JP5798357B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing technique for expressing an image as a vector. In particular, the present invention relates to a vector expression technique for expressing a raster image with a mesh.

  Conventionally, a technique for expressing an image as a vector is widely known. The vector-represented image has advantages such as little image quality deterioration when the image is enlarged / reduced, easy image editing, and high compression ratio.

  In the case of vector representation of illustrations and characters, a method of approximating the contour of an object with a Bezier, spline function or the like is used. The object area can be painted with a single color, linear gradation, radial gradation, etc., but it is difficult to express complex gradation.

  In the case of expressing an object including a complex gradation as a vector, an Adobe Illustrator (registered trademark) gradient mesh tool is generally used. In a gradient mesh, an object including a complex gradation can be drawn by generating a cubic function by giving a color and a gradient to the mesh (Patent Document 1).

  A gradient mesh is a curved mesh composed of four vertices, and a technique for expressing a complex gradation vector by a curved mesh composed of three vertices is known (Non-patent Document 1). Edges in the image are extracted as line segments (hereinafter referred to as edge lines), and vector expression is performed by constructing a three-vertex curve mesh that reproduces the edge lines. Here, the three-vertex curve mesh is assumed that the number of vertices of the mesh is 3, and the sides of the mesh connecting the vertices are straight lines or curves.

Special registration 04220010

Tian Xia, Binbin Liao, and Yizhou Yu, “Patch-Based Image Collection with Automatic Curvilinear Feature Alignment”, ACM SIGGRAPH Asia Via.9. 28, no. 5.

  Consider a vector representation of an object such as a natural image whose color changes in a complex manner using a curved mesh. Since a curved mesh approximates the shape and characteristics of an object with a curved line, the object can be approximated with a smaller number of meshes.

  When approximating an object using the above-described gradient mesh, a method of constructing a curved mesh composed of four vertices in a regular lattice shape is known. This method does not need to maintain mesh connection relations, so it is efficient as data representation. However, since mesh control cannot be performed locally, the number of meshes increases when approximating complex shapes, and the amount of data increases. End up.

  In the above-mentioned three-vertex curve mesh, mesh control can be performed locally, so that an object can be approximated with a small number of meshes. Conventionally, a three-vertex curve mesh is encoded in units of meshes, and vertices, sides, and colors shared between meshes are not taken into account, so that the encoding efficiency is low and the amount of data increases.

  The present invention has been made in view of the above problems, and an object thereof is to provide a technique for efficiently encoding a three-vertex curve mesh.

In order to achieve the object of the present invention, for example, the image processing apparatus of the present invention encodes an input image in units of a mesh composed of three vertices and the side connecting the three vertices is a straight line or curve An image processing apparatus that executes an encoding process,
The input image is divided into units of the mesh in the course of the mesh encoding process, the mesh is divided into a plurality of sub-mesh, and the position of the vertex in the input image with respect to each vertex of the sub-mesh A color assigning means for assigning colors in
Information assigning means for assigning unique information to each of mesh sides shared between two adjacent meshes on the input image, mesh sides not shared with any mesh,
For each divided mesh, information assigned by the information assigning means to each side constituting the mesh, and information indicating the color assigned by the color assigning means for each vertex of each sub-mesh in the mesh; Generate mesh data describing,
For each side to which the information assigning unit has assigned information, the information assigned by the information assigning unit to the side, the coordinates of both end positions of the side, and the side are generated from the edge line in the input image. Generating edge data that describes a flag value indicating whether or not a color image and information indicating the color assigned by the color assigning means for each vertex on the edge,
Generating means for generating, as encoded data of the input image, data including mesh data generated for each divided mesh and edge data generated for each edge to which the information assigning means has assigned information. And

  According to the configuration of the present invention, it is possible to efficiently encode a three-vertex curve mesh.

FIG. 3 is a block diagram illustrating an example of a functional configuration of the image processing apparatus. The figure explaining the result undesirable as an edge line. The flowchart of the process which the curved mesh production | generation part 103 performs. The figure explaining the overlap of meshes. The figure explaining the mesh production | generation method. The figure explaining shape information. The figure which shows the structural example of the encoding data which the encoding part 105 produces | generates. The figure which shows the example of a division | segmentation of a bezier patch. The block diagram which shows the hardware structural example of a computer. The block diagram which shows the function structural example of the apparatus which decodes. The figure explaining the shape information and color information of the 3 vertex curve mesh handled by 3rd Embodiment.

  Preferred embodiments of the present invention will be described below with reference to the accompanying drawings. The embodiment described below shows an example when the present invention is specifically implemented, and is one of the specific examples of the configurations described in the claims.

[First Embodiment]
Prior to the description according to the present embodiment, first, the curved mesh will be described. In the present embodiment, the curve connecting the three vertices constituting the curved mesh may be any curve such as a Bezier curve or a B-spline. In this embodiment, for explanation, a Bezier patch having three vertices, which is one of parametric curved surfaces, is used for each curved mesh. Among Bezier patches, in particular, a cubic Bezier patch has a merit that each side of a curved mesh becomes a cubic Bezier curve, so that the user can easily edit the curved mesh after generating the curved mesh. The points in the cubic Bezier patch having the vertices at P1, P2, and P3 are expressed by the following equations.

  Here, s, t, and u are barycentric coordinates in the patch. Further, p1, p2, and p3 are coordinate values of the vertices P1, P2, and P3, respectively. C1, c2, c3, c4, c5, c6, and c7 are parameters that determine the shape of the curved patch.

  Next, a functional configuration example of an image processing apparatus that executes mesh coding processing for coding an input image in units of a mesh (curved mesh) that includes three vertices and whose side connecting the three vertices is a straight line or a curve Will be described with reference to the block diagram of FIG.

  An input image of raster expression is input to the image input unit 101. The input image may be a color image, a gray image, or a black and white image. Further, the input image may be a rectangular image, and when it is desired to vectorize a part of the image, a clipped region such as an image in which the background portion is set to a transparent color may be given. Then, the image input unit 101 sends this input image to the feature line configuration unit 102 at the subsequent stage.

  The feature line configuration unit 102 configures a feature line in the input image received from the image input unit 101. In this embodiment, an edge line is used as the feature line. In this case, the feature line configuration unit 102 includes an edge extraction unit that extracts pixels (edge pixels) that form edges from the input image, and an edge configuration unit that connects edge pixels extracted by the edge extraction unit to form edge lines. And will have.

  The edge extraction unit extracts edge pixels from the input image using a known method called the Canny method. Of course, edge pixels may be extracted using other methods. Next, the edge forming unit forms a line segment group (edge line) that reproduces the edge by forming a line segment that connects adjacent ones of the extracted edge pixels. When a line segment group is formed by connecting adjacent edge pixels, there is a four-connection connection in which four pixels located above, below, left, and right of the target edge pixel are considered to be adjacent. In addition to these four directions, there are also eight connected connections that consider that the four pixels on the upper right, lower right, upper left, and lower left are considered to be adjacent. In this embodiment, 8-connected connection is used.

  However, if an edge pixel extracted by the edge extraction unit using the Canny method is connected to a pixel that can be considered to be adjacent by 8-connected connection, an undesirable result may be obtained as an edge line. This case will be described with reference to FIG.

  2A to 2C show a pixel group constituting a certain area in an image, and one square represents one pixel. In addition, a square indicated by hatching represents an edge pixel extracted by the edge extraction unit, and a square without hatching represents a pixel other than the edge pixel (non-edge pixel).

  As shown in FIG. 2B, when the edge pixel group shown in FIG. 2A is connected to adjacent edge pixels by 8-connection, a loop that connects the edge pixels 201, 202, and 203 is generated. End up. When an edge pixel as shown in FIG. 2A is detected, it is necessary to suppress the occurrence of such a loop. Therefore, in the present embodiment, this problem is solved by using a known technique called the Hiditch thinning method before performing 8-connection. When the thinning is performed on the edge pixel group shown in FIG. 2A, the edge pixel 201 shown in FIG. 2B becomes a non-edge pixel. As a result, as shown in FIG. 2C, the edge pixels (204 to 207) are connected by one edge line having no loop.

  In addition to the edge line, for example, the contour of an object region in the input image can be considered as the feature line. Here, the object refers to a character, an object image, or the like.

  As already mentioned, for the purpose of vectorizing a part of the input image, if a part of the input image is set to transparent color, the pixel at the boundary between the transparent color and the non-transparent color is the outline of the object. It becomes a pixel. The process of combining the contour pixels of the object into one line segment group can be realized by connecting adjacent pixels according to the same process as the process performed by the edge configuration unit. The same applies to pixels on the boundary line between regions having different attributes in the input image.

  Also, the feature line configuration unit 102 may divide the input image into regions and use the boundary lines as feature lines. Many methods are known for area division, such as those using the nearest neighbor method and those using an EM algorithm.

  Returning to FIG. 1, the curved mesh generation unit 103 divides the input image into mesh groups so that the feature lines can be reproduced and the meshes do not overlap each other. A process performed by the curved mesh generation unit 103 will be described with reference to FIG. 3 showing a flowchart of the process. Of course, the main subject of the processing in each step shown in the flowchart of FIG. 3 is the curved mesh generation unit 103.

  First, in step S301, an initial mesh group is generated for the input image. This initial mesh is generated as a mesh that reproduces an edge line. Further, the initial mesh is generated so that the mesh sides do not overlap even if the sides of the mesh reproducing the edge line are curved along the edge line.

  In Non-Patent Document 1, a straight line mesh having a sufficiently short line segment group that faithfully reproduces an edge line as a mesh side and having all pixels of a raster image as vertices is generated. This generation method will be described with reference to FIG. In FIG. 5A, solid lines indicate edge lines, dotted lines indicate other mesh sides, and black circles indicate pixels in the raster image. The edge lines 501, 502, 503, and 504 are all used as sides of the mesh.

  In another example, a straight mesh that faithfully reproduces a feature line, that is, a straight mesh in which all line segments constituting the feature line are sides of the mesh is generated. Since the straight mesh generated in this way faithfully reproduces the feature line without curving the side along the feature line, there is no problem that the meshes overlap with each other. This case will be described with reference to FIG.

  As in FIG. 5A, the solid line indicates the edge line, the dotted line indicates the side of the other mesh, and the black circle indicates the pixel in the raster image. The edge lines 511, 512, 513, and 514 are all used as sides of the mesh, but the pixels of the raster image are sampled in a scatter manner to form vertices. The sampling vertices at this time are determined from the feature amount of the raster image, but other methods may be used as long as a mesh that reproduces the edge line can be generated. Here, a straight mesh that faithfully reproduces the feature line is generated by using the Delaunay triangulation with constraints. The constrained Delaunay triangulation is an algorithm that, when given a point group and a constraint condition, generates a straight line mesh with the given point group as a vertex and satisfying the constraint condition. The constraint condition here is a line segment connecting a part of the point cloud. Of course, other methods may be used as long as a mesh that reproduces the edge line can be generated.

  Returning to FIG. 3, in step S307, a simplified curved mesh is generated. Generation of a simplified curve can be realized by following the same procedure as the method described in Non-Patent Document 1. That is, the process of integrating low-priority meshes of the initial meshes into a large mesh and the process of curving the sides of the mesh along the edge line are repeated. In the curving, the restriction that the meshes do not overlap is not provided, thereby preventing the meshes from overlapping.

  The overlapping of meshes will be described with reference to FIG. In FIG. 4, the solid line indicates the sides of the mesh representing the feature line, the dotted line indicates the sides of the mesh other than the feature line, and FIGS. 4 (a) and 4 (c) are prior to the integration of the mesh, and FIGS. ) After integration of the meshes of FIGS. 4 (a) and 4 (c), respectively.

  Consider the case where the vertices 402 on the feature line are integrated into the vertices 401 by taking FIGS. 4A and 4B as an example. The mesh sides 404 and 405 representing the feature lines are integrated and curved so as to approximate the integrated vertex 402, and become mesh sides 407 representing the feature lines. Also, as shown in FIG. 4B, mesh sides 406 other than the feature lines are connected to the vertex 401 by the integration of the vertex 402. Here, it can be seen that the side 407 of the mesh represented by a curve intersects the side 406 of the mesh represented by a straight line. Such a state is a state where the meshes overlap each other. Since mesh overlap is a major cause of image quality degradation when rendered in vector representation, it must be avoided as much as possible. 4C and 4D, when the same processing is performed, the mesh side 407 represented by a curve and the mesh side 406 represented by a straight line do not intersect. Such a state is a good state where the meshes do not overlap.

  Hereinafter, the simplified curve generation process (step S307) will be described in a form that compensates for parts not described in Non-Patent Document 1. The curved mesh generation process (step S307) basically includes a process of integrating low-priority meshes into a large mesh (step S303), and a process of curving mesh edges along the edge line (step S304). ) And are alternately repeated. The processes in step S303 and step S304 are known and may be performed according to the method described in Non-Patent Document 1. Note that Non-Patent Document 1 uses a known technique called a quadratic error metric method as a priority criterion for determining meshes to be integrated in step S303, but there are various methods such as preferential integration from a mesh with a small area. Conceivable. In Non-Patent Document 1, the feature line is approximated by a Bezier curve in step S304, but a B-spline curve or the like may be used as already described in the present embodiment. Details of the algorithm for setting the parameters of the curve along the feature line are not described in Non-Patent Document 1, but the least square method is used in this embodiment.

  Next, the process in step S305 will be described. The purpose of the process in step S305 is to determine whether or not the meshes overlap each other as the sides of the mesh are curved in step S304. Whether the meshes overlap each other is determined by determining whether the sides of the mesh intersect each other. Since it can be ensured that the line segments do not overlap among the sides of the mesh, it is not necessary to determine the overlap. It is a combination of a line segment and a curve, or a combination of a curve and a curve that needs to be determined.

  A method for determining the overlap between a line segment and a curve or between a curve and a curve will be described in detail. By replacing a curve with a line segment that approximates with sufficient accuracy, line segment and curve, curve and curve overlap judgment can be replaced with line segment and line segment group, line segment group and line segment group overlap judgment, respectively. The overlap determination can be realized by repeating the overlap determination between the line segments. The following (condition) is used to determine whether or not the line segment connecting the points (x1, y1) and (x2, y2) on the two-dimensional plane and the line segment connecting (x3, y3) and (x4, y4) overlap. .

(conditions)
{(X1-x2) * (y3-y1) + (y1-y2) * (x1-x3)} * {(x1-x2) * (y4-y1) + (y1-y2) * (x1-x4) } <0
And {(x3-x4) * (y1-y3) + (y3-y4) * (x3-x1)} * {(x3-x4) * (y2-y3) + (y3-y4) * (x3-x2) )} <0
If this condition is satisfied, it can be determined that the two line segments overlap, and if this condition is not satisfied, it can be determined that the two line segments do not overlap. By performing the overlap determination between the line segments described above, it is possible to determine the overlap between the line segment and the curve and between the curve and the curve.

  As a result of the determination in step S305, if it is determined that there is an overlap, the process proceeds to step S306. On the other hand, if it is determined that there is no overlap, the process returns to step S302.

  In step S306, the state of each mesh is returned to the state immediately before steps S303 and S304 performed last. In this way, when the integration is performed such that the meshes overlap each other, the state before the integration is restored to ensure that a mesh without overlapping is always created. When step S303 is executed again after step S306, the integration that causes the meshes to overlap each other must not be performed.

  In step S302, it is determined whether or not to end the process in step S307. For example, when the number of meshes is half that of the initial mesh, or when the number of meshes is 1/100 of the number of pixels included in the input image, the criterion for determining that the processing is finished may be considered. In addition, there may be a case where it is determined in step S305 that there is an overlap for a certain number of times (for example, 10 times) or more. Moreover, you may combine those conditions.

  Through the above processing, the input image can be divided into a plurality of meshes (a plurality of meshes are generated on the input image).

  Returning to FIG. 1, the color information generation unit 104 determines color information necessary for painting each mesh set on the input image by the curved mesh generation unit 103. Various methods are conceivable for generating color information. One possible method is to paint each mesh with a single color. As a method for determining one color, there is a method in which an average value of pixel values of pixels included in the mesh is used as color information of the mesh. As a method for expressing the colors in the mesh in a more complicated manner, a method for generating a function for expressing the color for each mesh may be considered. For example, in Non-Patent Document 1, a thin plate spline function that paints colors in a mesh is generated for three RGB components based on the color of the input image. In this embodiment, each mesh is further divided into meshes (sub-mesh) composed of three minute vertices, and the color of the pixel at the vertex of the minute mesh is acquired from the input image and assigned to the vertex (color assignment). When rendering, the colors in each small triangle can be supplemented by bilinear interpolation to paint each mesh color.

  The encoding unit 105 encodes the shape information of each mesh generated by the curved mesh generation unit 103 and the color information of each mesh generated by the color information generation unit 104, and generates encoded data for the input image.

  First, shape information and color information necessary for configuring each mesh will be described. FIG. 6A shows shape information related to a mesh (curved mesh) composed of three vertices and connected between the three vertices by a curve (any side is expressed by a curve). Reference numerals 601, 602, and 603 denote vertices of the curved mesh, and reference numerals 604 to 609 denote control points of the curved mesh.

  FIG. 6B shows color information related to the curved mesh shown in FIG. Reference numerals 610 to 624 denote vertexes (sampling points) of the sub-mesh in the curved mesh, and each sampling point has color information. Here, the sampling point is described as having RGB values as color information, but may also have color information in another color space such as CMYK. When the input image is a gray image, the sampling point holds the luminance information of the corresponding pixel as color information. Sampling points are derived by regularly dividing a Bezier patch. Here, 15 points are used as sampling points, but the number of Bezier patch divisions may be changed to change the number of sampling points. For example, in FIG. 8A, the number of divisions is 1, and the number of samplings is 3. In FIGS. 8B, 8C, and 8D, the division numbers are 2, 3, and 4, and the sampling numbers are 6, 10, and 15, respectively.

  Next, the encoding unit 105 generates by encoding the shape information of each mesh generated by the curved mesh generation unit 103 and the color information of each mesh generated by the color information generation unit 104. A data configuration example will be described with reference to FIG.

  FIG. 7A shows two adjacent meshes (curved meshes) 720 and 721, where 701, 702, 703, and 704 are vertexes of the curved mesh, 705 to 709 are sides of the curved mesh, and 710 to 719. Indicates the control points of the curved mesh. Here, the side 707 is not only one side of the curved mesh 720 but also one side of the curved mesh 721 and is a side shared by the respective curved meshes (shared side).

  FIG. 7B shows sampling points 722 to 736 for the curved mesh 720 and sampling points 737 to 751 for the curved mesh 721. However, the color information for the curved mesh 720 includes color information at each of the sampling points 722 to 736, and the color information for the mesh 721 includes color information at each of the sampling points 737 to 751. Here, as described above, the side 707 is shared by the curved mesh 720 and the curved mesh 721. However, it can be seen that the sampling point 722 and the sampling point 747 on the side 707 have the same coordinate value, and the curve mesh 720 and the curve mesh 721 share the same coordinates. The same applies to the respective sets of sampling points 724 and 748, sampling points 727 and 749, sampling points 731 and 750, and sampling points 736 and 751.

  On the other hand, if one side of the curved mesh is a side generated from an edge line (a side generated so that the edge line can be reproduced), each vertex on this side is one curve that shares this side Two colors are retained: the color of the mesh and the color of the other curved mesh. This represents a steep color change that occurs when crossing an edge line.

  For example, when the side 707 is a side on the edge line, the color information of the sampling points 722, 724, 727, 731, 736 on the mesh 720 side and the colors of the sampling points 747, 748, 749, 750, 751 on the mesh 721 side Information will be retained. That is, although the sampling point 722 and the sampling point 747 are at the same coordinate position, the respective sampling points hold different colors. The same applies to the respective sets of the sampling point 724 and the sampling point 748, the sampling point 727 and the sampling point 749, the sampling point 731 and the sampling point 750, and the sampling point 736 and the sampling point 751.

  On the other hand, when one side of the curved mesh is not a side generated from the edge line, each vertex on the one side holds one color. For example, when the side 707 is not on the edge line, the sampling point 722 and the sampling point 747 are at the same coordinate and retain the same color. The same applies to the sampling point 724 and sampling point 748, the sampling point 727 and sampling point 749, the sampling point 731 and sampling point 750, and the sampling point 736 and sampling point 751.

  In the present embodiment, the efficiency of mesh representation is improved by utilizing the characteristics of a three-vertex curve mesh that shares edges, shares sampling points on the edges, and sampling points on the edge line have two types of colors. .

  For this purpose, the encoding unit 105 generates encoded data as follows. First, specific information is assigned to each of mesh sides shared between two adjacent meshes on the input image and mesh sides not shared with any mesh (side information assignment). That is, a side shared between two adjacent meshes is also regarded as one side, and unique information is assigned to each side. In the case of FIG. 7A, unique information is assigned to each of the sides 705 to 709. However, one piece of unique information is also assigned to the side 707 shared by the mesh 720 and the curved mesh 721. This assignment may be performed before the operation of the encoding unit 105.

  The “unique information” (unique information) assigned to each side in such an information assignment process may be any information as long as the information can uniquely identify the side. For example, it may be an address on a memory that holds a side, or may be a number assigned without overlapping each side.

  And the encoding part 105 produces | generates the encoding data in which the following various information was described for every mesh. That is, encoded data in which unique information assigned to each side constituting one mesh and information indicating colors assigned to the vertices of each sub-mesh in the mesh are described as encoded data for the mesh. Generate. Such encoded data is generated for each mesh.

  In the present embodiment, encoded data is generated for each side to which specific information is assigned, but the encoded data for one side (side of interest) describes information to be listed next.

-Unique information assigned to the target side-Coordinates of the positions of both ends of the target side-Flag value indicating whether the target side is generated from the edge line in the input image-Each vertex on the target side ( Information indicating the color assigned to the sampling point)-Coordinates of the control point of the target side assigned to the target side during the mesh encoding process. Thus, encoded data with such information is generated for each side. To do. Then, the encoding unit 105 collects the encoded data generated for each mesh, the encoded data generated for each side, and the header (including the number of sides, the number of meshes, the number of mesh divisions), and collects the input image Is output as encoded data.

  Here, the encoded data for the side will be described by taking the side 707 as an example. When the encoded data of the side 707 is generated, the address of the memory holding the side 707 and the number assigned to the side 707 are described as “unique information assigned to the side 707”.

  As the “coordinates of both end positions of the side 707”, the x coordinate values and y coordinate values of the vertices 701 and 703 at both ends of the side 707 are described. “Flag value indicating whether or not the side 707 is generated from an edge line in the input image” is a value corresponding to whether or not the side 707 is generated from an edge line in the input image (For example, “It is generated” = 1, “It is not generated” = 0).

  “Information indicating the color assigned to each vertex (sampling point) on the side 707” varies depending on whether the side 707 is generated from an edge line in the input image. If the side 707 is generated from an edge line in the input image, the color information of the sampling points 722, 724, 727, 731, 736 (for the mesh 720), the sampling points 747, 748, 749, 750, 751 Describes color information (for mesh 721).

  At this time, the description order of the coordinates at both ends of the side is the side direction, and the color information is described in the order of the right side and the left side of the side. In the case of the side 707, when the coordinate value of the vertex 701 and the coordinate value of the vertex 703 are described in this order, the color information of the sampling points 722, 724, 727, 731 and 736 is set to the right side of the side from the vertex 701 to the vertex 703. Is described. Next, color information of sampling points 747, 748, 749, 750, and 751 is described on the left side. Of course, other description methods may be used as long as the right and left sides can be determined with respect to the side.

  On the other hand, when the side 707 is not generated from the edge line in the input image, only the color information of the sampling points 722, 724, 727, 731 and 736 is described. Here, in the present embodiment, the color information is described as representing RGB values, and therefore other representations such as using the RGB value difference between sampling points may be used as long as the RGB position can be derived.

  In addition, as the “coordinates of the control point of the side 707 assigned to the side 707 in the course of the mesh encoding process”, the x coordinate value and the y coordinate value of the control point 715 and the control point 714 are described. Here, as for the coordinate value of the vertex, any information may be adopted as long as even the x coordinate value and the y coordinate value can be derived. For example, other expressions such as holding a difference between vertices may be used. Similarly, regarding the coordinate value of the control point, any information may be used as long as the x coordinate value and the y coordinate value can be derived. For example, other expressions such as holding a difference in coordinate values with respect to a vertex or a difference in coordinate values between control points may be used.

  For example, when the coordinate values are described in the order of vertices 701 and 703, the control points are described in the order of the control point 715 corresponding to the vertex 701 and the control point 714 corresponding to the vertex 703. .

  Next, encoded data for the mesh will be described by taking the mesh 720 as an example. As the “unique information assigned to each side configuring the mesh 720”, the number and address of each side of the sides 705, 707, and 706 are described.

  The first two sides described at this time, the vertex 701 shared by the sides 705 and 707, is set as a reference point. By regularly dividing the Bezier patch from the vertex 701 serving as the reference point, sampling points are derived in order. For example, in the mesh 720, sampling points 722 to 736 are derived in order. However, as “information indicating the color assigned to the vertex of each sub-mesh in the mesh 720”, the color information of the sampling points 722 to 736 is described in the derivation order (sampling points 726, 729, 730,...). .

  Note that the encoding unit 105 performs zip encoding on the encoded data generated for each mesh, the encoded data generated for each side, and the header (including the number of sides, the number of meshes, and the number of mesh divisions). Although managed, other encoding methods including lossy encoding may be used.

  Conventionally, when encoding is performed in units of meshes, the coordinates and colors of sampling points on the side had to be encoded twice, whereas in the present embodiment, encoding can be performed once and the encoding efficiency Can be improved.

  The above processing is summarized as follows. First, in the course of mesh coding processing, the input image is divided into units of mesh, the mesh is divided into a plurality of sub-mesh, and the color at the position of the vertex in the input image is divided for each vertex of the sub-mesh. Assign (color assignment).

  Next, unique information is assigned to each of mesh sides shared between two adjacent meshes on the input image and mesh sides not shared with any mesh (information assignment).

  Next, for each divided mesh, mesh data describing information assigned to each side constituting the mesh and information indicating the color assigned to each vertex of each sub-mesh in the mesh Generate.

  Next, side data describing the following information is generated for each side to which information is assigned. That is, information assigned to the side, coordinates of both end positions of the side, a flag value indicating whether or not the side is generated from an edge line in the input image, and a color assigned to each vertex on the side Information indicating the coordinates of the control points of the side assigned to the side in the course of the mesh encoding process.

  Finally, data including mesh data generated for each divided mesh and side data generated for each side to which information is assigned is generated as encoded data of the input image.

  Next, an image processing apparatus that decodes the mesh coding result of the input image output by the coding unit 105 will be described. First, an image processing apparatus that performs this decoding process will be described with reference to the block diagram of FIG. The image processing apparatus having the configuration of FIG. 10 may be a separate apparatus from the image processing apparatus having the configuration of FIG. 1 or the same apparatus. In the latter case, this image processing apparatus is an apparatus having the configuration shown in FIG. 1 and the configuration shown in FIG. 10, and the decoding processing described below temporarily encodes the encoded data generated by itself. The encoded data is stored in the memory and then decoded.

  The encoded data input unit 1001 acquires input image encoded data that is an encoding result of the input image generated by the encoding unit 105, and sends the acquired input image encoded data to the subsequent shape information restoration unit 1002. To do.

  The shape information restoration unit 1002 restores the shape information of the three-vertex curve mesh from the input image encoded data. First, when the shape information restoration unit 1002 acquires the input image encoded data, it temporarily stores it in a memory (not shown) in the image processing apparatus. Then, the encoded data for each mesh is referred to. When the encoded data of the target mesh is referred to, “specific information assigned to each side constituting the target mesh” described in the encoded data is referred to. Then, the encoded data of the side specified by “unique information assigned to each side constituting the target mesh” is read from the input image encoded data. Then, the reference point is restored from the encoded data of the read side, and the target mesh is restored from the restored reference point.

  The processing up to this point will be described using the mesh 720 of FIG. 7 as an example. Reference is made to “unique information assigned to each side constituting the mesh 720” in the encoded data of the mesh 720. Here, the unique information described in the order of the side 705, the side 707, and the side 706 is referred to. Then, referring to the coordinate positions of both ends of the first two sides 705 and 707 in the description order, the coordinate position of one end of the side 705 and the coordinate position of one end of the side 707 are the same. . However, the vertex 701 which is the vertex with respect to the same coordinate position is set as a reference point. Then, the sides are connected counterclockwise from the vertex 701 serving as a reference point. First, the side 705 is selected from the positional relationship between the side 705 and the side 707 (which can be derived from the coordinate values at both ends). Then, by reading out the coordinate positions of the vertex 701, the control point 710, the vertex 702, and the control point 711 from the encoded data of the side 705, these are restored. Next, the coordinate positions of the control point 712, the vertex 703, and the control point 713 are read out from the encoded data of the side 706 to restore them. Finally, the coordinate positions of the control point 714 and the control point 715 are read out from the encoded data of the side 707 to restore them.

  By these restoration processes, information (ie, shape information) necessary for uniquely specifying the shape of the mesh 720 is restored, and as a result, the shape of the mesh 720 is restored. By performing such processing for each mesh, the shape of each three-vertex curve mesh is restored.

  The color information restoration unit 1003 restores the mesh color information using the shape information of the three-vertex curve mesh restored by the shape information restoration unit 1002. This process will be described with reference to FIG. 7 as an example. Sampling points are derived in order by regularly dividing a Bezier patch from a vertex 701 serving as a reference point. In mesh 720, sampling points 722-736 are derived.

  Next, a color is assigned to each sampling point. First, colors are assigned to sampling points on the sides of the mesh 720. The color is restored counterclockwise from the vertex 701 as the reference point. First, the order in which colors are arranged in the encoded data of the side 705 is confirmed. For example, it is assumed that the coordinate positions of both ends of the side 705 are arranged in the order of vertices 701 and 702 in the encoded data of the side 705. In this case, “information indicating the color assigned to each vertex (sampling point) on the side 705” in the encoded data of the side 705 is assigned in the order of the sampling points 722, 723, 725, 728, and 732. On the contrary, if the vertices 702 and 701 are arranged in this order, “information indicating the color assigned to each vertex (sampling point) on the side 705” in the encoded data of the side 705 is converted into sampling points 732, 728, 725, 723 and 722 are assigned in this order. Next, “information indicating the color assigned to each vertex (sampling point) on the side 706” in the encoded data of the side 706 is assigned in the order of sampling points 733, 734, 735, and 736. Finally, colors are assigned to the side 707 in the order of sampling points 731, 727, and 724.

  This color assignment process is a process when the side 707 is not a side on the edge line. If the side 707 is a side on the edge line, it must be determined whether the right or left color of the side 707 is used for the mesh 720. This determination uses the right color when the counterclockwise direction in which the color is restored is opposite to the order in which both ends (vertices) of the side are described. On the other hand, the left color is used. Taking FIG. 7 as an example, the counterclockwise direction for restoring the color is the direction from the vertex 703 toward the vertex 701. When the coordinate values are arranged in the order of the vertex 701 and the vertex 703 in the encoded data of the side 705, the right color is used as the color of the mesh 720. When the vertex 703 and the vertex 701 are arranged in this order, the left color is used as the color of the mesh 720. The color on the side of the mesh is restored by the above processing.

  Whether or not the side 707 is a side on the edge line is determined by “a flag value indicating whether or not the side 707 is generated from the edge line in the input image” in the encoded data of the side 707. You can refer to it.

  As the color at the sampling point in the remaining mesh 720, the “color assigned to the vertex of each sub-mesh in the mesh 720” described in the encoded data of the mesh 720 may be used. The color may be obtained by interpolation from the vertex color. Of course, the method of determining the color of the sampling point that is not on the side is not limited to this.

  Since the mesh shape information and color information have been restored by the above processing, the input image can be restored by using these pieces of information. However, the image output unit 1004 restores the input image using the mesh shape information and color information, and outputs the restored input image as an output image. The output destination is not particularly limited, and the output destination may be stored in an appropriate memory, may be output and displayed on a display device, or may be transmitted to an external device via a network. .

<Modification of First Embodiment>
1 and 10 may be configured by hardware, but may be implemented as software (computer program). In this case, the software is installed in a memory of a general computer such as a PC (personal computer). When the CPU of the computer executes the installed software, the computer realizes the functions of the above-described image processing apparatus (functions of each unit illustrated in FIGS. 1 and 10). That is, this computer can be applied to the above-described image processing apparatus. A hardware configuration example of a computer applicable to the image processing apparatus according to the first embodiment will be described with reference to the block diagram of FIG.

  The CPU 901 controls the entire computer using computer programs and data stored in the RAM 902 and the ROM 903, and executes the above-described processes described as being performed by the image processing apparatus. That is, the above-described processes are executed as performed by each unit shown in FIGS.

  The RAM 902 is an example of a computer-readable storage medium. The RAM 902 has an area for temporarily storing computer programs and data loaded from the external storage device 907 and the storage medium drive 908, data received from the external device via the I / F (interface) 909, and the like. Further, the RAM 902 has a work area used when the CPU 901 executes various processes. That is, the RAM 902 can provide various areas as appropriate. The ROM 903 is an example of a computer-readable storage medium, and stores computer setting data, a boot program, and the like.

  The keyboard 904 and the mouse 905 can be operated by a computer operator to input various instructions to the CPU 901. The display device 906 is configured by a CRT, a liquid crystal screen, or the like, and can display a processing result by the CPU 901 with an image, text, or the like. For example, the input image can be displayed, or a temporarily set mesh, a finally determined mesh, a restored image, or the like can be displayed.

  The external storage device 907 is an example of a computer-readable storage medium, and is a large-capacity information storage device represented by a hard disk drive device. In the external storage device 907, an OS (Operating System), computer programs and data for causing the CPU 901 to realize the functions of the units shown in FIGS. 1 and 10, data of the input image, information described as known information, and the like Is saved. Computer programs and data stored in the external storage device 907 are appropriately loaded into the RAM 902 under the control of the CPU 901 and are processed by the CPU 901.

  The storage medium drive 908 reads a computer program and data recorded on a storage medium such as a CD-ROM or DVD-ROM, and outputs the read computer program or data to the external storage device 907 or the RAM 902. Note that part or all of the information described as being stored in the external storage device 907 may be recorded on this storage medium and read by this storage medium drive 908.

  The I / F 909 is for connecting an external device to a computer. For example, a device such as a digital camera for acquiring the input image is connected to the I / F 909, and the input image is acquired from the external device to the RAM 902 or the external storage device 907 via the I / F 909. Also good. A bus 910 connects the above-described units.

  In the above configuration, when the computer is turned on, the CPU 901 loads the OS from the external storage device 907 to the RAM 902 according to the boot program stored in the ROM 903. As a result, an information input operation can be performed via the keyboard 904 and the mouse 905, and a GUI can be displayed on the display device 906. When the user operates the keyboard 904 or the mouse 905 to input an activation instruction for an image processing application program stored in the external storage device 907, the CPU 901 loads the program into the RAM 902 and executes it. As a result, the computer functions as the image processing apparatus.

  The image processing application program executed by the CPU 901 basically includes functions corresponding to the components shown in FIGS. Here, the encoded data is stored in the external storage device 907. It is apparent from the following description that this computer can be similarly applied to image processing apparatuses according to the following embodiments.

[Second Embodiment]
In the first embodiment, the efficiency of mesh representation is improved by utilizing the characteristics of a three-vertex curve mesh that shares edges, shares sampling points on edges, and sampling points on edge lines have two types of colors. I planned. Specifically, data redundancy is reduced using edge information and mesh information. In the present embodiment, information related to vertices is managed separately, and the redundancy of vertex information handled as edge information is reduced.

  The configuration of an apparatus for encoding an input image in the present embodiment is the same as the configuration shown in FIG. 1, but only the operation of the encoding unit 105 is different from that in the first embodiment. However, the operation of the encoding unit 105 according to the present embodiment will be described below. Note that the encoded data for each mesh is generated in the same manner as in the first embodiment.

  The encoding unit 105 allocates unique information for each vertex (vertex information allocation) to each vertex (actually used as a vertex of the mesh) set on the input image to generate a mesh. . The unique information may be any information as long as the information can uniquely identify each vertex. Then, the encoding unit 105 generates vertex data in which the following information is described for each vertex set on the input image for generating a mesh (actually used as a vertex of the mesh).

-Vertex unique information-Vertex coordinate position The encoding unit 105 generates encoded data for each side to which specific information is assigned, but the encoded data for one side (side of interest) includes: The following information is listed.

-Unique information assigned to the target edge-Specific information of the vertices at both ends of the target edge-Flag value indicating whether the target edge is generated from the edge line in the input image-On the target edge Information indicating the color assigned to each vertex (sampling point)-Coordinates of the control point of the target side assigned to the target side in the course of the mesh encoding process. To generate. The encoding unit 105 generates encoded data generated for each mesh, encoded data generated for each side, vertex data generated for each vertex, header (number of vertices, number of sides, number of meshes, mesh division Are included in the encoded data for the input image and output.

  The description of the vertex data will be described with reference to FIG. In the case of FIG. 7, the vertex data describes the x-coordinates and x-coordinates of the vertices 701, 702, 703, and 704 and the unique information of the vertices 701, 702, 703, and 704, respectively. As for the coordinate position of the vertex, other expressions such as holding the difference between the vertices may be used as long as the x coordinate and the x coordinate can be derived. Further, as long as the vertex-specific address is assigned to the information unique to the vertex, other expression methods may be used.

  The encoding unit 105 includes encoded data generated for each mesh, encoded data generated for each side, vertex data generated for each vertex, header (number of vertices, number of sides, number of meshes, number of mesh divisions) , And the like) are managed by zip encoding. However, management may be performed using other encoding methods including lossy encoding.

  In the first embodiment, the vertex coordinates have to be encoded by the number of sides connected to the vertex. However, in the expression of this embodiment, it can be expressed by one encoding and the encoding efficiency is improved. Improvement can be achieved.

  Next, an apparatus for decoding input image encoded data generated by encoding according to the present embodiment will be described. This decoding apparatus has the configuration shown in FIG. 10 as in the first embodiment, but differs only in the points described below.

  The mesh information 720 in FIG. 7 will be described as an example. The shape information restoration unit 1002 refers to “unique information assigned to each side constituting the mesh 720” in the encoded data of the mesh 720. Here, the unique information described in the order of the side 705, the side 707, and the side 706 is referred to. Then, the coordinate positions corresponding to the unique information of the vertices at both ends of the first two sides 705 and 707 in the description order are acquired from the coordinate information and referred to, and the vertex 701 is set as the reference point. Then, the side 705 is selected from the positional relationship between the side 705 and the side 707 (which can be derived from the coordinate values of both ends).

  Then, from the unique information of the vertices 701 and 702 in the encoded data of the edge 705, the vertex data of the vertices 701 and 702 is specified, and the specified vertex data and the encoded data of the edge 705 are used to specify the vertex 701, the control point 710, The coordinate positions of the vertex 702 and the control point 711 are read out. This restores them.

  Next, the vertex data of the vertex 703 is identified from the unique information of the vertex 703 in the encoded data of the edge 706, and the control point 712, the vertex 703, and the control point 713 are identified from the identified vertex data and the encoded data of the edge 706. Read the coordinate position. This restores them. Finally, the coordinate positions of the control point 714 and the control point 715 are read out from the encoded data of the side 707 to restore them.

  By performing such processing for each mesh, the shape of the three-vertex curve mesh is restored. The method for restoring the mesh color information is performed in the same manner as in the first embodiment.

[Third Embodiment]
In the first and second embodiments, the expression of a three-vertex curve mesh in which the sides of the mesh are curves has been described. However, in this embodiment, the three-vertex curve mesh in the case where the sides of the mesh are mixed with straight lines and curves. The expression of will be described.

  The configuration of the image processing apparatus that generates the input image encoded data has the configuration shown in FIG. 1 as in the first embodiment, but only the operation of the encoding unit 105 is different from that in the first embodiment. However, the operation of the encoding unit 105 will be mainly described below.

  The shape information and color information of the three-vertex curve mesh handled in this embodiment will be described with reference to FIG. The vertices constituting the mesh 1118 are vertices 1101, 1102, and 1103, a curve side is between the vertices 1101 and 1102, a curve side is between the vertices 1102 and 1103, and a vertex is between the vertices 1101 and 1103. Is the side of the straight line. Here, the sides of the curve are generated with control points. However, since the sides of the straight line can be generated as long as both ends thereof are known, the control points are unnecessary.

  In the present embodiment, as in the second embodiment, vertex data for each vertex, encoded data for each mesh, and encoded data for each side are generated, but only the configuration of the encoded data for each side is generated. Different. In the encoded data for one side (side of interest), the information listed below is described.

-Unique information assigned to the target side-Specific information on the vertices at both ends of the target side-Flag value indicating whether the target side is a curve or a straight line-The target side is generated from the edge line in the input image Flag value that indicates whether or not it is a target ・ Information indicating the color assigned to each vertex (sampling point) on the target side ・ The control point of the target side assigned to the target side during the mesh encoding process Coordinates (only when the side of interest is a curve)
Encoded data in which such information is described is generated for each side. For example, taking FIG. 11 as an example, since the side 1107 is a straight line, the encoded data for the side 1107 describes the information listed below.

Specific information assigned to the edge 1107 Specific information of the vertices 1101 and 1103 at both ends of the edge 1107 Flag value indicating whether the edge 1107 is a curve or a straight line (= flag value indicating a straight line)
Flag value indicating whether the edge 1107 is generated from an edge line in the input image. Each vertex (sampling point) 1120, 1122, 1125, 1129, 1134 on the edge 1107 (the edge 1107 is the input image). In addition to this, information indicating the color assigned to the vertices 1145 to 1149) in the case of the one generated from the middle edge line. NULL as coordinates
The encoding unit 105 generates encoded data generated for each mesh, encoded data generated for each side, vertex data generated for each vertex, header (number of vertices, number of sides, number of meshes, mesh division Are included in the encoded data for the input image and output.

  The encoding unit 105 includes encoded data generated for each mesh, encoded data generated for each side, vertex data generated for each vertex, header (number of vertices, number of sides, number of meshes, number of mesh divisions) , And the like) are managed by zip encoding. However, management may be performed using other encoding methods including lossy encoding.

  Next, an apparatus for decoding input image encoded data generated by encoding according to the present embodiment will be described. This decoding apparatus has the configuration shown in FIG. 10 as in the first embodiment, but differs only in the points described below.

  The mesh information 720 in FIG. 7 will be described as an example. The shape information restoration unit 1002 refers to “unique information assigned to each side constituting the mesh 720” in the encoded data of the mesh 720. Here, the unique information described in the order of the side 705, the side 707, and the side 706 is referred to. Then, the coordinate positions corresponding to the unique information of the vertices at both ends of the first two sides 705 and 707 in the description order are acquired from the coordinate information and referred to, and the vertex 701 is set as the reference point. Then, the side 705 is selected from the positional relationship between the side 705 and the side 707 (which can be derived from the coordinate values of both ends).

  Then, “flag value indicating curve or straight line” in the encoded data of side 705 is referred to. In the case of FIG. 7, since this flag value indicates “curve”, the vertex data of the vertices 701 and 702 are specified from the unique information of the vertices 701 and 702 in the encoded data of the side 705. Then, the coordinate positions of the vertex 701, the control point 710, the vertex 702, and the control point 711 are read out from the identified vertex data and the encoded data of the side 705 to restore them.

  Next, “flag value indicating curve or straight line” in the encoded data of side 706 is referred to. In the case of FIG. 7, since this flag value indicates “curve”, the vertex data of the vertex 703 is specified from the unique information of the vertex 703 in the encoded data of the side 706. Then, the coordinate positions of the vertex 703, the control point 712, and the control point 713 are read out from the identified vertex data and the encoded data of the side 706 to restore them.

  Finally, the “flag value indicating a curve or a straight line” in the encoded data of the side 707 is referred to. In the case of FIG. 7, since this flag value indicates “curve”, the control point 714 and the control point 715 are read out from the encoded data of the side 707 to restore them.

  By performing such processing for each mesh, the shape of the three-vertex curve mesh is restored. The method for restoring the mesh color information is performed in the same manner as in the first embodiment. It should be noted that the first to third embodiments can be used in appropriate combinations, and the combination method can be appropriately considered by those skilled in the art.

(Other examples)
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed.

Claims (8)

An image processing apparatus that executes mesh encoding processing that encodes an input image in units of a mesh that includes three vertices and a side connecting the three vertices is a straight line or a curve,
The input image is divided into units of the mesh in the course of the mesh encoding process, the mesh is divided into a plurality of sub-mesh, and the position of the vertex in the input image with respect to each vertex of the sub-mesh A color assigning means for assigning colors in
Information assigning means for assigning unique information to each of mesh sides shared between two adjacent meshes on the input image, mesh sides not shared with any mesh,
For each divided mesh, information assigned by the information assigning means to each side constituting the mesh, and information indicating the color assigned by the color assigning means for each vertex of each sub-mesh in the mesh; Generate mesh data describing,
For each side to which the information assigning unit has assigned information, the information assigned by the information assigning unit to the side, the coordinates of both end positions of the side, and the side are generated from the edge line in the input image. Generating edge data that describes a flag value indicating whether or not a color image and information indicating the color assigned by the color assigning means for each vertex on the edge,
Generating means for generating, as encoded data of the input image, data including mesh data generated for each divided mesh and edge data generated for each edge to which the information assigning means has assigned information. An image processing apparatus. An image processing apparatus that executes mesh encoding processing that encodes an input image in units of a mesh that includes three vertices and a side connecting the three vertices is a straight line or a curve,
The input image is divided into units of the mesh in the course of the mesh encoding process, the mesh is divided into a plurality of sub-mesh, and the position of the vertex in the input image with respect to each vertex of the sub-mesh A color assigning means for assigning colors in
Side information assigning means for assigning unique information to each of mesh sides shared between two adjacent meshes on the input image, mesh sides not shared with any mesh,
Vertex information assigning means for assigning unique information to each vertex set to generate a mesh on the input image;
For each divided mesh, information assigned by the side information assigning unit to each side constituting the mesh, and information indicating the color assigned by the color assigning unit for each vertex of each sub-mesh in the mesh And generate mesh data describing
For each side to which the side information assigning unit has assigned information, information assigned by the side information assigning unit to the side and information assigned by the vertex information assigning unit to vertices located at both ends of the side And a flag value indicating whether or not the side is generated from an edge line in the input image, and information indicating the color assigned by the color assigning unit for each vertex on the side Generated edge data,
For each vertex set to generate a mesh on the input image, generate vertex data describing the information assigned by the vertex information assigning means for the vertex and the coordinate position of the vertex,
Mesh data generated for each divided mesh, side data generated for each side to which the side information assigning unit has assigned information, vertex data generated for each vertex set to generate a mesh on the input image, An image processing apparatus comprising: generating means for generating data including the data as encoded data of the input image. The generating means includes
When the target side is generated from an edge line in the input image, information indicating the color assigned by the color assigning unit for the vertex on the target side of one mesh sharing the target side and the target Information indicating the color assigned by the color assigning unit for the vertex on the target side of the other mesh sharing the side is described in the side data of the target side,
If the target side is not generated from an edge line in the input image, information indicating the color assigned by the color assigning unit for the vertex on the target side of one mesh that shares the target side, The image processing apparatus according to claim 1, wherein the image processing apparatus is described in edge data of the edge of interest. Said generating means, according to claim, characterized in that managing divided mesh data generated for each mesh and the information assigning unit generated edge data for each side has assigned information, the data, including by zip coded 1 An image processing apparatus according to 1.   The edge data further describes a flag value indicating whether the edge is a straight line or a curve. When the flag value indicates that the edge is a curve, the edge data further includes the mesh data. The image processing apparatus according to any one of claims 1 to 4, wherein coordinates of control points of the side assigned to the side in the course of encoding processing are described. An image processing method performed by an image processing apparatus that executes mesh coding processing for coding an input image in units of a mesh composed of three vertices and whose side connecting the three vertices is a straight line or a curve,
The color allocation means of the image processing device divides the input image into units of the mesh in the course of the mesh encoding process, divides the mesh into a plurality of sub-mesh, and for each vertex of the sub-mesh Assigning a color at the position of the vertex in the input image; and
The information allocation unit of the image processing apparatus allocates unique information to each of mesh sides shared between two adjacent meshes on the input image and mesh sides not shared with any mesh. Information allocation process;
The generation means of the image processing apparatus assigns the information assigned in the information assigning step to each side constituting the mesh for each divided mesh, and the color assignment for each vertex of each sub-mesh in the mesh Generate mesh data describing information indicating the color assigned in the process,
For each side to which information is assigned in the information assignment step, the information assigned to the side in the information assignment step, the coordinates of both end positions of the side, and the side are generated from the edge line in the input image. Generating edge data that describes a flag value indicating whether or not a color image and information indicating the color assigned in the color assignment step for each vertex on the edge,
A generation step of generating data including mesh data generated for each divided mesh and side data generated for each side to which information is assigned in the information allocation step, as encoded data of the input image. An image processing method. An image processing method performed by an image processing apparatus that executes mesh coding processing for coding an input image in units of a mesh composed of three vertices and whose side connecting the three vertices is a straight line or a curve,
The color allocation means of the image processing device divides the input image into units of the mesh in the course of the mesh encoding process, divides the mesh into a plurality of sub-mesh, and for each vertex of the sub-mesh Assigning a color at the position of the vertex in the input image; and
The edge information allocating means of the image processing device provides unique information for each of the mesh edges shared between two adjacent meshes on the input image and mesh edges not shared with any mesh. Side information assignment step to be assigned;
Vertex information assigning means of the image processing device assigns unique information to each vertex set to generate a mesh on the input image, and vertex information assigning step,
The generation means of the image processing apparatus assigns the information assigned in the side information assigning step to each side constituting the mesh for each divided mesh, and the color for each vertex of each sub-mesh in the mesh. Generate mesh data describing information indicating the color assigned in the assignment process,
For each side to which information is assigned in the side information assignment step, information assigned in the side information assignment step to the side, and information assigned in the vertex information assignment step to vertices located at both ends of the side And a flag value indicating whether or not the side is generated from an edge line in the input image, and information indicating the color assigned in the color assignment step for each vertex on the side Generated edge data,
For each vertex set to generate a mesh on the input image, generate vertex data describing the information assigned in the vertex information assignment step for the vertex and the coordinate position of the vertex,
Mesh data generated for each divided mesh, edge data generated for each edge assigned information in the edge information assigning step, vertex data generated for each vertex set to generate a mesh on the input image, An image processing method comprising: a generation step of generating data including the data as encoded data of the input image.   A computer program for causing a computer to function as each unit of the image processing apparatus according to any one of claims 1 to 5.

Images