Abstract: Methods and apparatus for: identifying perimeter lines in an image, e.g., document border lines forming a non-rectangular quadrilateral corresponding to a scanned document, determining document corner points, and correcting the input image based on determined corner points are described. Lines are identified in the input image and a line weight is generated for each identified line. The identified lines are classified into two different groups based on direction, e.g., predominately vertical or predominately horizontal. For each group at least two, and sometimes more, candidate perimeter lines are determined to be kept, based on line weight and line length, and other lines in the group are eliminated. Two best candidate perimeter lines are identified from each group, e.g., based on distance between the remaining lines in the group. Four document corner points are determined based on the remaining lines in the two groups.

Abstract: Methods and apparatus for identifying lines in an image are described. An image to be processed is divided into a plurality of tiles, and processing is performed on a per tile basis. Lines are identified in tiles and a weight is assigned to each line based on among other things, the length of the line. Quantized first and second parameter values, e.g., values defining where lines enters and leave an area, are used in defining the identified lines. A set of lines is selected based on the weight information and output or used in image processing the image including the lines.

Abstract: Methods and apparatus for detecting the presence and location of punch holes in a scanned image are described. The punch hole detection methods and apparatus rely at least in some embodiments on whether a portion of the scanned image, referred to as a component, corresponds to a punch hole by comparing one or more characteristics of the component such as its circularity, aspect ratio, black to white pixel ratio, density, height and/or width to one or more thresholds before making a decision as to whether or not the component is a punch hole. In some embodiments the components which pass component-level checks are grouped and one or more group-level checks are performed on the components to determine if the components are punch holes. Once a punch hole is detected, in some embodiments, the image is processed to remove the detected punch hole.

Abstract: Methods and apparatus for processing one or more images, e.g., images representing pages including text, to detect and in some instances correct the orientation of the page. In some embodiments the methods and apparatus for processing image data comprise generating a histogram of foreground pixel counts corresponding to a current line of text of the image being processed with the foreground pixel counts corresponding to different rows of pixels corresponding to the current line of text and identifying based on statistical analysis of the generated histogram whether the current page of text is oriented in an inverted or non-inverted position. In some embodiments analysis is performed on multiple lines of text with cumulative statistics being used in to determine the orientation of the page. In some embodiments, a page whose orientation is determined to be upside down is re-oriented to be right-side up.

Abstract: Methods and apparatus for detecting the presence and location of punch holes in a scanned image are described. The punch hole detection methods and apparatus rely at least in some embodiments on whether a portion of the scanned image, referred to as a component, corresponds to a punch hole by comparing one or more characteristics of the component such as its circularity, aspect ratio, black to white pixel ratio, density, height and/or width to one or more thresholds before making a decision as to whether or not the component is a punch hole. In some embodiments the components which pass component-level checks are grouped and one or more group-level checks are performed on the components to determine if the components are punch holes. Once a punch hole is detected, in some embodiments, the image is processed to remove the detected punch hole.

Abstract: Methods and apparatus for binarizing images represented by sets of multivalent pixel values in a computationally efficient manner are described In a grayscale image to be binarized, one group of pixel values represents “foreground”, e.g., text to be converted to black, while another group represents a shaded “background” region to be converted, e.g., to white. The difference between foreground and background is often a function of the scale of the image components, e.g., text and/or other images. Filters in the form of morphological operators, computationally efficient quick-open and quick-close morphological operators are employed to binarize images, e.g., grayscale images. The methods and apparatus effectively handle both smooth and sharp image background structures in a computationally efficient manner.

Abstract: Methods and apparatus for binarizing images represented by sets of multivalent pixel values in a computationally efficient manner are described In a grayscale image to be binarized, one group of pixel values represents “foreground”, e.g., text to be converted to black, while another group represents a shaded “background” region to be converted, e.g., to white. The difference between foreground and background is often a function of the scale of the image components, e.g., text and/or other images. Filters in the form of morphological operators, computationally efficient quick-open and quick-close morphological operators are employed to binarize images, e.g., grayscale images. The methods and apparatus effectively handle both smooth and sharp image background structures in a computationally efficient manner.

Abstract: Methods and apparatus for comparing blank forms represented in a digital format to digitized filled-in forms are described. Different errors are attributed different weights when attempting to correlate regions of blank and filled-in forms. Foreground pixels in the blank form which are not found in a corresponding portion of a filled-in form are attributed greater error significance than foreground pixels, e.g., pixels which may correspond to added text, found in the filled-in form which correspond to a background pixel value in the blank form. A virtual filled-in form including content, e.g., pixel values, from the filled-in form is generated from the content of the filled-in form and pixel value location mapping information determined from comparing the blank and filled-in forms. Various analysis is performed on a block basis, but in some embodiments the final pixel mapping to the virtual form is performed on a pixel by pixel rather than a block basis.

Abstract: Color image transform coding methods and apparatus which take advantage of redundant information which is normally present in the luminance and chrominance components of a color image are described. The transform coding method of the present invention, can normally provide increased compression and/or better image quality at a given coding data rate than conventional JPEG or JPEG 2000 coding which does not use the transform of the present invention as part of the color component coding process. These benefits are normally achieved without generating more coefficients than would be generated in the case of more conventional coding such as JPEG or JPEG 2000 coding. In fact, the transform of the present invention can be applied to increase the coding efficiency of coding systems which perform spatial transforms and modeling and entropy coding operations of the type normally used in JPEG and/or JPEG 2000 coding systems.

Abstract: Improved methods and apparatus for implementing JPEG 2000 MQ encoding operations are described. The following features of the invention may be used alone or in combination to implement JPEG 2000's coefficient bit modeling 1) Lists of coefficients to be processed in one or more subsequent coding passes are generated as coefficients are processed. The list is allows processing in the subsequent coding pass to be limited to coefficients on the list; and 2) Generation, updating and use of a neighborhood descriptor value table that stores a neighborhood descriptor value for each coefficient of a block being processed. The neighborhood descriptor values provide information about the state of the coefficient and its adjacent neighbors. Neighborhood descriptor values are used to access a lookup table that provides one or more context values and/or sign value to be used in MQ encoding.