Ricoh Company, Ltd.

Google Inc.

Maya Rani Gupta

Hiroshi Takahashi

Kathrin Berkner

Michael J Gormish

David Geoffrey Stork

Martin P Boliek

Charles Joseph Rosenberg

Samy Bengio

Thomas J Duerig

Zhen Hao Zhou

Neil G Alldrin

Maks Ovsjanikovs

Methods, systems and apparatus for assigning labels to images. In general, in one aspect, a method includes determining, for an image, a first set of labels, each label being determined to be indicative of subject matter of the image based on content feature values of the image, for each label in the first set of labels, determining a second set of labels, each label in the second set of labels determined to be semantically related to the label in the first set of labels, assigning a score to each label in the second sets of labels, and based on the scores assigned to each label in the second sets of labels, assigning one or more of the labels in the second sets of labels to the image.

Assigning labels to images

A method and an apparatus to perform replacement halftoning is disclosed. In one embodiment, the method comprises dithering multiple color channels of a pixel of an image to generate a plurality of halftone outputs. Each of the halftones corresponds to a distinct color channel of the pixel. The method further includes creating a virtual color channel from two or more color channels of the plurality of color channels and applying error diffusion to the virtual color channel to generate an error diffusion halftone outputs, and then replacing of the dither output error diffusion halftone outputs.

Replacement halftoning

A method and apparatus for block quantization for color halftoning is described. In one embodiment, the method comprises dividing an input image and an output image into blocks, wherein each block in the output image corresponds to one block in the input image. The method also includes dividing each block into subblocks if an edge exists in said each block, calculating a color average of each input block, and calculating a set of output colors for each block in the output image to match the color average of its corresponding block in the input image.

Block quantization method for color halftoning

A method and apparatus for performing error diffusion of a multiresolutional representation of an image is disclosed herein. In one embodiment, an image is received as a pixel representation. The image is converted from a pixel representation to a multiresolutional representation having multiresolutional coefficients. Each multiresolutional coefficient is quantized and an error is passed to one or more other multiresolutional coefficients as specified by an error filter. Finally, the resulting quantized image is produced as output.

Error diffusion of multiresolutional representations

A method and apparatus for performing ink reduction error diffusion are described. In one embodiment, the method comprises performing vector error diffusion and mapping anchors to printing colors, where performing vector error diffusion includes quantizing each input pixel value to a closest anchor in an anchor set, calculating errors based on differences between the input pixel values and the anchors, and performing error filtering based on calculated errors.

Ink reduction error diffusion

A method and apparatus for performing color halftoning using block quantization is disclosed. In one embodiment, the method comprises dividing an input image into a plurality of blocks, clustering pixels in each of the plurality of blocks into a plurality of color groups, block quantizing each of the plurality of color groups for each of the plurality of blocks to a representative color value, and creating an output halftone image having a distinct block associated with each block in the plurality of blocks, with the representative color value for each group in each distinct block being approximated over a spatial area covered by the group in said each distinct block.

Growing community of inventors

Menlo Park, CA, United States of America

Maya Rani Gupta