Accusoft.ScanFixXpress8.Net
Glossary
A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

A

Aliasing
Aliasing is the appearance of jagged lines caused by digital sampling of curves or diagonal lines. Aliasing is most prevalent in low resolution, low color depth images.

Anti-aliasing
Anti-aliasing is a process by which the effects of aliasing are reduced by blending the colors of pixels along sharp borders.

Artifact
An artifact is a shape that appears in an digitized image that was not in the original, but was created during sampling or image processing steps. The jagged edges produced by aliasing are a type of artifact caused by sampling, while the specks near edges in JPEG images are an artifact caused by the DCT compression algorithm.

Aspect Ratio
There are two kinds of aspect ratio that you need to be aware of when working with 2D images; the image aspect ratio and the pixel aspect ratio.

Average Filter
See Box Filter.

B

Bitonal

A bitonal image is an image that consists of only two colors, black and white. Bitonal images are commonly dealt with in document imaging processing, because:

BMP
The Microsoft Windows Bitmapped file format is based on the Microsoft Windows internal bitmap data structures. The BMP format is very widely used and it is quick and easy to read. Although most BMP files are uncompressed, a type of run length encoding is supported.

Box Filter
Box filters are used for smoothing, and provide fast, aggressive smoothing to the image. A box filter, or average filter, is a convolution filter with a matrix of all 1's, which has the result of outputting the average pixel value for the window. However, large box filters also tend to produce visibly boxy results due to the use of a rectangular window.

BPP
Bits per pixel. The color depths supported by ScanFix® Xpress are: 1 bpp for bitonal images, 8 bpp for gray, and 24 bpp for color.

Brightness
The intensity of a pixel, usually expressed in the range of 0 (black) to 255 (white or fully bright color). Brightness of a color image may be defined in a number of ways, depending on the colorspace used.

C

CCITT
CCITT is the most widely used compression scheme for transmitting binary (black and white) image data. CCITT was established by the International Telegraph and Telephone Consultative Committee. The Group 3 and Group 4 CCITT compression standards are widely used for transmitting fax data. CCITT compressed images are typically stored in the TIFF image format.

Color Depth
The color depth of an image, expressed in bits per pixel (BPP), is the amount of space needed to store the information to describe the color of a single pixel. ScanFix® Xpress supports bit depths of 1, 8 and 24 BPP.

Color Space
A colorspace is a means of representing a set of colors, typically by mapping a subset of the realm of visible colors onto a three dimensional space so that each possible color can be represented by a set of three values. There are many different colorspaces which serve different purposes. ScanFix® Xpress uses the RGB (Red, Green, Blue) and HSV (Hue, Saturation, Value) colorspaces.

Comb
A comb is a set of boxes, with closed or open tops, that are used on forms, where each box is typically intended to be filled with a single character. The "teeth", or short vertical lines in the combs, are removed during the comb removal function, leaving only the contents of the boxes.

Container
The container for the component. An object implementing the IContainer interface that represents the component's container, or a null reference (Nothing in Visual Basic) if the component does not have a site.

Continuous Tone
A continuous tone image is one where all the colors in the gamut are evenly distributed, allowing blending of pixel values. Bitonal and paletted images are not considered continuous tone, but 8-bit grayscale and 24-bit color images are.

Contrast
Contrast is the difference between the darkest and the lightest valued pixels in a given region of an image. Contrast does not apply to bitonal images, which have only pure black and pure white pixels. Contrast is very important in color and gray images, because low contrast data can be difficult to read and process.

Convolution Filter
A convolution filter alters each pixel's color based on its current color and the colors of neighboring pixels within the filter window, using a matrix of values to determine the weight given to each pixel in the area. When a pixel is processed, its color value is multiplied by the coefficient in the center of the matrix, and any pixels within the window are multiplied by the corresponding coefficients in the matrix. The sum of the products, divided by the Divisor, provides the output value for the center pixel. The convolution filter is typically used for sharpening, softening, and edge enhancement.

D

DCT
A DCT, or discrete cosine transform, is a method of turning sampled data into a collection of cosine waves of varying frequencies. The DCT is most commonly used in the JPEG compression format, where it is used to convert blocks of pixels into waves, and then remove low amplitude elements that do not contribute significantly to the image. The remaining frequencies are stored, and by reversing the DCT, convert the waves back into pixel values can be used to reconstruct a visually similar image. The DCT is typically used as a lossy compression format, which yields very high compression ratios with little loss in quality of the restored image.

DCX
An image format containing multiple PCX images. A DCX file can contain a maximum of 1024 images.

DDB
Microsoft Device Dependent Bitmap. The actual in-memory information used to display an image.

DEP
Data Execution Prevention (computer security).

Depth, Color
See Color Depth.

Despeckle
A process intended to remove specks from an image. See Speck.

DIB
Microsoft Device Independent Bitmap. An alternative file extension used by the Microsoft Windows Bitmap.

DICOM
Digital Imaging and Communication, a standard for handling digital medical images.

Difference of Gaussian
See Unsharp Mask Filter.

Dilate
Dilate is an operation that "grows" objects in an image. In ScanFix® Xpress, the bitonal image processing definition is used, which means that the dilate operation makes black objects get larger, and white objects get smaller.

Dispose
Releases the resources used by the Component.

Dithering
A technique used in computer graphics to create the illusion of varying shades of gray on a monochrome display or printer, or additional colors on a color display or printer. See also Dot Shading and Halftoning.

DoG
See Unsharp Mask Filter.

Dot Shading
Dot shading refers to the use of halftoning or dithering to create an area of lighter color in a printed document. Dot shaded regions are commonly found in printed forms, and provide special difficulties in removal. ScanFix® Xpress offers a function designed specifically to remove the dots from dot shaded regions in bitonal images. For continuous tone images, a Gaussian filter will often convert dot shading into a solid color.

E

Eight-way Connected
Eight-way connected refers to a method of grouping bitonal pixels into an object. An eight-way connected object is one in which pixels that meet on the diagonals are considered connected. See also, Four-way Connected.

Erode
Erode is an operation that "shrinks" objects in an image. In ScanFix® Xpress, the bitonal image processing definition is used, which means that the erode operation makes black objects get smaller, and white objects get larger.

EXIF
Exchangeable Image File (Japan Electronic Industry Development Association, JEIDA), compressed image format used in digital cameras.

F

Float
Denotes a simple type that stores 32-bit floating-point values.

Four-way Connected
Four-way Connected refers to a method of grouping bitonal pixels into an object. A Four-way Connected object is one in which pixels that meet on the top, bottom, or sides are considered connected, but diagonals are not. See also, Eight-way Connected.

Frequency
Frequency in imaging terms refers to the rate of change in value of a line of pixels. High frequency data consists of rapidly changing brightness or colors, while low frequency data consists of gradually changing brightness or color.

G

Gamut
All possible colors representable by a given colorspace and/or bit depth.

Gaussian Filter
A Gaussian filter is a type of convolution filter that uses matrix values generated using a Gaussian curve. Gaussian filters are used to smooth an image, and provide high quality results. However, Gaussian filters tend to be slower than other filters, such as the box filter, because a larger filter window needs to be used to provide the same degree of smoothing as, for example, the box filter.

GDI
Acronym for Graphical Device Interface. In Windows, a graphics display system used by applications to display or print bitmapped text (TrueType fonts), images, and other graphical elements. The GDI is responsible for drawing dialog boxes, buttons, and other elements in a consistent style on screen by calling the appropriate screen drivers and passing them the information on the item to be drawn. The GDI also works with GDI printers, which have limited ability to prepare a page for printing. Instead, the GDI handles that task by calling the appropriate printer drivers and moving the image or document directly to the printer, rather than reformatting the image or document in PostScript or another printer language. 

GIF
Graphical Interchange File, originally developed by CompuServe as a machine independent format. GIF is one of the most popular image formats for storing 8-bit digitized images. GIF files are typically smaller than uncompressed formats like BMP due to the LZW compression algorithm that GIF uses.

Grayscale or Greyscale
A format of an image that contains only brightness information, but differs from bitonal in that it contains more than 2 colors. Grayscale images supported by ScanFix® Xpress use 8 bits of brightness information, for a range of 0 (black) to 255 (white).

H

Halftone
Pertaining to pictures in which gradation of tone in the photograph is reproduced by a graduated system of dotted and checkered spots, usually nearly invisible to the unaided eye, produced by the interposition between the camera and the object of a screen. The name alludes to the fact that this process was the first that was practically successful in reproducing the half tones of the photograph. Halftoning is often used for dot shaded regions in document images, and can be difficult to remove with despeckle algorithms because the halftone dots are often connected. ScanFix® Xpress provides a special dot shading removal algorithm to handle these cases. See also Dithering, Dot Shading.

High Pass Filter
A high pass filter is a filter that preserves high frequency information, while removing low frequency information. In imaging, high pass filters are used for edge detection and sharpening. See also Low Pass Filter, Frequency.

HSV
HSV is a colorspace consisting of three components called Hue, Saturation, and Value (or Volume). This colorspace is often used by artists, because it provides an easier and more intuitive means of describing color than most other colorspaces. The HSV colorspace is also used by a number of features in ScanFix® Xpress, where the separation of hue, saturation, and value are useful. See also Hue, Saturation, and Value.

Hue
One of the 3 components of the HSV color space. The hue represents where on the color wheel a color falls. While Hue is often represented with values from 0 to 359, indicating degrees of a circle, ScanFix® Xpress uses values between 0 and 255, so that the representation will fit in one byte of data.

I

Inverse
To reverse something or change it to its opposite. For example, to invert the colors on a monochrome display means to change light to dark and dark to light. In a digital electrical signal, to replace a high level by a low level and vice versa. This type of operation is the electronic equivalent of a Boolean NOT operation.

IOD
Information Object Definition.

J

Jaggies
See Aliasing.

JIF
An alternative image file extension used for a JFIF compliant JPEG file. Because many early versions of JPEG contained a JPG extension but did not adhere to the JFIF specifications, the JIF extension was created to identify JPEG files that are JFIF compliant. As the popularity of JPEG grew, JFIF compliance became synonymous with the JPG extension and thus the JIF extension is rarely used.

JPEG
The JPEG (Joint Photographic Experts Group) standard is one of the most important image data compression standards. It is widely used in many industries including graphic arts, desktop publishing, medical imaging, and many more. It is also becoming one of the most widely used image formats on the Internet. Although JPEG supports a lossless mode, it most commonly uses lossy compression, wherein a DCT is used to eliminate less important frequencies from the image. The lossy compression allows JPEG to achieve very high compression ratios with little visual impact, though at the pixel level it can result in a significant number of artifacts in the image.

L

Laplacian Filter
The Laplacian filter is a convolution filter that works as a high pass edge detection filter, detecting edges in all directions. See also High Pass Filter, Sobel Filter, Prewitt Filter.

Lossless Compression
Lossless compression is a compression method that preserves all the data in the original, so that the reconstructed image will be exactly the same as the original image. Lossless compression is very poorly suited to dealing the noise typically found in sampled continuous tone images, so it is usually reserved for simpler data such as bitonal images. Most bitonal TIFF compression formats are lossless formats.

Lossy Compression
Lossy compression is a compression method that discards relatively unimportant data during compression, so that the reconstructed image will be very similar, but not exactly the same as the original image. Lossy compression algorithms are very common when dealing with audio, video, and image data, as they provide extremely high compression with minimal loss in perceptible quality. Formats that can or do use lossy compression include the JPEG image format, MPEG video, and MP3 audio.

Low Pass Filter
A low pass filter is a filter that preserves low frequency information, while removing high frequency information. In imaging, low pass filters are used for smoothing. See also, High Pass Filter, Frequency.

LZW
A compression algorithm named after Abraham Lempel and Jakob Ziv (creators of Lempel Ziv compression) and LZW designer Terry Welch that makes use of repeating strings of data in its compression of character streams into code streams. It is also the basis of GIF compression.

M

Median Filter
A median filter is a filter that uses the statistical median value of a window of pixels to generate the output pixel. Median filters operate in a manner similar to high pass filters, but they have the advantage of preserving sharp edges, while eliminating high contrast specks far more effectively than a comparably sized box or Gaussian filter.

Morphology
Morphology means shape or form. In imaging, morphology refers to the shape of groups of pixels, and algorithms that can alter these shapes. An example of a morphological operation is dilate, which makes black objects get larger, or deskew, which rotates the image contents.

N

Negate
Negate reverses the brightness of each pixel in an image, creating a photo negative of the original image.

Noise
Noise is any data in an image that is considered irrelevant and obscures the desirable data in the image. Noise can take many forms, but the most common forms encountered in imaging are the result of either sampling issues or artifacts of processing performed on the image. Sampling issues are caused by dust or other foreign matter on the scanned document or scanner bed, or defects inherent in the document, such as highly textured paper or shaded regions that interfere with text recognition. Artifacts may be cause when thresholding, either on the scanner or by later processing software, produces dots out of shaded regions or sampled noise, or when lossy compression methods such as that used by JPEG results in specks surrounding sharp edges in the image.

O

Object
An object is any group of connected pixels in a bitonal image; objects in ScanFix® Xpress are generated using Four-way Connection. Objects are important in many ScanFix Xpress functions, such as despeck, blob removal, line removal, and color and blank detection.

Overscan
Overscan refers to the region outside the page but within the scanned region of a document image. Overscan may be black or white.

P

Palette
A palette is an indexed list of colors used to generate an image. Palettes are typically found in non-continuous tone color images, though some image formats, such as the BMP format, do not discriminate between 8 bit continuous tone gray images and 8 bit palette color images. ScanFix® Xpress only handles a limited set of paletted images; those which contain only black and white, and those which contain only 256 shades of gray.

PBM
Portable bitmap image format, a very simple uncompressed 1-bit (black and white) image format. Data can be stored in either binary or ASCII format.

PCX
The PCX image format was originally developed by ZSoft for PC Paintbrush. A version of PC Paintbrush called Microsoft Paintbrush is bundled with Microsoft Windows which established the importance of PCX support. The PCX format uses an RLE encoding scheme which attempts to reduce the size of the image, however, detailed images may result in negative compression which means the file size may be actually be larger after compression.

PGM
Portable Gray Map image format. PGM is a very simple uncompressed 8-bit grayscale image format. Data can be stored in either binary or ASCII format.

Pixel
A pixel, short for "picture element", is a single sampled value from a digital image. Document images are typically sampled at a resolution of between 100 and 300 pixels per inch (PPI).

PNG
Portable Network Graphics (pronounced "ping") provides a portable, legally unencumbered (royalty-free), well-compressed, well-specified standard for lossless bitmapped image files. The initial motivation for developing PNG was to provide a royalty-free alternative to GIF. The design retains many of the features of GIF that make it an attractive format for transmission of images, such as lossless compression and progressive display capabilities. Additionally, it provides several advantages including 24-bit image support and generally better compression, although decompression is typically somewhat slower than GIF. PNG's success will be ultimately determined by its acceptability on the Internet. At the time of this writing, however, PNG has not been incorporated as part of the major Internet browsers.

PPM
Portable Pixmap image format, a very simple uncompressed 24-bit image format. Data is stored in RGB triplets and can be stored in either binary or ASCII format.

Prewitt Filter
The Prewitt filter is a convolution filter used for detecting horizontal or vertical edges. Pixels corresponding to an appropriate edge will show as bright values in the output image, while non-edge pixels will show as dark. The Prewitt filter provides a strong response to horizontal or vertical lines, but less response to diagonals. See also, Sobel Filter, Laplacian Filter.

R

Resolution
The resolution of a digital image is the rate at which the original was sample, and in document images is expressed in terms of the number of pixels per unit of measure, typically pixels per inch (PPI). Document images are typically scanned at resolutions between 100 and 300 PPI.

RGB
Red, Green Blue. Bitmapped images are most generally stored as RGB data. Each pixel or data point in a true color (24-bit) image contains 3 components, red, green and blue, that describe the color of the pixel. Each of the components can have a value between 0 and 255, 0 being the darkest shade of color and 255 being the lightest shade of color. Since each component can be stored in one byte (8-bits), the 3 components take up 3 bytes or 24-bits per pixel.

RLE
RLE stands for Run Length Encoding. It is a lossless compression algorithm that only offers decent compression ratios in specific types of data, typically bitonal or continuous tone images with large regions of flat color.

S

Sampling
Sampling is the process of measuring values at certain intervals to generate a digital reproduction of an item. Images are sampled by breaking the image down into a rectangular grid of pixels, then measuring the color or brightness of each pixel.

Saturation
One of the 3 components of the HSV colorspace. Saturation controls the purity of a color. Higher saturations result in richer, deeper colors. Lower saturations add gray, resulting in paler colors. Saturation is expressed as a value from 0 to 255.

Sharpen
Sharpening is the use of a high pass filter to emphasize the high frequency elements of an image. Sharpening and smoothing are opposite operations, and most sharpening filters are created by subtracting a smoothing filter from a no-change filter. See also, Unsharp Mask Filter, High Pass Filter.

Size of an Image

The term "size" can refer to many things when applied to a digital image. These include:

Smooth
Smoothing is the process of de-emphasizing sharp, high contrast edges in the image, typically by using a low pass filter. Smoothing algorithms are often used to remove noise in continuous tone images. See also Sharpen.

Sobel Filter
The Sobel filter is a convolution filter used for detecting horizontal or vertical edges. Pixels corresponding to an appropriate edge will show as bright values in the output image, while non-edge pixels will show as dark. The Sobel filter provides a stronger response to diagonals than the similar Prewitt filter. See also, Prewitt Filter, Laplacian Filter.

Speck
A speck is small group of pixels that appear as high contrast noise. In bitonal images, specks usually appear is one or more black pixels, while in continuous tone images they may appear as any color that contrasts with the background. For bitonal images, specks may be removed by the Despeckle or dot shading removal algorithms, and for continuous tone images, a smoothing or median filter can be used.

T

TGA
Targa Image File format was developed by TrueVision for its Targa line of products. TGA is one of the earliest image formats and was the first popular format to support 24-bit color. TGA supports uncompressed and RLE compressed formats.

TIFF
Tag Image File Format, originally released by Aldus Corporation is a standard file format found in most paint, imaging, and desktop publishing programs. TIFF is extremely powerful, flexible, and extensible. However, because of its many nuances and its wide scope, TIFF can be a difficult format to understand as well as a difficult format to support.

U

Unsharp Mask Filter
The unsharp mask filter, also called the difference of Gaussian or DoG filter, is a type of convolution filter used for sharpening an image. The unsharp mask gets its backwards sounding name from the fact that it takes a Gaussian filtered version of the image and takes the difference between it and the original image to de-emphasize the low frequency components, and thus emphasize the high frequency components of the image.

V

Value/Volume
The brightness of a pixel in the HSV colorspace.

W

WMF
The Microsoft Windows Metafile format is used to store vector and bitmap image data. Because of its popularity and wide use, many Windows and non-Windows applications support WMF.

 

 


©2016. Accusoft Corporation. All Rights Reserved.

Send Feedback