Canon Kabushiki Kaisha

Yu-Ling Chen

James Philip Andrew

Steven Richard Irrgang

James Austin Besley

Eric Wai-Shing Chong

Trevor Lee McDonell

David Robert James Monaghan

Dixon De Sheng Deng

Peter William Mitchell Ilbery

Michael John Webb

Timothy John Wark

Michael Jan Lawther

Kai Huang

Ian Geoffrey Combes

Paul Raymond Higginbottom

Yoong-Chert Foo

Ping Liu

Julie Yan Zhu

Peter Leslie Bell

Anthony John Arnold

Kailash Nandish Ballu

Disclosed is a method of generating an electronic document having an enclosed region with a fill color. The method receives a digital representation of a source document, the source document containing an enclosed region and at least one corresponding background region, the enclosed region overlapping at least a portion of the background region. A fill color is determined for the enclosed region from the digital representation as is a reference background color for the enclosed region from the corresponding background region of the digital representation. The method assigns a transparent fill color to the enclosed region based on a comparison of the determined fill color for the enclosed region with the reference background color and stores enclosed region with the transparent fill color to generate an electronic document.

Determining transparent fills based on a reference background colour

A method () is disclosed for modifying colors of text extracted from an image of a document page to improve visual quality. The document image comprises text and image regions of various colors. The method divides () the document image into a plurality of connected components, the connected components comprising pixels of similar color that are closely spatially located, and selects from the image a plurality of the connected components that represent text. The plurality of text connected components are grouped to form a logical structure grouping of said page comprising one or more text blocks wherein each said text block represents a text line or paragraph. One or more logically consistent color sets are then formed () for each text block based on the colors of the connected components in the text block. The logically consistent color sets represent all text of visually similar color of the corresponding text line or paragraph. The method then applies color enhancement () to the logically consistent color sets to enhance visual quality and correct image defects in the document image.

Colour correcting foreground colours for visual quality improvement

Disclosed is a method of classifying segmented contents of a scanned image of a document. The method comprise partitioning the scanned image into color segmented tiles at pixel level. The method then generates superpositioned segmented contents, each segmented content representing related color segments in at least one color segmented tile. Statistics are then calculated for each segmented content using pixel level statistics from each of the tile color segments included in segmented content, and then determines a classification for each segmented content based on the calculated statistics. The segmented content may be macroregions. The macroregions may form part of a multi-layered document representation of the document. Each of a plurality of tiles of predetermined size of the image are converted into a representation having a plurality of layers, the representation corresponding to at least one said tiles comprising multiple colored layers, each tile comprising a superposition of the corresponding colored layers. For each of the colored layers, merging is performed with adjacent ones of the tiles, thereby generating a multi-layered document representation.

Colour document layout analysis with multi-level decomposition

Methods (), apparatuses, and computer program for finding a region containing text in a color bitmap image comprising pixels are provided. Connected components (CCs) are generated () from the color bitmap image by grouping substantially similarly colored and adjacent pixels. Independently of color, which of the connected components are text connected components are determined () dependent upon attributes of the generated connected components. For each text CC) a value is assigned () to each tile that comprises at least a portion of the text connected component. The value indicates the presence of text overlapping the tile. Each tile comprises pixels of the color bitmap image. The region containing text in the color bitmap image is determined () dependent upon the assigned values of the tiles.

Finding text regions from coloured image independent of colours

A method detects noise in a document image. The image includes pixels representing at least text and image data forming the document. The noise includes at least one of halftone, bleeding and texture noise. The method partitions the image into tiles, formed of pixels, determines a set of dominant colors for each tile, associates each pixel with one of the determined dominant colors, and then determines for each dominant color a fragmentation statistic, and applies a threshold to each statistic to detect the noise in the corresponding dominant color in the tile. Detected noise in the image is reduced by (a) merging colors with noise to one color without noise if the noise is halftone noise or texture noise; and (b) removing the colors with noise if the noise is bleeding noise. The pixels of the tile are then quantized to those remaining dominant colors of the set.

Block-based noise detection and reduction method with pixel level classification granularity

A Hybrid Compressor and method of compressing a tile of pixels within an image are disclosed. An Edge Processor of the Hybrid Compressor analyzes pixel runs of the tile and generates boundary data defining one or more regions comprising pixels of substantially identical color. A Palette Processor of the Hybrid Compressor generates data defining a color value for each region. An Image Processor of the Hybrid Compressor generates a representation of pixels not included in the one or more regions.

Method and apparatus for hybrid image compression

A method of text extraction in color compound documents is described. The method connects similarly colored pixels of an image of a color compound document into connected components (CCs); classifies each CC as either text or non-text; refines the text CC classification for each text CC using global color context statistics; groups text CCs into text blocks; recovers misclassified non-text CCs into a nearby text block; and removes extraneous CCs from each text block using local color context statistics to thereby provide the extracted text in the text blocks. Also described is a method of locating graphics objects in a color compound document image.

Object extraction in colour compound documents

A method (), an apparatus, and a computer program product for automatically producing a compact representation of a colour document are disclosed. In the method, a digital image of a colour-document page is segmented () into connected components in one-pass, block raster order. The digital image of the page is partitioned into foreground and background images using layout analysis () based on compact, connected-component statistics of the whole page. At least one portion of the background image where at least one portion of the foreground image obscures the background image is inpainting () in one-pass block raster order. The foreground and background images are combined () to form a compact document. A method, an apparatus, and a computer program product for segmenting a digital image comprising a plurality of pixels are also disclosed.

Segmenting digital image and producing compact representation

A method of rendering a graphical image described by an expression tree is disclosed. The expression tree has one or more parent nodes and one or more leaf nodes, with each parent node representing a graphical operator and having branches to respective descendent nodes. Each of the leaf nodes represents a graphic object. The method determines regions of one or more pixel locations in an output space of the image, with each region corresponding to a combination in the image of one or more of the graphical objects. Next, for each of the regions, an activity state of corresponding parent nodes is determined using a predefined function for each corresponding operator. The method then generates, in parallel, for each active one of the operators, compositing instructions and pixel generation instructions. The compositing instructions are stored in an operator queue and pixel colors are generated using the pixel generation instructions, with the pixel colors being buffered in an operand queue. Lastly, the generated pixel colors are composited, for each pixel in the region, using the operators defined in the operator queue, to output a composited pixel value for each pixel in the region.

Tree-based compositing system

Disclosed is a method of detecting a predetermined mark embedded in an image. The mark () comprises a predetermined arrangement of a plurality of elements (), each element having a predetermined colour characteristic and predetermined shape. The method processes () the image to provide an encoded representation of the image at a predetermined resolution (eg. 200 dpi). The coordinate positions for substantially each of the elements of the mark embedded in the image are detected (), wherein the detection is characterised by applying at least one mask () to substantially each pixel of the encoded representation. From the coordinate positions, a set of spatial features representing a spatial arrangement of the detected elements are then determined (). The determined set of spatial features is then compared () to a known set of spatial features to provide a confidence level measure for a degree of matching between the known set and the determined set of spatial features.

Method and apparatus for mark detection

A method, an apparatus () and a computer program product for entropy coding of discrete wavelet transform (DWT) coefficient bits that are arranged into code blocks and coded in bitplane order using three coding passes for each bitplane are disclosed. In the method, transform coefficients of a code block () in sign-magnitude form are pre-analyzed. Statistical data about the coefficients is stored, preferably with the coefficients. Significance state data (), coded data (), magnitude refinement data (), bit data (), and sign data () for the code block are buffered. More preferably, the buffering is implemented using a FIFO (), located between a context generation module () and an arithmetic coder (). Based upon the statistical data, at least one command for at least one sequence of bit and context pairs is generated for arithmetic encoding.

Entropy coding

Disclosed is method of detecting a predetermined mark ( ) forming part of an image. A digitized form ( ) of the image ( ) comprising a plurality of picture elements is provided and within a current region of the image a set of features is determined. The features typically comprise a dominant color feature ( ) within the current region; and at least one moment feature ( ) based on picture elements pertaining to the predetermined mark within the current region. Each determined feature is compared ( ) with a corresponding predetermined feature ( ) of the predetermined mark. Using the comparison, a confidence level for a detection of said predetermined mark within the current region calculated. Based on the confidence level, whether or not said predetermined mark is detected in the current region can then be determined. Adjunct to the above method, also disclosed are methods for providing multiple encoded representations ( ) of the image, detecting a center location ( ) of a predetermined pattern in the image, and determining ( ) a moment feature of a portion of the image.

Image pattern detection method and apparatus

A compositing system ( ) that comprises an optimized alpha-channel compositing module ( ) is disclosed. The compositing system ( ) optimizes the range correction in the alpha-channel compositing module ( ) by using a constant divider ( ) and delaying the correction of an alpha value until after the summation of the alpha values from all contributing regions. Uncorrected opacity values are used to index a look-up table in order to reduce the data processing path to a multiplier ( ) for processing after premultiplication of a color values from all contributing regions.

Alpha-channel compositing system

The apparatus comprises a discrete wavelet transform (DWT) engine, a code block manager, and an entropy encoder. The code block manager comprises at least one controller, which losslessly compresses the transform coefficients and stores them in a code block storage for buffering. The entropy coder comprises at least one entropy encoder, each comprising a decoder for decoding the losslessly compressed transformed coefficients prior to entropy encoding.

Growing community of inventors

Epping, Australia

Yu-Ling Chen