1 (en) Method and system for presenting similar photos based on homographies
CN101539937A (en) Method for fast display of vast images
JP2004021430A (en) Image searching device, image searching method, and image searching program
EP2966613A1 (en) Method and apparatus for generating a super-resolved image from an input image
JP2016045600A (en) Image processing device and image processing method
CN113256484B (en) Method and device for performing stylization processing on image
CN114820988A (en) Three-dimensional modeling method, device, equipment and storage medium
US20120042268A1 (en) Processing user interfaces

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E701 Decision to grant or registration of patent right
GRNT Written decision to grant