lelap/gist:ed02822f3654da3505ad

## gistfile1.txt
Useful davinci image processing scripts - All from Christensen, 2010 Remote Sensing Lab, ASU

1) a = read(“filename”)
Read in a file into array ‘a’
Examples:
t = read(“Atlanta_AST_11_25_2006_surface_temp.vic")
Reads the 830 pixel by 700 pixel by 1-band temperature image into ‘t’
em = read(“Atlanta_AST_11_25_2006_emissivity.vic")
Reads the 830 pixel by 700 pixel by 5-band emissivity image into ‘em’

2) display(a[ , , 2])
Display all rows and all columns of band 2 an image (must specify either 1 or 3 bands.  Image must be in byte format).

3) b = sstretch(t, ignore = 0)
Perform a gaussian stretch on image ‘a’ (see write-up for more options).  Ignore= 0 option ignores zero (black) pixels when computing the stretch parameters. This is very useful for converting a floating-point image (e.g. temperature) into a 0-255 DN byte image.

4) display(sstretch(t, ignore=0))
Stretch and display all rows, all columns of band 2 in single step.

5) display(sstretch(em[ , , 2], ignore=0))
Stretch and display all rows, all columns of band 2 in single step. (Remember, display only displays 1 or 3 bands)

6) pplot(em[ , 100, 3],  y1 = .8,  y2 = 1.02)
Plot  all elements of the 100th row of band 3 of image ‘em’.  Set y plot range to go from 0.8 to 1.02.  See pplot write-up for lots more useful options

7) pplot(t[75,  , 1])
Plot all elements of the 75th column of band 1 of image ‘t’

8) pplot(em[130,90])
Plot the emissivity spectrum for pixel 130, 90

9) c = boxfilter(t, 7,  ignore=0)
Perform a 7x7 boxcar (low-pass) filter on image ‘t’.  Ignore zero values when doing the filter.

10) d = t – c
Difference two images (or arrays).  If ‘t’ was the original image and ‘c’ was a low-pass filter image, then ‘d’ would be a high-pass filtered image

11) h = histogram(t[ , , 1], steps=256)
pplot(h[2], xaxis=h[1])
Compute histogram of all x and y elements in band 1 of image in 256 bins.  The min and max values in the image are found automatically.  The histogram value are in h[2]; the bin (x-axis) values are in h[1].  Ploting using pplot as shown plots the data in their true bins.
	Useful davinci image processing scripts - All from Christensen, 2010 Remote Sensing Lab, ASU

	1) a = read(“filename”)
	Read in a file into array ‘a’
	Examples:
	t = read(“Atlanta_AST_11_25_2006_surface_temp.vic")
	Reads the 830 pixel by 700 pixel by 1-band temperature image into ‘t’
	em = read(“Atlanta_AST_11_25_2006_emissivity.vic")
	Reads the 830 pixel by 700 pixel by 5-band emissivity image into ‘em’

	2) display(a[ , , 2])
	Display all rows and all columns of band 2 an image (must specify either 1 or 3 bands. Image must be in byte format).

	3) b = sstretch(t, ignore = 0)
	Perform a gaussian stretch on image ‘a’ (see write-up for more options). Ignore= 0 option ignores zero (black) pixels when computing the stretch parameters. This is very useful for converting a floating-point image (e.g. temperature) into a 0-255 DN byte image.

	4) display(sstretch(t, ignore=0))
	Stretch and display all rows, all columns of band 2 in single step.

	5) display(sstretch(em[ , , 2], ignore=0))
	Stretch and display all rows, all columns of band 2 in single step. (Remember, display only displays 1 or 3 bands)

	6) pplot(em[ , 100, 3], y1 = .8, y2 = 1.02)
	Plot all elements of the 100th row of band 3 of image ‘em’. Set y plot range to go from 0.8 to 1.02. See pplot write-up for lots more useful options

	7) pplot(t[75, , 1])
	Plot all elements of the 75th column of band 1 of image ‘t’

	8) pplot(em[130,90])
	Plot the emissivity spectrum for pixel 130, 90

	9) c = boxfilter(t, 7, ignore=0)
	Perform a 7x7 boxcar (low-pass) filter on image ‘t’. Ignore zero values when doing the filter.

	10) d = t – c
	Difference two images (or arrays). If ‘t’ was the original image and ‘c’ was a low-pass filter image, then ‘d’ would be a high-pass filtered image

	11) h = histogram(t[ , , 1], steps=256)
	pplot(h[2], xaxis=h[1])
	Compute histogram of all x and y elements in band 1 of image in 256 bins. The min and max values in the image are found automatically. The histogram value are in h[2]; the bin (x-axis) values are in h[1]. Ploting using pplot as shown plots the data in their true bins.