Abstract: A method for determining the orientation of a digital image, includes the steps of: employing a semantic object detection method to detect the presence and orientation of a semantic object; employing a scene layout detection method to detect the orientation of a scene layout; and employing an arbitration method to produce an estimate of the image orientation from the orientation of the detected semantic object and the detected orientation of the scene layout.

Abstract: A system and method for automatically altering an image having pixels is disclosed. An image is segmented into regions and various features of the regions are extracted and used to calculate a confidence factor, or belief value, that a particular region is a main subject of the image. The belief values are assembled into a belief map that is used to define how pixel values are to be altered. Various color and appearance values of the pixel are altered in a predetermined way to emphasize the main subject of the image. The emphasis is applied in a binary fashion or as in a gradient of values. Either the main subject can be emphasized, or the secondary or background subjects can be de-emphasized. The system can be implement on a variety of computer systems.

Abstract: A method for producing a new image of a desired aspect ratio from an original digital image comprises the steps of: (a) determining a cropping window of the desired aspect ratio and a predetermined size; (b) obtaining a main subject belief map including an array of belief values indicating the location and relative importance of subject matter in the original digital image; (c) using the main subject belief map, determining a location of the cropping window in the original digital image that satisfies a predetermined set of criteria; and (d) producing a cropped image of the desired aspect ratio.

Abstract: A method for processing a digital color image for display and transmission includes the steps of (a) resizing the image to a lower display resolution to produce a display-sized image, (b) quantizing the display-sized image to a display palette with an error diffusion process to produce an error diffused image, (c) losslessly compressing the error diffused image to produce a compressed image having a particular file size, (d) comparing the file size to a maximum file size and if less than the maximum file size, transmitting the digital image to the display; (e) if the file size is greater than the maximum file size but less than a another predetermined file size, reducing the error diffusion rate by a predetermined amount in order to reduce the file size; and (f) if the file size is greater than the maximum file size and the predetermined file size, simplifying the resized image by reducing image detail to produce a simplified image and repeating the preceding steps on the simplified image.

Abstract: A method for processing a digital color image includes the steps of: providing a subject matter detector for distinguishing between target and background subject matters; applying the subject matter detector to the image to produce a belief map indicating the degree of belief that pixels in the image belong to target subject matter; providing an image enhancement operation that is responsive to a control signal for controlling the degree of image enhancement; and applying image enhancement to the digital image by varying the control signal according to the belief map to produce an enhanced image.

Abstract: Disclosed is a method for detecting subject matter regions in a digital color image having pixels of (red, green, blue) values, comprising the steps of: assigning to each pixel a belief value as belonging to the subject matter region based on color and texture features; forming spatially contiguous candidate subject matter regions by thresholding the belief values; analyzing the spatially contiguous regions based on one or more unique characteristics of the subject matter to determine the probability that a region belongs to the subject matter; and generating a map of detected subject matter regions and associated probability that the regions belongs to the subject matter.

Abstract: A method for segmenting an anatomical structure having portions occluded in a digital image wherein the digital image is produced by a projection-based medical imaging modality. The method includes the steps of: outlining an unoccluded portion of the anatomical structure in a training image; augmenting the occluded portion of the anatomical structure according to a pre-determined atlas of the anatomical structure; generating an augmented training example using the outlined and augmented portions; repeating the steps of outlining, augmenting, and generating for a plurality of training images to train a statistical model of the structure; accessing the digital image; and segmenting the anatomical structure in the accessed digital image according to the statistical model.

Abstract: A method for determining a candidate lesion region within an ultrasound image. The method includes the steps of: accessing a digital ultrasound image of anatomical tissue; segmenting spatially contiguous pixels in the ultrasound image into a plurality of regions in accordance with substantially similar intensity values and spatial smoothness constraints; and selecting, from the plurality of regions, one or more candidate lesion regions having an intensity value lower than a pre-determined intensity value. In one arrangement, the one or more candidate lesion region is classified into at least one of the following classes: benign, malignant, or unknown.

Abstract: A method for classifying radiographs. The method includes the steps of: accessing a radiograph; categorizing the radiograph into pre-determined classes based on gross characteristics of the radiograph, and recognizing the image contents in the radiograph.

Abstract: A method for recognizing the projection view of radiographs comprising the steps of correcting the orientation of the input radiograph, locating a region of interesting in the radiograph, recognizing the projection view of the radiograph.

Abstract: A method for detecting and correcting the orientation of a radiograph. The method generally includes the steps of accessing the radiograph; locating a region of interest in the radiograph; and using the located region of interest and a classifier to detect the orientation of the radiograph. In a preferred embodiment, the radiograph is preprocessed.

Abstract: A system and method for automatically altering an image having pixels is disclosed. An image is segmented into regions and various features of the regions are extracted and used to calculate a confidence factor, or belief value, that a particular region is a main subject of the image. The belief values are assembled into a belief map that is used to define how pixel values are to be altered. Various color and appearance values of the pixel are altered in a predetermined way to emphasize the main subject of the image. The emphasis is applied in a binary fashion or as in a gradient of values. Either the main subject can be emphasized, or the secondary or background subjects can be de-emphasized. The system can be implement on a variety of computer systems.

Abstract: A method for detecting objects in a digital image includes the steps of generating a first segmentation map of the digital image according to a non-object specific criterion, generating a second segmentation map of the digital image according to an object specific criterion, and detecting objects in the digital image using both the first and second segmentation maps. In a preferred embodiment of the invention, the non-object specific criterion is a color homogeneity criterion and the object specific criterion is an object specific color similarity, wherein the object specific color is skin color and the method further comprises the step of detecting red-eye in the detected skin color regions.

Abstract: A novel multi-resolution block sampling based texture analysis/synthesis algorithm. A reference texture is assumed to be sample from a probability function. The synthesis of a similar, but distinctive, synthetic texture is handled in a process and by an apparatus that first estimates and then resamples the probability function. In order to achieve good and fast estimation of the probability function for a reference texture and in order to retain the texel structural information during the synthesis, a novel concept of block sampling and a corresponding novel texture synthesis scheme based on multi-resolution block sampling is employed. As a result of this novel approach, the computational complexity of the present invention is much lower than that of other approaches to the problem. In addition, for textures that exhibit a high degree of directionality, a process, which integrates estimation of dominant texture direction and the synthesis algorithm is employed to handle directional textures.

Abstract: A technique for producing a foreground mask from a digital image is based on receiving two digital images, wherein a first digital image is derived from a first original scene image including a foreground photographed against a background having arbitrary and unknown content matter, and a second digital image is derived from a second original scene image including substantially the same background but without the foreground. The second digital image is automatically aligned to the first digital image without user intervention to generate an aligned digital image including the background without the foreground. Then, using the first digital image and the aligned digital image, a foreground mask is produced relating to the spatial regions in the first digital image pertaining to the foreground, whereby the foreground mask is subsequently used to generate a composite digital image.

Abstract: Disclosed is a method for converting an input digital color image having a set of possible input colors to an output digital color image having a set of palette colors, the number of palette colors being less than the number of possible input colors, wherein the set of palette colors is determined based on the distribution of colors in the input digital image boosted by a distribution of important colors contained in the input digital image.

Abstract: A method of storing an image in a digital camera, comprising the steps of: capturing the image using the selected quantization table. A method of storing an image in a digital camera, wherein the step of selecting a quantization table comprises the steps of: selecting a quality setting; compressing the image using a quantization table corresponding to the selected quality setting; decompressing the image; evaluating the decompressed image with the image quality metric; and, adjusting the quantization table such that the quality metric matches the selected quality setting.

Abstract: Disclosed is a method for converting an input digital color image having a set of possible input colors to an output digital color image having a set of palette colors, the number of palette colors being less than the number of possible input colors, wherein the set of palette colors is determined based on the distribution of colors in the input digital image supplemented by a distribution of important colors.

Abstract: A method is disclosed for providing photofinishing services for images having a main subject and background, comprising the steps of producing a digital image of a customer's photographic image; producing a main subject belief map from the digital image; employing the main subject belief map to produce a modified digital image; producing a print of the digital image; producing a print of the modified digital image; and delivering both prints to the customer. The step of employing the main subject belief map to produce a modified digital image includes zoom and crop steps that maintain a main subject within a resulting cropped image. The step of employing the main subject belief map to produce a modified digital image may include a subject emphasizing step.

Abstract: A method of detecting sky in a digital color image having pixels is disclosed. The method 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.