Lab Instructions

Lab 4, due 5:30 pm October 1, 2007

EES5053/ES4093: Remote Sensing, UTSA

Student Name: ___________________

Digital Image Processing I: Atmospheric Correction, Radiance,

Reflectance, NDVI from Landsat Image

Objective: In this lab, you will learn the basic procedure of digital image
processing from converting the digital number to radiance, and then to
reflectance by using band math. Based on reflectance, you will learn how
to calculate the NDVI, and classify the NDVI using Density slice.

Part I: Concepts and short questions:

1. Give a four bands image of 4 x 3 as the following, please replace them into BSQ,
BIP, and BIL formats.

1, 4, 5, 6 2, 5, 8, 9 105, 79, 103, 114 20, 25, 31, 26

3, 4, 4, 5 10, 3, 2, 1 114, 100, 124, 134 29, 19, 29, 50
2, 3, 9, 10 9, 8, 7, 6 130, 125, 109, 150 25, 30, 33, 50

2. Calculate the two multivariate parameters (covariance and correlation) of the

above 4 bands. Identify which two bands have the best correlation.

3. Summarize the differences and similarities of supervised and unsupervised

classifications. You may use examples to illustrate them.

4. Assume that two road intersections shown on a photograph can be located on a

1:25,000 scale topographic map. The measured distance between the intersections
is 47.2 mm on the map and 94.3 mm on the photograph. (a) What is the scale of
the photograph? (b) at that scale, what is the length of a fence line that measures
42.9 mm on the photograph?

5. The area of a lake is 52.2 cm2 on a 1:7500 vertical photograph. find the ground
area of the lake in m2, ha, and acres

1
Part II

1. Preparation:
(1). Create a Lab4 directory under c:\Fall2007-RS\YourName\. Today we will use the
atmospherically corrected image (with mask applied) that you processed in Lab3:
Mask_DOS_020708.tif. Save your results to Lab4 directory.

2. Spectra radiance calculation

Equation 1 is the basic equation for calculating spectral radiance from the Digital
Number (DN) of Landsat 4, 5 and 7:

 LMAX  LMIN 
L    * ( DN  QCALMIN )  LMIN
 QCALMAX  QCALMIN  (1)

where, DN is the Digital Number of each pixel in the image (in this lab, it will be
Mask_DOS_020708.tif, although a simple DOS atmospheric correction has been performed, it is
still DN, not radiance yet), LMAX and LMIN are the calibration constants, and
QCALMAX and QCALMIN are the highest and the lowest points of the range of
rescaled radiance in DN.

For Landsat 7, However, there is a more simple way for calculating L (Landsat 7
Science User Data Handbook Chap.11, 2002). This is what we will use in this lab.

L  gain * DN  offset
(2)

In Equation 2, the “gain” corresponds to the “Gain” in the header file, and the “offset”
corresponds to the “Bias” in the header file. The unit of radiance is W m-2 sr-1 um-1

The image NE_p27r40_020708_12347.tif you worked on in Lab 3 was just a portion

(or a subscene) of the original full image of LE70270400000218950. This image can be
downloaded free from the TexasView Remote Sensing Consortium that UTSA (LRSG) is
the member of the Consortium. From the LRSG website (www.utsa.edu/LRGS/), there is
a link to the website. Now I would like you to explore this website (there are many cool
stuff for you to explore) and to know how you may download data from there for your
own research. Also you can see the header file information you need for this lab. When
are in the TexasView website, click Landsat Data under Data Archives on the left hand
column. Now you should a figure like this:

2
 This is the available Landsat 7 images for the Texas and its adjacent states. You
can see the Path (east to west) from 24-33 and Row (north to south) from 35-42. Path
tells you each satellite orbit, row tells the descending or ascending of the orbits.

Question1, from this figure, is Landsat 7 a descending or ascending satellite for

the day time passing Texas (usually 10:00 am). What is descending or ascending?

Path and Row together define the one Landsat image. For San Antonio, it is
Path27, Row40 or 2740 in the image. Click the 2740 in the image, you will see there are
4 years of images available for San Antonio: y1999, y2001, y2002, y2003. The image we
are using is in July 8, 2002. Click y2002, you will see three images in different days
(Julie day): d189, d317, d365). The Julie Day for July 8, 2002 is the day 189. so click
189. you will see two different types of images. What we want to use is the ‘nlaps’. Click
the ‘nlaps’, you will see the many image files there: the same file name with different file
extension. H1, H2, H3, HI are header files; I1, I2, …, I9 are real images. To understand
what they mean, please click the README.TXT file. I would encourage you to
understand all of them using your spare time, since it is the first step for you to really get
familiar with a satellite sensor and what many parameters mean. If you have any
question about them, you are always welcomed to ask me.

You should be able to find the gain and offset information from those files for the
Lab. But to save your time, I list them for you to use.

Band | Ref DN to Radiance Default

| Detector gain offset Abs Calib?

-------------------------------------------------------
1 | 15 0.775686 -6.20000 FALSE
2 | 12 0.795686 -6.39999 FALSE
3 | 8 0.619216 -5.00000 FALSE
4 | 7 0.965490 -5.10001 FALSE
5 | 14 0.125725 -0.99999 FALSE
6 | 8 0.066823 0.000000 FALSE

3
 7 | 10 0.043726 -0.35000 FALSE
8 | 27 0.971765 -4.70000 FALSE
9 | 8 0.037059 3.200000 FALSE

In this lab, we only do band 3 and band 4. You can use the Band Math tool to do so.
From the main ENVI menu, click Basic Tools -> Band Math, type the equation for band
3 as the figure 1 below, and click OK. A new window will popup, select the atmospheric
corrected band 3 (Mask_DOS_020708.tif) as b3, then save this image to your directory. In
the similar way, you can do the band 4. Output as Radn_189_b3.tif and Radn_189_b4.tif

Figure 1. Band math expressions

Question 2, calculate and show the basic statistics of the two radiance images, and give a brief
analysis of the statistics.

3. Spectra reflectance calculation

The reflectance for band  is computed by the following equation (Markham and
Barker,1986 and Landsat 7 Science User Data Handbook Chap.11, 2002):

  L  d 2
 
ESUN   cos  (3)

where L is at satellite spectral radiance which is the outgoing radiation energy of the
band observed at the top of atmosphere by the satellite (in this lab, we use the results
calculated from step 3), d is the Earth-Sun distance in astronomical units (), ESUN  is

4
mean solar exoatmospheric irradiances for the band , and cos is the cosine of the
solar incident angle. Supposing a horizontal land surface is flat, the cosine of solar
incident angle (cos) can be calculated from the Sun Elevation cos(90-SunElevation).
The Sun elevation angle for the image is 65.26º (you can get this from the head file of
LE7027040000218950.H1 mentioned above)

Since the inverse of d2 (which is 1/d2) in Equation 3 is equivalent to “inverse squared

relative distance Earth-Sun, dr“, the Equation 3 can be rewritten as:

  L
 
ESUN   cos  d r (4)

The annual averaged value of dr is 1.0, and it ranges from about 0.97 to 1.03. You can
find a real number for a special date (such as the189 day: July 8 for this image used is
1.0167) from Table 11.4 of this link here at:
http://ltpwww.gsfc.nasa.gov/IAS/handbook/handbook_htmls/chapter11/chapter11.html

The values for ESUN in Equation 4 are given in Table 3. The value of ESUN  for band 6
is not available.

Table 3. ESUN for Landsat 4 and 5 TM in mW/cm2/μm (Markham and Barker, 1986),
and for Landsat 7 ETM+ in W/m 2/μm (Landsat 7 Science User Data Handbook Chap.11,
2002)

Band1 Band2 Band3 Band4 Band5 Band6 Band7

Landsat-4 TM 195.8 182.8 155.9 104.5 21.91 - 7.457
Landsat-5 TM 195.7 182.9 155.7 104.7 21.93 - 7.452
Landsat 7 ETM+ 1969 1840 1551 1044 225.7 - 82.07

In this lab, we only calculate the reflectance of band 3 and band 4, using the Band
Math tool as in step 2, output as Reflt_189_b3 and Reflt_189_b4.

Question 3. Calculate and compare the basic statistics of reflectance Band 3 and 4 .

4. Calculate NDVI

NDVI stands normalized difference of vegetation index: the difference between the near
infrared band (~0.83 µm) and the red band (0.66 µm). For Landsat image (TM or ETM+
image), they are band 4 and band 3, respectively. Thus, the NDVI can be calculated based
on this equation: (b4-b3)/(b4+b3), using the reflectance band 3 and band 4 as input, name
as NDVI_02189.

Question 4. Show the basic statistics of NDVI.

5
5. Density Slice

In the image window, click Overlay->Density Slice. In the density slice window, select
one of the item, then edit the data range and color, click apply. Then click file->Save
Range to your folder as NDVI_class.

Question 5. Upload an image as RGB742, link it with NDVI Density Slice (gray).
Please make a simple discussion/comparison about the spatial distribution of the NDVI
values (you can edit the range of NDVI to match the RGB) and vegetation coverage.