Abstract: In a digital image scanning system and method, a sample of tone values from the image is used to estimate the exposure or central tendency of the recorded image. The estimate is then used to select a tone correction function used to process the image prior to printing, transmission, or CRT display.

Abstract: Apparatus for enhancing and thresholding scanned microfilm images, specifically for removing single pixel noise therefrom, and accompanying methods for use therein are disclosed. Specifically, this apparatus convolves each one of a pre-defined group of incoming pixel values with a corresponding one of a pre-defined set of coefficients in order to generate an convolved pixel value. The convolved pixel is then compared against a threshold level to produce a thresholded pixel value indicative of the results of the comparison. The inventive apparatus also includes a noise processing circuit that contains a noise tracker and a noise detector and filter circuit. The noise tracker is used to produce a noise value that tracks a noise level present in the threshold pixel value.

Abstract: Apparatus and an accompanying method are disclosed herein for improving the compressibility of an enhanced digital video signal, illustratively comprised of a serial bit stream, such as that generated by scanning a video image produced, for example, by a microfilm scanner. Specifically, whenever a video transition occurs in the a stream of incoming pixels, this apparatus momentarily varies a threshold level, in an appropriate direction, during the transition in order to reduce the noise content in threshold pixel values and thereby enhance the compressibility of these pixels.

Abstract: Apparatus, and accompanying methods for use therein, for locating edges contained within each scan line of a scanned video image and also for locating horizontal boundaries (leading and trailing edges) of a portion of a document contained within the scan line are described herein. Specifically, this apparatus first provides a background value representative of a background level in a scanned image and associated with an incoming pixel value. If the background value exceeds a first reference value, a first signal is generated. If the incoming pixel value exceeds a second reference value, a second signal is generated. Now, an edge transition is detected based upon a pre-defined logical combination of the first and second signals.

Abstract: A method for extracting (masking off) a desired video window, such as that which forms a scanned image of a document or a portion thereof, from a overscanned image stored in a frame buffer is disclosed herein. Specifically, this method involves first determining horizontal pixel locations for left and right edges of the document for each of a plurality of scan lines located within said scanned image. Then, histograms of the left and right edge locations are fabricated. Once this has occurred, the positions of top and bottom edges of an escribing rectangle is found. The escribing rectangle is a rectangular area, located within the frame buffer, which includes the entire document. Now, with the top and bottom edge locations known, the inventive method locates the right and left edge locations of the escribing rectangle by suitably analyzing the histogram results.