Showing papers by "Andrew C. Gallagher published in 2004"

Detecting irises and pupils in images of humans

Abstract: In an eye detection method, a plurality of candidate eyes are located within a digital image. Pixels of interest having a predetermined characteristic and a predetermined association to respective eyes are found. Pixels of interest associated with each eye are grouped. Parametric boundaries are fit on each of the groups to define a plurality of detected eye features. The boundaries have a predefined uniform shape and a size matched to a respective group. Each of the detected eye features is scored as to a geometric relationship between the respective boundary and pixels of interest associated with the respective eye to provide eye feature scores.

Image processing based on direction of gravity

Abstract: A method of processing a digital image to produce an improved digital image, includes receiving the digital image captured with a camera; determining a first vanishing point associated with the digital image; determining a second vanishing point associated with the digital image corresponding to a direction orthogonal the first vanishing point; determining a transform for modifying the digital image based on the first vanishing point and the second vanishing point; and applying the transform to the digital image to produce an improved digital image.

Detection of sky in digital color images

Abstract: A method of detecting sky in a digital color image having pixels includes identifying pixels from the digital color image representing an initial sky region, developing a model based on the identified sky pixels wherein such model is a mathematical function that has inputs of pixel position and outputs of color, and using the model to operate on the digital color image to classify additional pixels not included in the initial sky region as sky.

Improved blue sky detection using polynomial model fit

Abstract: Blue sky commonly appears in photographs. Reliable detection of blue sky can aid semantic image retrieval, image understanding and image enhancement. A novel algorithm is proposed based on a two-dimensional polynomial model of the image of blue sky. An initial sky detection is used to establish high-confidence blue sky regions. Candidate sky regions are found and a two-dimensional polynomial model is used to compute the belief that a given region is also sky. On a database of 83 images, the algorithm correctly classified 31 additional regions as blue sky while adding 8 minor false positives, most of them reflections of sky. Further, the algorithm correctly increased the belief values on 6 regions found by the initial sky detector.

Determining the age of a human subject in a digital image

Abstract: The age class of a human subject is ascertained in a digital image. The subject has a redeye defect pair; each defect having one or more defect pixels. In the method, two regions of pixels corresponding to the defects are identified. The distance between the regions is measured. A region size is determined, based upon the size of at least one of the regions. An age class is determined from the distance and region size.

Method and apparatus to extend the effective dynamic range of an image sensing device

Abstract: An image sensor generates an image signal with a differential response to image light. The image sensor has an array of photosites divided into standard photosites and non-standard photosites. A limiter provides the standard photosites with a predetermined standard response to a light exposure and the non-standard photosites with a predetermined slower response to the same light exposure. The standard photosites and nonstandard photosites both sparsely sample the array in a predetermined pattern.

Applying an adjusted image enhancement algorithm to a digital image

Abstract: A method is described for applying an image enhancement algorithm to input digital images represented in different input color spaces including identifying the input color space of an input digital image, and applying a color space transformation to the input digital image represented in the input color space to form a corresponding input digital image in a reference color space. The method also includes adjusting one or more algorithm parameters of the image enhancement algorithm in response to the identified input color space, and applying the image enhancement algorithm with the one or more adjusted algorithm parameters to the corresponding input digital image in the reference color space to produce an enhanced digital image in the reference color space.

Correction of redeye defects in images of humans

Abstract: In a method and system for correcting redeye in a digital image of a human, a defect pair separation is measured The members of the defect pair are each redeye defects in the image Each defect has one or more pixels The defects are adjusted, that is, retained unchanged or reduced in size to provide adjusted defects The adjustment can follow growth of initial seed defects The adjustment is responsive to the defect pair separation and can include trimming of pixels beyond a size limit calculated using the defect pair separation Following the adjustment, the color of the adjusted defects is changed to reduce redeye

Detection of hanging wires in digital color images

Abstract: A method of improving a digital color image having pixels, the method includes identifying pixels from the digital color image representing one or more sky regions; detecting sky occlusion regions by examining the sky regions; developing a model based on the identified sky pixels, wherein such model is a mathematical function that has inputs of pixel position and outputs of color; and using the model to operate on the digital color image to replace the values of pixels from the sky occlusion regions with values predicted by the model.

Adjusting the brightness of a digital image

Abstract: A method of processing a digital image, including the steps of : assigning belief values to pixels in the digital image, wherein the belief values are proportional to a belief that the pixels are associated with human skin; calculating a brightness adjustment amount based on a statistic of the pixels with belief values exceeding a threshold and the assigned belief values of the pixels; and applying the brightness adjustment amount to the digital image to produce a processed digital image having improved overall brightness.

Digital image sharpening method

Abstract: PROBLEM TO BE SOLVED: To provide an improved image sharpening method by which a quantity of sharpening is adjusted to avoid amplification of D min noise. SOLUTION: The method of sharpening a digital image having an image pixel value includes steps of: determining a D min value indicating a value corresponding to minimum possible exposure for a system that produced the digital image; applying a gain coefficient to impart a smaller gain factor to a pixel value having a value nearer the D min value according to a difference between the pixel value and the D min value; and sharpening the digital image by using the gain factors. COPYRIGHT: (C)2004,JPO