Abstract: Registration data embedded in an image is used to determine and compensate for geometric transformation of the image. The registration data may include frequency domain features of the image that are used to compute the rotation and/or scaling of the image. The registration method may be used in conjunction with digital watermarking and other signal processing applications.

Abstract: The invention provides an improved method and system for processing mail items on bar code sorting machines. Such a method includes the steps of scanning a surface of each mail piece with at least one imaging camera to obtain at least one image of the surface represented by image data, analyzing the image data to locate a bar code in the image, and analyzing the bar code to recognize a first destination code. If the first destination code meets predetermined criteria, then the mail piece is processed further in a manner dependent on the first destination code. If the first destination code cannot be recognized or fails to meet the predetermined criteria, the image data is analyzed to locate destination address lines in the image, and the address lines are analyzed to determine a second destination code. The mail piece is sorted according to the second destination code, or a result determined by arbitrating the first and second destination codes.

Abstract: A first masking image of an input image is created by assuming that pixels in an area including the area of a target in all sample images are valid pixels while pixels in the remaining area are invalid pixels. The first masking image is used in a first comparison between the input image and each sample image to select an image that provides a best match as a provisionally selected sample image. In a second comparison, a second masking image corresponding to the provisionally selected sample image is used to select a final sample image. The second masking image is created from an individual sample image by assuming that pixels in an area including the area of the target are valid pixels while pixels in the remaining area invalid pixels.

Abstract: A method for assisting a visually impaired person with a document job. The method includes scanning a document that includes at least one sheet, detecting at least one characteristic of the scanned document, and providing a signal identifying the at least one characteristic.

Abstract: A method is for calibration of perfusion parameter images. The image data from tomographic imaging measurements are analyzed using a perfusion model, and perfusion parameter values obtained from the analysis are calibrated with a calibration factor to a physiological normal value. To perform the calibration, an image area which contains several different tissue compositions, and in which approximately normal perfusion conditions can be assumed, is selected from the obtained perfusion parameter image. A frequency analysis of the perfusion parameter values contained in this image area is then performed, and at least one frequency interval is selected which corresponds at least approximately to a frequency interval of the occurrence of perfusion parameter values of a known tissue composition.

Abstract: This disclosure presents an improved method for registering anatomical medical images and functional medical images. The example deals with the registration of x-ray computer tomography images with positron emission tomography images. The process is characterized by clinically useful registration with minimal computer calculations and minimal delay for computation. A nonrigid B-Spline free form deformation is used in both a preliminary coarse registration and the finished fine registration. Additional steps are used to insure accurate and complete registrations.

Abstract: A computer-implemented method of planning orthopaedic surgery comprises providing a library of templates representing orthopaedic prostheses, displaying and scaling one or more patient images such as X-ray images, allowing a user to reconfigure geometrical constructs displayed over the images to match the construct to anatomical features shown in the image; and selecting one or more templates from the library in accordance with parameters of the reconfigured constructs. The templates correspond to the orthopaedic prosthesis or prostheses which are most suitable for the patient. Hip replacement surgery can be planned using a single patient image. Knee surgery can be planned using two patient images showing different views of the anatomical features, in which case geometrical constructs for use with each view are provided. The library of templates is accessible via the Internet so as to be accessible by users in any location and readily updateable.

Abstract: A method and related apparatus for sensing selected emotions or physical conditions in a human subject. The technique employs a two-dimensional camera to generate a facial image of a human subject. Then, an image processing module scans the image to locate the face position and extent, and then scans for selected critical areas of the face. The size and activity of the selected critical areas are monitored by comparing sequential image frames of the subject's face, and the areas are tracked to compensate for possible movements of the subject. The sensed parameters of the selected critical areas are compared with those stored in a database that associates activities of the critical areas with various emotional and physical conditions of the subject, and a report or assessment of the subject is generated.

Abstract: A method of analyzing a bundle of banknotes. The method includes the steps of providing a bundle of banknotes, which bundle comprises at least one surface defined by the edges of banknotes, illuminating the surface of said bundle, providing a two-dimensional image of the bundle by making use of an optical sensor, and providing an output signal that represents the result of the analysis.

Abstract: An object positioning solves said problems encountered in machine vision, which employs electro-optic (EO) image sensors enhanced with integrated laser ranger, global positioning system/inertial measurement unit, and integrates these data to get reliable and real time object position. An object positioning and data integrating system comprises EO sensors, a MEMS IMU, a GPS receiver, a laser ranger, a preprocessing module, a segmentation module, a detection module, a recognition module, a 3D positioning module, and a tracking module, in which autonomous, reliable and real time object positioning and tracking can be achieved.

Abstract: A system and method in a multi-channel detection system for multi-rate filter bank applications for converting overlapping rectangular two-dimensional (2D) regions into a new set of non-overlapping rectangular regions for the efficient reconstruction of a signal wherein each non-overlapping region has a maximum extent in a major dimension is described. Overlapping regions are split into marked regions in a non-uniform grid and merged along the major dimension and along the minor dimension to form non-overlapping regions wherein no two non-overlapping rectangular regions have an adjacent edge orthogonal to the major dimension thereby increasing the efficiency of data compression and reducing error-rates.

Abstract: Methods for adapting the operation of a system in response to the detection of embedded watermarks from a digital host content are provided. A digital host content is received and examined for the presence of watermarks. In response to the detection of embedded watermarks and in accordance with the value, type, density or spacing of the detected watermarks, one or more system reactions may take place. These reactions include conditionally allowing the system to resume its normal operation, prohibiting the system from resuming its normal operation, degrading the quality of the digital host content or changing the security status of the digital host content. In response to the extraction of weak watermarks that do not meet the desired system requirements, the extraction operation may be modified or extended to enable the detection of strong watermarks.

Abstract: A method for immediately determining incorrect pixel values of a captured image on an optical mouse, which determines a quality of an image in a video stream captured by the optical mouse and accordingly finds a motion distance for the optical mouse. The method first computes a set of match values for possible motion vectors of a sample image with respect to a reference image and applies a filtering operation to compute a respective smooth coefficient for each match value. The method further computes a locally minimum number in the set of match values according to a first rule, such that the sample image is regarded as a qualified image when the locally minimum number is smaller than a threshold and accordingly a global minimum in the set of match values is found as a motion vector of the sample image.

Abstract: A method and apparatus for determining camera pose from point correspondences. Specifically, an efficient solution to the classical five-point relative pose problem is presented. The problem is to find the possible solutions for relative camera motion between two calibrated views given five corresponding points. The method consists of computing the coefficients of a tenth degree polynomial and subsequently finding its roots. The method is well suited for numerical implementation that also corresponds to the inherent complexity of the problem. The method is used in a robust hypothesize- and-test framework to estimate structure and motion in real-time.

Abstract: Techniques for computing a globally consistent set of image feature correspondences across a wide range of viewpoints suitable for interactive walkthroughs and visualizations. The inventive approach takes advantage of the redundancy inherent in a dense set of images captured in a plane (or in higher dimensions, e.g., images captured in a volume, images captured over time, etc). The technique may detect features in a set of source images and track the features to neighboring images. When features track to the same position in the same image, they are flagged as potential correspondences. Among the potential correspondences, the technique selects the maximal set using a greedy graph-labeling algorithm (e.g., best-first order). Only correspondences that produce a globally consistent labeling are selected. After globalization is done, a set of features common to a group of images can be quickly found and used to warp and combine the images to produce an interpolated novel view of the environment.

Abstract: Methods and apparatus are provided for subjecting an original, pixel domain image to an Integer Wavelet Transform (IWT) to obtain a matrix of wavelet coefficients; selecting at least one bit plane between a least significant bit plane and a most significant bit plane of the matrix of wavelet coefficients; compressing the at least one selected bit plane to produce free space in the at least one selected bit plane; embedding hidden data in the free space of the at least one compressed bit plane; and subjecting the at least one embedded bit plane to an Inverse IWT to produce a marked pixel domain image. Methods and apparatus are also provided for reversing this process.

Abstract: A method for detecting, quantifying, staging, reporting, and/or tracking of a disease includes providing analysis software configured to detect, quantify, stage, report, and/or track a disease utilizing images of a patient. The analysis software is executable on a personal computer of a patient. Patients are then imaged utilizing a medical imaging apparatus and medical images of the patient produced by the imaging apparatus are downloaded to the personal computer of the patient. The imaging and downloading are repeated a plurality of times at intervals selected to provide the analysis software with sufficient images to detect, quantify, stage, report, and/or track the disease in the patient.

Abstract: A for embedding a digital watermark on a wavelet lowest subband includes: setting a DC component region of a multi-stage wavelet-transformed original copy image to a watermark embedment region, and high-frequency filtering an original picture LLn of the embedment region; generating index information for designating a pixel position, and a watermark sequence to be embedded; calculating an embedment strength ? for each position of the watermark embedment region; in case the watermark sequence is sequentially embedded on an embedded position designated by the index information, mutually comparing the original picture LLn coefficient value for each embedded position with a mirror picture LLn? coefficient value, and then altering the original picture LLn coefficient value; and in case the original picture LLn coefficient value is differentiated above a predetermined value with reference to the coffesponding embedment strength ?, skipping the watermark embedment for the position.

Abstract: A method for creating a message template used for embedding hidden messages, the method includes the steps of determining a message template performance metric comprising a dispersal measure having both a spatial domain function and a frequency domain function; developing a numerical optimization algorithm containing the message template performance metric as a basis for optimization; determining the message template geometric configuration comprising: (i) determining a message template capacity; (ii) determining a message template area; and applying the numerical optimization algorithm to the message template geometric configuration which results in an optimal message template.

Abstract: Methods and apparatus for digitally enhancing images are disclosed. In one embodiment, a method includes performing a wavelet transformation process on an acquired image to provide a low pass spatial frequency. A dynamic range and a mixing value are then determined from the low pass spatial frequency, and the mixing value is applied to provide a transformed output image. The transformed output image is then reformatted to provide a mixed output image, and the mixed output image is subtracted from the acquired image to provide an enhanced image. In an alternate embodiment, a method includes analyzing an acquired image to provide a high sensitive low light image and a low sensitive high light image, and then summing the high sensitive low light image and the low sensitive high light image to create an enhanced image.