Abstract: Systems and methods for backlit scene detection are described. The method includes performing a pre-test on an electronic image to determine that the electronic image does not include a disqualifying characteristic. Next, the method includes performing a type test on the electronic image to determine an image type of the electronic image from a first set of predetermined backlit image types. Next the method includes performing a post test on the electronic image to determine that the image darkness density does not exceed a predetermined density. Finally, the method includes performing a second post test on the electronic image to determine that the electronic image is one of a second set of predetermined backlit image types or that the edge percentage of the electronic image exceeds a predetermined percentage.

Abstract: Image tone adjustment systems and methods are disclosed. The system of the present invention includes an input to accept an input image. A curve generator is also disclosed that is used to generate a tone reproduction curve including a contrast stretch of the input image, a white stretch of the input image and a black stretch of the input image. Backlit correction is also applied by the tone adjustment corrector to thereby generate a tone adjusted image.

Abstract: The subject application is directed to a system and method for detecting backlit images. Encoded color image data is first received into a computer having a processor and associated data storage. Histogram data is then calculated from the received encoded color image data, and a mid-tone range in normalized histogram data is then detected. A zone of normalized pixel counts within the mid-tone range is then selected. Data representing an entry point and an exit point of the normalized histogram data relative to the selected zone is then generated. A plateau area is detected in the histogram data in the selected zone between the entry point and the exit point. Thereafter, a backlit image detection signal is generated indicating whether a backlit image portion is present in the color image data according to the plateau detection.

Abstract: Systems and methods for image contrast enhancement are disclosed. A tone reproduction curve function is generated for a plurality of pixel values, each including a brightness value, based on a contrast adjustment value, the brightness value of each of the pixels and a power-law function, the tone reproduction curve taking the form of an S-shaped curve. The S-shaped curve includes a bulging curve portion corresponding to an area wherein a brightness of pixels having values representative of relative brightness is increased, a sagging curve portion corresponding to an area wherein a brightness of pixels having values representative of relative darkness is decreased, and an inflection point serving as a demarcation between the bulging area and the sagging area of the tone reproduction curve. The exponent of the power-law function is generated as a quadratic polynomial function f(x) in a range 0?f(x)?1, wherein x is defined as a curvature factor.

Abstract: Systems and methods for backlit scene detection are described. The method includes performing a pre-test on an electronic image to determine that the electronic image does not include a disqualifying characteristic. Next, the method includes performing a type test on the electronic image to determine an image type of the electronic image from a first set of predetermined backlit image types. Next the method includes performing a post test on the electronic image to determine that the image darkness density does not exceed a predetermined density. Finally, the method includes performing a second post test on the electronic image to determine that the electronic image is one of a second set of predetermined backlit image types or that the edge percentage of the electronic image exceeds a predetermined percentage.

Abstract: The subject application is directed to a system and method for image fog scene detection. Electronic image data is first received and divided into image regions, with each region consisting of a plurality of pixels. Next, a comparison matrix is generated corresponding to each image region based upon a comparison of minimum intensity values associated with corresponding pixels to a threshold value. An entry of at least one of the comparison matrices is then tested for a preselected value. The received electronic image data is then identified as inclusive of a fog scene based upon the output of the test on the entry of the comparison matrix. Based upon the identification, fog scene detection data representing an identified fog scene is then generated.

Abstract: The subject application is directed to a system and method for digital image classification. A first data set that includes a data range that corresponds to a first flesh tone range derived from measured flesh tones of subjects in a first ethnicity class is stored memory. A second data set having a data range that corresponds to a second flesh tone range derived from measured flesh tones of subjects in a second ethnicity unique to the first ethnicity is stored in memory. The data sets being encoded in a three dimensional color space. Image data is then received that includes a human subject. A facial region is isolated and image data in the facial region is compared with each data set via a processor operable with the memory. An image correction signal is generated representing whether the facial region aligns with the first data set or the second data set.

Abstract: The subject application is directed to a system and method for image contrast enhancement. Input image comprised of digitally encoded image data is received into a data processor, including an associated data storage, and histogram data is generated corresponding to received image data. Acceptability of received image data is determined in accordance with a comparison of the histogram data to at least one preselected threshold value and a contrast adjustment including a white stretch, a black stretch, and a contrast stretch is performed on the input image in accordance with the histogram data.

Abstract: Image tone adjustment systems and methods are disclosed. The system of the present invention includes an input to accept an input image. A curve generator is also disclosed that is used to generate a tone reproduction curve including a contrast stretch of the input image, a white stretch of the input image and a black stretch of the input image. Backlit correction is also applied by the tone adjustment corrector to thereby generate a tone adjusted image.

Abstract: The subject application is directed to a system and method for detecting backlit images. Encoded color image data is first received into a computer having a processor and associated data storage. Histogram data is then calculated from the received encoded color image data, and a mid-tone range in normalized histogram data is then detected. A zone of normalized pixel counts within the mid-tone range is then selected. Data representing an entry point and an exit point of the normalized histogram data relative to the selected zone is then generated. A plateau area is detected in the histogram data in the selected zone between the entry point and the exit point. Thereafter, a backlit image detection signal is generated indicating whether a backlit image portion is present in the color image data according to the plateau detection.

Abstract: The subject application is directed to a system and method for backlit face image correction. Image data is first received that includes at least one facial region that defines an image human face. At least one facial region is detected via a processor that operates in accordance with software and an associated data storage. Skin tone characteristics of the at least one facial region are then detected. Pixel count data and histogram data are then calculated from the received image data. A plateau is then detected in a function of the pixel count data relative to the histogram data. A correction factor, based upon a property of the detected plateau, is then calculated. Pixel values of the at least one facial region are adjusted in accordance with the calculated correction factor. Thereafter, a corrected image is output that includes adjusted pixel values corresponding to at least one facial region.

Abstract: A method and apparatus for enhancing electronic images allows for improved characteristics between light areas and dark areas, and is particularly effective for backlit images. A transition between light and dark image portions is detected. A determination is made from an analysis of spectral distributions as to whether an image portion is backlit. Upon detection, image data is adjusted to lighten or darken image portions to allow for improved image viewing. Use of cumulative probability distribution data associated with an electronic image facilitates isolation of backlit image portions and object image portions.

Abstract: In accordance with the present invention, a system and method are described for producing pure gray tones on a multi-color document output device that includes a system for operating on device independent color data having multiple color components, which color data corresponds to one or more associated electronic documents. The system and method function to extract gray data by the use of tristimulus values in a range specified by a functional relationship of a selected tristimulus value.

Abstract: The subject application is directed to backlit image adjustment. First, image data is received that includes an image portion defined by an associated backlit region, from which a tonal curve is then isolated. At least one anchor point on the isolated tonal curve is selected based upon backlighting characteristics in the received image data. A sectional bulging operation is applied on the isolated tonal curve in accordance with the selected anchor point.

Abstract: The subject application is directed to a system and method for global darkness image enhancement. Image data encoded in a multidimensional color space is first acquired. Thereafter, histogram data is calculated according to the acquired image data. A ramp zone associated with the calculated histogram data is then detected. A black level associated with the acquired image data is then selectively stretched in accordance with the detected ramp zone so as to generate enhanced image data. The enhanced image data is then output to an associated data storage or an associated display.

Abstract: The subject application is directed to a system and method for brightness adjustment of electronic images. Image data including a plurality of pixels is first received, with each pixel having an associated image value. Histogram data corresponding to a histogram of at least one component value corresponding to each of the pixels is then calculated. An image correction value is then calculated according to at least one characteristic of the calculated histogram data. A determination is then made regarding the application of the calculated image correction value from the calculated histogram data. Corrected image data is thereafter generated based upon the application of the calculated image correction value to each pixel.

Abstract: The subject application is directed to a system and method for artistic scene image detection. First, image data is received that is encoded in a multi-dimensional color space. From the received image data, histogram data is then calculated. Dominant spike regions in the calculated histogram data are then identified. An N-sum value is then calculated from the identified dominant spike regions in the calculated histogram data. Testing of the calculated N-sum value against a predetermined threshold value then occurs. The received image data is thereafter classified as an artistic scene, a tinted artistic scene, or a sepia tone range artistic scene according to the testing results.

Abstract: The subject application is directed to a system and method for isolating near achromatic pixels of a digital image. First, image data encoded as a plurality of pixels is received, with each pixel having component values in a multi-dimensional color space. From the received image data, chroma data is extracted. Histogram data is then calculated from the extracted chroma data. An adapted threshold value is then determined based upon the calculated histogram data. Component values of the plurality of pixels are then compared to the determined adapted threshold value. Isolation of pixels as near achromatic then occurs in accordance with the output of the comparison of the adapted threshold value to the component values.

Abstract: A system and method for converting an image to a constrained color palette representation receives a specified color or hue from which an image is to be defined. A conversion is then made from an input color space to a CIELAB representation. Data is converted in CIELAB space, in accordance with a specified hue, to a color space set by the specified color. Once converted, data is converted to a color space corresponding to that of an associated document rendering device.

Abstract: The subject application is directed to a system and method for image fog scene detection. Electronic image data is first received and divided into image regions, with each region consisting of a plurality of pixels. Next, a comparison matrix is generated corresponding to each image region based upon a comparison of minimum intensity values associated with corresponding pixels to a threshold value. An entry of at least one of the comparison matrices is then tested for a preselected value. The received electronic image data is then identified as inclusive of a fog scene based upon the output of the test on the entry of the comparison matrix. Based upon the identification, fog scene detection data representing an identified fog scene is then generated.