4SpatialFilteringLectureNotes - 1 GIS 4037c Dr Charles...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
1 GIS 4037c Dr. Charles Roberts Spatial Filtering So far, the contrast enhancement techniques that we have discussed involve moving pixels around on the grey scale. Each operation affects all pixels with the same grey scale, without reference to their location. Using contrast enhancement techniques, we can combine two BVs together, or we can create contrast between them by moving them further apart. However, we cannot break up the pixels in a single BV. For instance, no matter what mathematical operation we preform on pixels with a BV of, say, 47, they will all end up in the same class. Furthermore, any enhancement we make to BV 47 will affect every pixel with that BV, without regard to context. Spatial filtering is a way of enhancing spectral information by taking into account the surrounding pixels. It is also a technique that breaks up the number of pixels in any BV class, say 47. Basically, there are two things we can do with spatial filters. We can enhance details or we can suppress details. A digital image consists of both low and high spatial frequency information. Their sum constitutes the original image. Spatial frequency is the number of BV changes per unit distance. The following row of pixels have low frequency information: 5 5 5 4 5 5 5 5 6 6 6 6 6 6 6 Few changes = low frequency, similar regions, image looks smooth Low frequency land covers change brightness slowly as you move across the image: - Water - Forest Cover - Agriculture - Forest The next row has high frequency information: 5 9 1 2 5 5 5 7 0 3 7 1 0 1 0 Many changes in brightness values, image looks "rough". Note that there is no reference to the magnitude of change, just the notion that there is frequent change. In High Frequency imagery, land cover brightness changes frequently: - Specular reflectance on waves (water) - Slope contrasts (shadows) - Lithologic contacts
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 - Drainage networks - Urban cultural features Algorithms that perform spatial frequency enhancement are called filters because they pass or emphasize some spatial frequencies and suppress others. Filters that enhance high frequency information are called HIGH PASS FILTERS ; They enhance detail. They emphasize frequent changes in brightness values as you move from one pixel to another as you move across the image. We will examine these filters next week. Filters that enhance low frequency information are called LOW PASS FILTERS or smoothers, because they suppress detail, and emphasize continuous tones. They create regions of homogeneous brightness values. So filters can enhance or suppress detail, and they can actually reorganize brightness values around the values of adjacent pixels. .but what exactly is a filter? The Moving Window Concept
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/21/2010 for the course GIS 4037c taught by Professor Roberts during the Fall '10 term at FAU.

Page1 / 9

4SpatialFilteringLectureNotes - 1 GIS 4037c Dr Charles...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online