Abstract: A digital image acquisition system having no photographic film, such as a digital camera, has a flash unit for providing illumination during image capture and a red-eye filter for detecting a region within a captured image indicative of a red-eye phenomenon, the detection being based upon a comparison of the captured image and a reference image of nominally the same scene taken without flash. In the embodiment the reference image is a preview image of lower pixel resolution than the captured image, the filter matching the pixel resolutions of the captured and reference images by up-sampling the preview image and/or sub-sampling the captured image. The filter also aligns at least portions of the captured image and reference image prior to comparison to allow for, e.g. movement in the subject.

Abstract: A smart-focusing technique includes identifying an object of interest, such as a face, in a digital image. A focus-generic classifier chain is applied that is trained to match both focused and unfocused faces and/or data from a face tracking module is accepted. Multiple focus-specific classifier chains are applied, including a first chain trained to match substantially out of focus faces, and a second chain trained to match slightly out of focus faces. Focus position is rapidly adjusted using a MEMS component.

Abstract: A method of generating one or more new spatial and chromatic variation digital images uses an original digitally-acquired image which including a face or portions of a face. A group of pixels that correspond to a face within the original digitally-acquired image is identified. A portion of the original image is selected to include the group of pixels. Values of pixels of one or more new images based on the selected portion are automatically generated, or an option to generate them is provided, in a manner which always includes the face within the one or more new images. Such method may be implemented to automatically establish the correct orientation and color balance of an image. Such method can be implemented as an automated method or a semi automatic method to guide users in viewing, capturing or printing of images.

Abstract: A wide field angle camera module that incorporates field curvature correction suitable for an electronic device is described herein. By way of example, the wide field angle camera module comprises micro optical lenses adapted to capture field angles of greater than thirty degrees, and is further adapted to provide field curvature correction for a range of object distances, and for a range of zoom configurations include 1×, 2× and 3× zoom configurations. Furthermore, the field correction can be implemented alone or in conjunction with signal processing that corrects barrel distortion introduced by the micro optical lenses. In particular aspects, the field correction can be automated based on image quality analysis to provide a high quality image for object distances substantially 30 centimeters or larger.

Abstract: A method and system for detecting facial expressions in digital images and applications therefore are disclosed. Analysis of a digital image determines whether or not a smile and/or blink is present on a person's face. Face recognition, and/or a pose or illumination condition determination, permits application of a specific, relatively small classifier cascade.

Abstract: A method of detecting a face in an image includes performing face detection within a first window of the image at a first location. A confidence level is obtained from the face detection indicating a probability of the image including a face at or in the vicinity of the first location. Face detection is then performed within a second window at a second location, wherein the second location is determined based on the confidence level.

Abstract: A portable digital image acquisition device includes multiple lenses and/or multiple flashes. A main digital image and first and second reference images are acquired. The first and second reference images are acquired with different flash-lens combinations that have different flash-lens distances. One or more flash eye defects are detected and corrected in the main image based on analysis of the first and second reference images.

Abstract: An image processing apparatus for tracking faces in an image stream iteratively receives an acquired image from the image stream including one or more face regions. The acquired image is sub-sampled at a specified resolution to provide a sub-sampled image. An integral image is then calculated for a least a portion of the sub-sampled image. Fixed size face detection is applied to at least a portion of the integral image to provide a set of candidate face regions. Responsive to the set of candidate face regions produced and any previously detected candidate face regions, the resolution is adjusted for sub-sampling a subsequent acquired image.

Abstract: A method and apparatus for providing image processing. For one embodiment of the invention, an image processing apparatus is arranged to process a first relatively underexposed and sharp image of a scene, and a second relatively well exposed and blurred image, nominally of the same scene, the first and second images being derived from respective image sources. The apparatus provides a portion of the relatively first underexposed image as an input signal to an adaptive filter; and a corresponding portion of the second relatively well exposed image as a desired signal to the adaptive filter. The adaptive filter produces an output signal from the input signal and the desired signal; and an image generator constructs a first filtered image from the output signal, relatively less blurred than the second image.

Abstract: A method of determining face recognition profiles for a group persons includes determining with a multi-classifier face detector that a face region within a digital image has above a threshold probability of corresponding to a first person of the group, and recording probability scores which are analyzed for each classifier, including determining a mean and variance for each classifier for the first person. The process is repeated for one or more other persons of the group. A sub-set of classifiers is determined which best differentiates between the first person and the one or more other persons. The sub-set of classifiers is stored in association with the first person as a recognition profile.

Abstract: A system and method for reducing noise in images is disclosed. The present invention reduces noise and preserves contrast of an image to be displayed, the image having pixels, by (1) comparing a value of a first pixel to values of a set of other pixels; (2) comparing values of pixels neighboring the first pixel to values of further pixels neighboring the set of other pixels; (3) determining, for each pixel in the set of other pixels, a weight based on results of steps (1) and (2); (4) calculating a de-noised pixel value based on the weights of each pixel in the set of other pixels; and (5) replacing the value of the first pixel with the de-noised pixel value.

Abstract: A technique of enhancing a scene containing one or more off-center peripheral regions within an initial distorted image captured with a large field of view includes determining and extracting an off-center region of interest (hereinafter “ROI”) within the image. Geometric correction is applied to reconstruct the off-center ROI into a rectangular or otherwise undistorted or less distorted frame of reference as a reconstructed ROI.

Abstract: Different distances of two or more objects in a scene being captured in a video conference are determined based on applying a face model to a detected face within a digitally-acquired image. At least one portion of the scene other than the face is identified as comprising a background object that is a different distance from the video camera component than the face. The technique involves blurring or otherwise rendering unclear the background object.

Abstract: A device is trained for face recognition. A first acquired digital image of a scene includes a face. Face image data is extracted and stored in a face image library along with an unique identifier. A second acquired digital image may or may not include the face of the same person as the face in the first acquired digital image. Face recognition is applied to extracted face data of the second digital image, and the face of the first digital image is displayed as a match to the face of the second digital image when the first and second images are determined to match or the second digital image or a face therein is identified with biometric data stored along with the first digital image when the faces in the first and second images are determined to match.

Abstract: A technique is provided for generating sharp, well-exposed, color images from low-light images. A series of under-exposed images is acquired. A mean image is computed and a sum image is generated each based on the series of under-exposed images. Chrominance variables of pixels of the mean image are mapped to chrominance variables of pixels of the sum image. Chrominance values of pixels within the series of under-exposed images are replaced with chrominance values of the sum image. A set of sharp, well-exposed, color images is generated based on the series of under-exposed images with replaced chrominance values.

Abstract: A digital image acquisition system having no photographic film, such as a digital camera, has a flash unit for providing illumination during image capture and a red-eye filter for detecting a region within a captured image indicative of a red-eye phenomenon, the detection being based upon a comparison of the captured image and a reference image of nominally the same scene taken without flash. In the embodiment the reference image is a preview image of lower pixel resolution than the captured image, the filter matching the pixel resolutions of the captured and reference images by up-sampling the preview image and/or sub-sampling the captured image. The filter also aligns at least portions of the captured image and reference image prior to comparison to allow for, e.g. movement in the subject.

Abstract: An image acquisition device having a wide field of view includes a lens and image sensor configured to capture an original wide field of view (WFoV) image with a field of view of more than 90°. The device has an object detection engine that includes one or more cascades of object classifiers, e.g., face classifiers. A WFoV correction engine may apply rectilinear and/or cylindrical projections to pixels of the WFoV image, and/or non-linear, rectilinear and/or cylindrical lens elements or lens portions serve to prevent and/or correct distortion within the original WFoV image. One or more objects located within the original and/or distortion-corrected WFoV image is/are detectable by the object detection engine upon application of the one or more cascades of object classifiers.

Abstract: A face recognition technique includes using a multi-classifier face detector to determine above a threshold probability that region of an image includes a face. Further probability values are determined for a set of classifiers for the region to provide a recognition profile. Face detection and recognition probabilities are determined for at least one classifier of the set. The recognition profile is compared against a predetermined recognition profile to determine a degree of match.

Abstract: In an embodiment, a device comprises a plurality of elements configured to apply a filter to multiple groups of pixels in a neighborhood of pixels surrounding a particular pixel to generate a matrix of filtered values; compute, from the matrix of filtered values, a first set of gradients along a first direction and a second set of gradients along a second and different direction; determine how many directional changes are experienced by the gradients in the first set of gradients and the gradients in the second set of gradients; compute a first weighted value for a first direction and a second weighted value for a second direction; and based, at least in part, upon the first and second weighted values, compute an overall texture characterization value for the particular pixel, wherein the overall texture characterization value indicates a type of image environment in which the particular pixel is located.

Abstract: An image registration method involves computing horizontal and vertical integral projection vectors for first and second distorted or partially distorted images or distortion-corrected images, or both. The images are registered by applying a translation, rotation and/or scale factor estimation between the first and second images on the horizontal and vertical integral projection vectors.