Enhanced Vectorization of Raster Images

• US 20170372455A1
• Filed: 06/22/2016
• Published: 12/28/2017
• Est. Priority Date: 06/22/2016
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Winnemoeller, Holger, Li, Wilmot Wei-Mau, Schiller, Stephen N., Xie, Jun
• Granted Patent: US 9,972,073 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 11/203 Drawing of straight lines or curves
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/33 using feature-based methods
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 10/30 Noise filtering
• CPC Class Codes: G06V 10/469 Contour-based spatial representations, e.g. vector-coding

Abstract

Enhanced vectorization of raster images is described. An image vectorization module converts a raster image with bitmapped data to a vector image with vector elements based on mathematical formulas. In some embodiments, spatially-localized control of a vectorization operation is provided to a user. First, the user can adjust an intensity of a denoising operation differently at different areas of the raster image. Second, the user can adjust an automated segmentation by causing one segment to be split into two segments along a zone marked with an indicator tool, such as a brush. Third, the user can adjust an automated segmentation by causing two segments to be merged into a combined segment. The computation of the vector elements is based on the adjusted segmentation. In other embodiments, semantic information gleaned from the raster image is incorporated into the vector image to facilitate manipulation, such as joint selection of multiple vector elements.

Claims

1. In a digital medium environment to enhance picture vectorization by facilitating conversion of raster images to vector images based on spatially-localized user control, a method implemented by a computing device, the method comprising:

displaying a raster image corresponding to a picture having multiple semantic elements, the raster image including multiple image features;

segmenting the raster image into multiple segments, each segment of the multiple segments based on at least one image feature of the multiple image features, the multiple image features serving as a proxy for the multiple semantic elements of the picture;

presenting a user interface feature to provide localized control of a vectorization of the picture, the localized control pertaining to a portion of the raster image;

detecting a user control signal indicative of the portion of the raster image for which a localized vectorization characteristic is to be adjusted;

producing a vector image corresponding to at least part of the picture based on the multiple segments and an adjusted localized vectorization characteristic; and

displaying the vector image.

2. The method as described in claim 1, wherein the segmenting comprises:

detecting multiple edges of the picture of the raster image using a thresholding procedure that determines variances in colors of adjacent pixels that exceed a color difference threshold;

recognizing the multiple image features of the raster image based on color or texture;

tuning the multiple detected edges using the multiple image features to group edges that correspond to individual ones of the multiple image features; and

deriving the multiple segments of the raster image based on the multiple tuned edges.

3. The method as described in claim 1, wherein:

the segmenting comprises producing a hierarchical segmentation tree including the multiple segments of the raster image; and

the presenting comprises presenting a user interface feature to change a displayed scale for the multiple segments to increase or decrease a level of detail displayed for the multiple segments in accordance with the hierarchical segmentation tree.

4. The method as described in claim 1, wherein:

the presenting comprises providing a brush tool to indicate the portion of the raster image; and

the detecting comprises ascertaining pixels of the raster image over which the user applies the brush tool to determine the portion of the raster image for which the localized vectorization characteristic is to be adjusted.

5. The method as described in claim 1, wherein:

the detecting comprises detecting an area of the raster image as indicated by the user control signal; and

the producing comprises;

adjusting the localized vectorization characteristic by changing an intensity of a denoising operation that is applied to the indicated area of the raster image; and

producing the vector image after the denoising operation is applied to the indicated area in accordance with the changed intensity.

6. The method as described in claim 1, wherein:

the detecting comprises detecting a zone of a segment of the multiple segments as indicated by the user control signal; and

the producing comprises;

adjusting the localized vectorization characteristic by splitting the segment into two segments for the multiple segments based on the indicated zone; and

producing the vector image based on the multiple segments including the two segments.

7. The method as described in claim 1, wherein:

the detecting comprises detecting two segments of the multiple segments as indicated by the user control signal; and

the producing comprises;

adjusting the localized vectorization characteristic by merging the two indicated segments into a combined segment for the multiple segments responsive to the user control signal; and

producing the vector image based on the multiple segments including the combined segment.

8. The method as described in claim 1, further comprising:

generating semantic metadata based on the multiple image features of the raster image,

wherein the producing comprises incorporating the semantic metadata into the vector image.

9. The method as described in claim 8, wherein the incorporating comprises associating an image feature of the multiple image features with multiple vector elements of the vector image.

10. At least one computing device operative in a digital medium environment to enhance picture vectorization by facilitating conversion of raster images to vector images based on spatially-localized user control, the computing device comprising:

a segmentation module configured to segment a raster image into multiple segments, the raster image including multiple image features and corresponding to a picture having multiple semantic elements, each segment of the multiple segments based on at least one image feature of the multiple image features, the multiple image features of the raster image serving as a proxy for the multiple semantic elements of the picture;

a localized user control module configured to;

present a user interface feature to provide localized control of a vectorization of the picture, the localized control pertaining to a portion of the raster image; and

detect a user control signal indicative of the portion of the raster image for which a localized vectorization characteristic is to be adjusted; and

a vector image production module configured to produce a vector image corresponding to at least part of the picture based on the multiple segments and an adjusted localized vectorization characteristic.

11. The computing device as described in claim 10, wherein the localized user control module comprises a denoising module configured to adjust the localized vectorization characteristic by changing an intensity of a denoising operation applied to the indicated portion of the raster image.

12. The computing device as described in claim 10, wherein the localized user control module comprises a segment splitting module configured to adjust the localized vectorization characteristic by splitting one segment into two segments for the multiple segments based on the indicated portion of the raster image.

13. The computing device as described in claim 10, wherein the localized user control module comprises a segment merging module configured to adjust the localized vectorization characteristic by merging two segments into a combined segment for the multiple segments based on the indicated portion of the raster image.

14. The computing device as described in claim 10, wherein the localized user control module is configured to merge segments or to split a segment based on the indicated portion of the raster image and an intensity of an edge in a contour map corresponding to the picture of the raster image.

15. The computing device as described in claim 10, further comprising:

a semantic information incorporation module configured to;

generate semantic metadata based on the multiple image features of the raster image; and

incorporate the semantic metadata into the vector image.

16. In a digital medium environment to enhance picture vectorization by facilitating vector image manipulation using semantic information obtained from a conversion of a raster image, a method implemented by a computing device, the method comprising:

displaying a raster image corresponding to a picture having multiple semantic elements, the raster image including multiple image features, the multiple image features serving as a proxy for the multiple semantic elements;

analyzing the raster image to recognize the multiple image features;

determining multiple vector elements representative of at least part of a vector version of the picture;

generating semantic metadata including associating the multiple vector elements with at least one image feature of the multiple image features; and

producing a vector image including the multiple vector elements and the associated semantic metadata.

17. The method as described in claim 16, wherein the associating comprises linking a vector element of the multiple vector elements with an indication of a corresponding image feature of the multiple image features.

18. The method as described in claim 16, wherein the associating comprises:

for each vector element of the multiple vector elements, determining a semantic element of the multiple semantic elements that corresponds to the vector element based on at least one image feature of the multiple image features; and

organizing the multiple vector elements into multiple semantic groups, each semantic group corresponding to a particular semantic element of the multiple semantic elements and including those vector elements that correspond to the particular semantic element.

19. The method as described in claim 16, further comprising:

manipulating the multiple vector elements of the vector image based on the associated semantic metadata.

20. The method as described in claim 16, further comprising:

displaying the at least part of the vector version of the picture using the multiple vector elements in conjunction with a visual indication of vector element groupings that are based on the associated semantic metadata.