Abstract: A document authentication method employs Krawtchouk decomposition to analyze and compare document images. When printing an original document, the original document image is segmented into image patches, which preferably correspond to individual symbols of the document. Krawtchouk decomposition is applied to each image patch. The image patches are classified into image patch classes using their Krawtchouk coefficients. The locations of all image patches belonging to each class are obtained and stored along with the Krawtchouk coefficients for each class. When authenticating a target document, the same segmentation, Krawtchouk decomposition and classification steps are applied to the target document image, and the locations of all image patches belonging to each class are obtained. The image patch classes and the locations of image patches belonging to each class for the original and target document image are compared to detect alterations present in the target document.

Abstract: A method for separating foreground and background contents in a document image is provided. The method first computes a pixel-wise map of maximal local features (e.g., local variance, local contrast, etc.), which is binarized to generate a mask for potential foreground. In order to utilize color information effectively, the local feature map is computed using all color channels of the image. Then the background image is obtained by inpainting the mask regions from the non-mask regions of the original document image. Adaptive thresholding is applied to the difference between the original document image and the background image to obtain the binary foreground image. Post-processing of the binary foreground image can further remove undesirable elements. Finally, a more accurate background image can be obtained by inpainting the original document image using the binary foreground image as a mask.

Abstract: A method for separating foreground and background contents in a document image is provided. The method first computes a pixel-wise map of maximal local features (e.g., local variance, local contrast, etc.), which is binarized to generate a mask for potential foreground. In order to utilize color information effectively, the local feature map is computed using all color channels of the image. Then the background image is obtained by inpainting the mask regions from the non-mask regions of the original document image. Adaptive thresholding is applied to the difference between the original document image and the background image to obtain the binary foreground image. Post-processing of the binary foreground image can further remove undesirable elements. Finally, a more accurate background image can be obtained by inpainting the original document image using the binary foreground image as a mask.

Abstract: A method for encoding and decoding color barcodes to increase their data capacity. The encoding steps include determining a shape, a foreground color and a background color for each data cell, wherein a combination of the shape, foreground and background colors for the data cell is chosen from a plurality of such combinations in accordance with a value of the digital data to be encoded; and coloring some pixels in the data cell with a foreground color and other pixels with a background color, in accordance with the shape, foreground and background colors for the data cell determined above. The decoding steps include segmenting the data cells, recognizing a shape, a foreground color of the shape and a background color of the data cell, and obtaining digital data from a combination of the shape and foreground and background colors in each data cell.

Abstract: A binary image downsampling method, including the steps of generating a gray-scale image from a binary image having a background and one or more foreground portions, locating skeleton pixels in the one or more foreground portions, manipulating values of certain foreground pixels in the gray-scale image such that the differences between the values of the skeleton pixels and the background pixels become more significant, downsampling the gray-scale image with the manipulated values of the certain foreground pixels, and generating a downsampled binary image from the downsampled gray-scale image.

Abstract: A method and program for encoding and decoding color barcodes to increase their data capacity. The encoding steps include determining a shape and a color for each data cell to encode digital data, wherein a combination of the shape and the color for the data cell is chosen from a plurality of combinations of shapes and colors in accordance with a value of the digital data to be encoded, and coloring a subset of the plurality of pixels in each data cell in accordance with the shape and the color for the data cell determined above. The decoding steps include segmenting the data cells in a color barcode, recognizing a shape formed by a subset of pixels in each data cell and the color of the shape, and obtaining digital data from a combination of the recognized shape and color in each data cell.

Abstract: A method for encoding and decoding color barcodes to increase their data capacity. The encoding steps include determining a shape, a foreground color and a background color for each data cell, wherein a combination of the shape, foreground and background colors for the data cell is chosen from a plurality of such combinations in accordance with a value of the digital data to be encoded; and coloring some pixels in the data cell with a foreground color and other pixels with a background color, in accordance with the shape, foreground and background colors for the data cell determined above. The decoding steps include segmenting the data cells, recognizing a shape, a foreground color of the shape and a background color of the data cell, and obtaining digital data from a combination of the shape and foreground and background colors in each data cell.

Abstract: A method for decoding digital data in a color barcode having a plurality of data cells, including the steps of: scanning the color barcode of the hardcopy document, separating color image of the color barcode into print primary color planes, computing peaks of each print primary color plane, projecting, for at least one of the print primary color planes, the data cells along a horizontal direction and a vertical direction at the peaks of the at least one of the primary color planes in each direction which represent data cell center locations respectively, and creating a grid where each of its intersection is a respective data cell center location, assigning a color to each grid intersection which corresponds to a respective data cell by examining values of the print primary color planes at such location, and decoding digital data from the data cells based on the respective color assigned to each data cell.

Abstract: A method is described for binarizing a gray scale document image, in particular, a document image containing both text and non-text contents. Phase congruency maps are calculated from the gray scale image, and used to segment the text and non-text areas of the gray scale image. The phase congruency maps are also used to extract long lines such as table lines, which can be optionally removed from the image. The text and non-text areas of the gray scale image are divided into image patches; for the text areas, connected components obtained from the phase congruency map are used to generate image patches, so that each image patch contains a text character. The image patches are binarized individually using individual threshold values, and then combined to generate a binary image of the gray scale image. The method can also be used for purposes of OCR or document authentication.

Abstract: A method for authenticating a printed document which carries barcode that encode authentication data, including word bounding boxes for each word in the original document image and data for reconstructing the original image. The printed document is scanned to generate a target document image, which is then segmented into text words. The word bounding boxes of the original and target document images are used to align the target document image. Then, each word in the original document image is compared to corresponding words in the target document image using word difference map and Hausdorff distance between them. Symbols of the original document image are further compared to corresponding symbols in the target document image using feature comparison, symbol difference map and Hausdorff distance comparison, and point matching. These various comparison results can identify alterations in the target document with respect to the original document, which can be visualized.

Abstract: A method for compressing a bi-level document image containing text is disclosed. The document image is segmented into symbol images each representing a letter, numeral, etc. in the document. The symbol images are classified into a plurality of classes, each class being associated with a template image and a class index. Classification is done by comparing each symbol to be classified with template of existing classes, using a number of image features including zoning profiles, side profiles, topology statistics, and low-order image moments. These image features are compared using a tolerance based method to determine whether the symbol matches the template. After classification, certain classes that have few symbols classified into them may be merged with other classes. In addition, the template images of the classes are down-sampled, where the final sizes of the template images are dependent on the likelihood of confusion of the template with other templates.

Abstract: A binary image downsampling method, including the steps of generating a gray-scale image from a binary image having a background and one or more foreground portions, locating skeleton pixels in the one or more foreground portions, manipulating values of certain foreground pixels in the gray-scale image such that the differences between the values of the skeleton pixels and the background pixels become more significant, downsampling the gray-scale image with the manipulated values of the certain foreground pixels, and generating a downsampled binary image from the downsampled gray-scale image.

Abstract: A method for decoding digital data in a color barcode having a plurality of data cells, including the steps of: scanning the color barcode of the hardcopy document, separating color image of the color barcode into print primary color planes, computing peaks of each print primary color plane, projecting, for at least one of the print primary color planes, the data cells along a horizontal direction and a vertical direction at the peaks of the at least one of the primary color planes in each direction which represent data cell center locations respectively, and creating a grid where each of its intersection is a respective data cell center location, assigning a color to each grid intersection which corresponds to a respective data cell by examining values of the print primary color planes at such location, and decoding digital data from the data cells based on the respective color assigned to each data cell.

Abstract: An image based document index and retrieval method is described. During document indexing, each source document is analyzed to generate index information at document, page, region and unit levels. Region and unit level index information is generated by segmenting each text region into units, constructing unit length or unit density histograms, and analyzing the units in a few most frequent bins of the histogram. The index information and the source document images are stored in a database. During document retrieval, a target document is analyzed to generate target index information in the same way as during document indexing. The target index information is compared to stored index information in a progressive manner (from higher to lower levels) to identify source documents with index information that matches the target index information. Fuzzy logic is used in the comparison steps to increase the robustness of the document retrieval.

Abstract: A method is described for binarizing a gray scale document image, in particular, a document image containing both text and non-text contents. Phase congruency maps are calculated from the gray scale image, and used to segment the text and non-text areas of the gray scale image. The phase congruency maps are also used to extract long lines such as table lines, which can be optionally removed from the image. The text and non-text areas of the gray scale image are divided into image patches; for the text areas, connected components obtained from the phase congruency map are used to generate image patches, so that each image patch contains a text character. The image patches are binarized individually using individual threshold values, and then combined to generate a binary image of the gray scale image. The method can also be used for purposes of OCR or document authentication.

Abstract: A pair of spectacles that can automatically change its power so that a fixation region of interest (ROI) of the user is always in focus. The automatic accommodative spectacle device includes focusing elements, scene sensors, scene analyzer, focus engine, focusing element controller, and power supply. The scene analyzer determines the fixation ROI in the scene by analyzing the scene images. The fixation ROI is used to determine powers for the focusing elements in order to bring the fixation ROI into focus. The focusing element controller carries out the needed optical power adjustment to apply to the focusing elements. Optional light sources may be provided to the scene sensors. Additional optional eye sensor(s), eye light source and line of sight detector can be used to help the scene analyzer extract the fixation ROI.

Abstract: A document authentication method employs Krawtchouk decomposition to analyze and compare document images. When printing an original document, the original document image is segmented into image patches, which preferably correspond to individual symbols of the document. Krawtchouk decomposition is applied to each image patch. The image patches are classified into image patch classes using their Krawtchouk coefficients. The locations of all image patches belonging to each class are obtained and stored along with the Krawtchouk coefficients for each class. When authenticating a target document, the same segmentation, Krawtchouk decomposition and classification steps are applied to the target document image, and the locations of all image patches belonging to each class are obtained. The image patch classes and the locations of image patches belonging to each class for the original and target document image are compared to detect alterations present in the target document.

Abstract: A method is described to obtain a binary image from the print-and-scan process to best match the known original. A point-spread function (PSF) of the PAS process is first obtained from its knife-edge responses, and deblurring is carried out on the scanned images using deconvolution. After image deskewing and preliminary registration, a supervised adaptive thresholding procedure is utilized to binarize the scanned image such that a measure of difference (e.g. the Euclidean distance) between the original and binarized images is minimized. The supervised adaptive thresholding procedure divides the scanned images into many rectangular sub-images. Otsu's method is used to find a starting threshold for each scanned sub-image. An optimal threshold is found around the Otsu's threshold via iterative search to minimize the measure of difference between the original sub-image and scanned sub-image. The sub-images are binarized using the optimal threshold. This method may be used in document authentication.

Abstract: An image based document index and retrieval method is described. During document indexing, each source document is analyzed to generate index information at document, page, region and unit levels. Region and unit level index information is generated by segmenting each text region into units, constructing unit length or unit density histograms, and analyzing the units in a few most frequent bins of the histogram. The index information and the source document images are stored in a database. During document retrieval, a target document is analyzed to generate target index information in the same way as during document indexing. The target index information is compared to stored index information in a progressive manner (from higher to lower levels) to identify source documents with index information that matches the target index information. Fuzzy logic is used in the comparison steps to increase the robustness of the document retrieval.

Abstract: A pair of spectacles that can automatically change its power so that a fixation region of interest (ROI) of the user is always in focus. The automatic accommodative spectacle device includes focusing elements, sensors, line of sight detector, focus engine, focusing element controller, and power supply. The line of sight detector determines the line of sight for the left and right eyes of the user using data from the sensors. The focus engine uses the lines of sight for left and right eyes to determine the user's fixation ROI. The fixation ROI is used to determine powers for the focusing elements in order to bring the fixation ROI into focus. The focusing element controller carries out the needed optical power adjustment to apply to the focusing elements. Optional light sources may be provided.