Chapter 8: Drone questions

- Can fly up to 120m, or 120m above the tallest building

- Minimum of 3-mile visibility
- Remote ID: module that broadcasts signal for drone info
- Fly within 1 hour of local solar noon
- Min 75% overlap
- Radiometric calibration: converts DN to reflectance
- All remotely sensed images have errors/ distortions
o Shape, size, scale distortions
- Orthophotos: Photos that have been corrected & celebrated to coord system

Chapter 9:

Error types

- Systematic
o Always there, unavoidable
o Earth's rotation, curvature, atmospheric refraction, topographic effects,
Relief displacement
Tall objects displaced away from center of image
- Random
o Changes in flight altitude, attitude or/and velocity

Geometric Correction: aligning pixels to correct planimetric (x, y) map locations

- Operators
o Georeferencing
No spatial reference (coord system)
o Georectification
Already georeferenced, but adjusted for correction
o Image to map rectification

Steps:

1. Choosing a reference source

a. Select source w/ real world geographic coords
i. Must be larger than image
2. Select ground control points
a. Easily identifiable on air photo & reference photo.
 b. Known coordinates and spatial reference
c. Good GCP
i. Stable & well defined (building corners)
d. Bad GCP
i. Moveable (shorelines, vehicles)
e. Should be evenly distributed
f. Minimum of 3 GCP, the more the better
g. For homogeneous areas, must create own markers for GCP
3. Image Transformation
a. Uses GCPs to derive transformation equations
b. Rotation, skewing, translation and/or scaling
4. Accuracy Assessment
a. Root mean square error (RMSE), how well images align
b. Lower value the better

Radiance (Lλ)

- Amount of radiant flux per unit solid angle and per unit projected area.
o Watts per square meter per Steradian
- Indicated the amount of energy per pixels on average (Digital number)

Reflectance

- 𝒓𝝀 = 𝑬𝑹(𝝀) / 𝑬𝑰(𝝀)
- Helps describe unique spatial characteristics
- EI is found by specific sensors on platforms (DLS)

Radiometric Calibration
Empirical Line Calibration (ELC) method

- Pseudo-invariant reference targets. One dark one bright

o Compares new measurements to black (0) and white (100%) targets
o Higher resolution is needed to remove over saturation effect
Ideally it is a little lower then 255 (8-bit) to know that there is not
saturation
- R=a*DN+b
o Points on graph should be 0,0 and 100, ~255.
Find slope (y2-y1)/(x2-x1) ---> ~1/255
- All is needed is a panel with known reflectance and percentages
o With assumption of 0,0
 - Relative calibration method

Chapter 10:

Definition
The science and tech of obtaining reliable spatial measurements and geometric info about
spatial objects/phenomena

Photogrammetry vs RS, 3 differences

1. Imaging system: RS uses multispectral. Photogrammetry uses an imaging system

(i.e LiDAR sensor)
2. Data: photogrammetry doesn't require images to be georectified, can be 2D or 3D
3. Application: photogrammetry involves determining spatial properties and
dimensions of objects

Metric Photogrammetry:
- makes precise measurements and evaluates exact sizes, shapes, positions,
distance, angles, areas, volumes, and elevations of surface features.
Interpretative Photogrammetry:
- deals principally in recognizing and identifying surface objects and judging their
significance rather than making direct measurements and calculations.

Photo scale & Characteristics

- Unitless
- Photo scale IS map scale

Equation:

- Focal length, f
- Flying altitude above sea level, H
- Mean elevation of the terrain, h
- Flying altitude above terrain, H – h

- ∴ (𝑎′𝑏′)/ (𝐴𝐵) = 𝑓/ (𝐻 − ℎ)
- ∵ ∆𝐴𝐿𝐵 ∼ ∆𝑎′𝐿𝑏′
 o F/(H-h)

Ex.

- F=8mm, flying altitude AGL=120m (AGL above ground level, = H-h)

o 8mm/120m *1000/1000 (convert to meters) 8m/120,000m --> 1/15,000

Finding cell size:

- Ex: given cell size of 1/15,000, sensor size= 7.12x5.33

o Photo L/ Ground L --> 7.12/ (1/15,000) --->7.12*15,000 =106,800mm --->
106.8m
Ground L is photo scale
- Can take cell size and divide it by image resolution (how many pixels in image)
o Prob need to convert to cm due to small number (*100)

With changing terrain (elevation)

- Need average scale S(avg)

o S(avg) = f/ (H-h(avg))
Going in reverse

- Image to photo scale & spatial resolution

o Need to know # of pixels in image (image dimensions)
o For an estimate: A known measurement in image
Ground control point w/ known measurement
Feature w/ known measurement. i.e. football field
o For precise:
Measure a feature on an image (i.e. parking garage length) then go to
google earth to find the same feature and measure.
For best results measure on the ground in the field (in situ)
o RS at best should always be paired w/ in situ data
Measured image/real world measure = scale
Photo distance/ ground distance = scale

Area Measurement
- Irregular shapes split into regular shapes
o Convert measurements to real world distances using photo scale
S=1/60 area= 5*4 ---> (5*60) *(4*60) =72,000cm
 Can also: photo area * S^2, (area is squared i.e. m^2)

Chapter 11:

[The simple camera

- Aperture: how wide the opening is, how much light enters the camera (d)
o f-stops
Smaller number= larger opening
o Effects depth of field- range of distances that appear sharp
Larger aperture allows for a blurry background
Areial photography uses a small aperture to create sharp images
- Shutter: How long is the opening open. Duration of exposure (t)
o Slow shutter creates motion blur
o High shutter speed creates motion freeze
- Exposure: total amount of light energy (E) exposed onto the sensor per unit area
- ISO: Camera’s sensitivity to light (like radiometric resolution)
o Effects brightness
Too high ISO will lead to a grainy texture/ noise
Best to keep ISO at base unless necessary
o Exponential brightness between values

Exposure for Aerial Photography

- Better to be slightly underexposed. Small Apture, fast shutter speed

o Creates a dark image, but prevents data loss due to hitting upper limit
Brightness can be retroactively adjusted

Filters
- Transparent materials placed in the optical path in front of a camera lens to control
which EMR wavelengths can enter
- Allows for more precision with sensors
- In black & white photography, yellow filters are used to block blue
o Increases contrast

Filter types

- Low-pass: allow short wavelengths

 o Notch filters
- High-pass: allow long wavelengths
o Wratten High-pass filters: used in RS to block short wavelengths
- Band-pass: allows specific range of wavelengths
- Band-stop: blocks specific range of wavelengths
- UV Filters: reduces scattering, reduces salt & pepper effect & haziness

DSLR Camera

- Digital single-lens reflex camera

o Reflex- A mirror directs incoming light. Allow light to hit sensor directly

Single-lens Frame Camera

- Larger windows for bands. Overlap of bands. Cannot separate

Multi-lens frame camera

- Separate narrow bands

Chapter 12:

Thermal in Remote sensing

- Thermal radiation is a type of longwave EMR within the infrared portion of the
spectrum, ranging from 3 to 14 μm.
- All objects on Earth emit thermal radiation, with the amount of TIR energy emitted
per unit area directly proportional to their temperature.
o Allows TIR sensors to measure an object’s surface temperature.

Surface temp:

- Derived from Stefan-Boltzmann Law: M= 𝝈T^4

Calculating surface temperatures from remotely sensed images

1. Convert digital numbers into top of atmosphere radiance (TOA)

2. TOA radiance to TOA brightness temperature
3. Calculate surface emissivity
a. Emissivity: effectiveness in emitting energy as thermal radiation
i. Ranges from 0-1
ii. Black Body has an emissivity of 1
 1. Black body: Absorbs 100% of incident energy. 100%
transferred into thermal energy
iii. Ratio between Thermal energy radiated and contained in a material
1. How well a substance retains heat
iv. Ratio of thermal radiation emitted by a material to the radiation
emitted by a black body at the same temperature
1. How efficient to convert short wave to long wave radiation
4. Derive the LST

Raster example:

- In the summer there are pixel values in the negatives for surface temperature
o From clouds, can produce negative temperature even in summer
Ice crystals
o Water may still be seen as dark, but only due to contrast for easy viewing.

Urbanization

- Land use land cover change (LULCC)

o Land change, human-induced modification to the earth's surface

Impact of urbanization on climate

1. Increased air & surface temps

a. Absorbed during day and slowly released at night
b. Industrial process releases heat
2. Increased surface runoff
a. More impervious surfaces than vegetation can absorb water
3. Decrease in biodiversity
a. Replacement of native ecosystems
b. Introduction of non-native ornamental vegetation
4. Downgraded Air Quality
a. Pollution --> ozone
b. LULCC contributed to higher aerosols

Urban heat island effect (UHI

- Urban areas exhibit higher temps compared to surrounding rural areas

 o Increased energy & water use --> degraded air quality
o Extended duration of heat waves
- Heat island Intensity: Change in temp, urban-rural
- Types of UHI- All dependent on how data is gathered
o Boundary layer
100-2,000m above ground. Highest
Uses Weather balloons
o Surface layer
Highest point within AOI (tallest point in city)
Rooftop weather stations or Eddy-Covariance flux towers
o Canopy layer
Air layer within urban canopy
Typically, 2m above the surface
Issue: Temp is affected by the land cover
o Surface UHI
Remotely sensed TIR imagery
Ensures distribution
Aka- Skin Temperature
Issue: Temporal, spatial resolution, cloud coverage.
RS sees top of everything. Leads to inconsistency