爱华网Inimageprocessing,normalizationisa process that changes the range ofpixelintensityvalues. Applications include photographs withpoorcontrastdueto glare, for example. Normalization is sometimes called contraststretching or histogram stretching. In more general fields of dataprocessing, such asdigitalsignal processing, it is referred toasdynamicrangeexpansion.[1]
The purpose of dynamic range expansion in the various applicationsis usually to bring the image, or other type of signal, into arange that is more familiar or normal to the senses, hence the termnormalization. Often, the motivation is to achieve consistency indynamic range for a set of data, signals, or images to avoid mentaldistraction or fatigue. For example, a newspaper will strive tomake all of the images in an issue share a similar rangeofgrayscale.
Normalization transforms an n-dimensionalgrayscaleimagewithintensity values in the range (Min,Max), into a newimagewithintensity values in the range (newMin,newMax).
Thelinearnormalizationof agrayscaledigitalimageis performed according to theformula
For example, if the intensity range of the image is 50 to 180 andthe desired range is 0 to 255 the process entails subtracting 50from each of pixel intensity, making the range 0 to 130. Then eachpixel intensity is multiplied by 255/130, making the range 0 to255.
Normalization might also be non linear, this happens when thereisn't alinearrelationshipbetweenand.An example of non-linear normalization is when the normalizationfollows asigmoidfunction, in that case, the normalized image is computedaccording to the formula
Wheredefinesthe width of the input intensity range,anddefinesth e intensity around which the range is centered.[2]
Auto-normalization in image processing software typicallynormalizes to the full dynamic range of the number system specifiedin the image file format. The normalization process will produceiris regions, which have the same constant dimensions, so that twophotographs of the same iris under different conditions will havecharacteristic features at the same spatial location.
Fromwiki
