IEEE Geoscience and Remote Sensing Magazine - March 2019 - 66

Count of Instruments

250
200

4
3
2
1
0

100
50

600

Eight

Seven

Six

Five

Four

Three

Two

VHR

Number of Channels
fIGURE 6. The average group valuations for the numbers of
separate spectral channels. The values on the x-axis might look
equally distributed, but they are not. Because they represent only
the values from the sample, they have a discrete character.

2) The spread between the maximum and minimum bounds
of wavelengths associated with higher-valuation companies lies between 380 and 470 nm (right part of Figure 5).
3) The trendline built using the least-squares method
exhibits a negative slope.
We can, therefore, conclude that the modern EO market has consolidated applications with established spectral ranges, the use of which is valued more highly than
other ranges. The range of interest lies within rather narrow limits but covers a large number of spectral channels:
visible light in all channels and the NIR spectrum. We can
suppose that this wavelength range is driven primarily by

TaBLE 4. cLaSSIFIcaTION BY maSS [29].
cLaSS

LaUNch maSS (kg)

Femtosatellites

10.1

Picosatellites

0.1-1

nanosatellites

1-10

microsatellites

10-100

minisatellites

100-500

medium satellites

500-1,000

Large satellites

21,000

TaBLE 5. cLaSSIFIcaTION BY SPr [30].

Nano
Micro
Mini
Medium
Nano
Micro
Mini
Medium
Nano
Micro
Mini
Medium
Nano
Micro
Mini
Medium

150

One

Mean Group Valuations (US$ Millions)

300

cLaSS

raNGE (m)

VHr

1-5

5-30

230

fIGURE 7. The count of instruments in each mass group over

SPR groups.
the application areas for satellite imagery of interest to
the market.
From each start-up, we collected information about applications for which its satellite imagery could be used. The
words most often used by EO start-ups in the description
of applications are agriculture, disaster, intelligence, monitoring,
infrastructure, and forestry. It can be assumed that these applications are the most popular in the market and determine
the technical capabilities required by satellite sensors. For
each of these areas, images on the visible spectrum are valuable sources of information, while data from other spectral
bands play more auxiliary roles, e.g., by complementing optical information with additional insights (consider, for instance, multispectral applications aimed at material detection and building monitoring in the construction industry).
These highly valued spectral ranges determine the number
of separate channels where imagery is available, which is the
next feature we analyzed (Figure 6). The area of visible light
is divided into three standard channels (RGB) and the NIR
channel as a fourth source of information.
We can observe that missions with five separate channels
are the most highly valued by venture capitalists. The fifth
channel in such cases is a panchromatic channel, which gives a
picture in gradations of gray; but, in many cases, it has a higher
SPR. Some companies from the sample provide an additional
(sixth) channel of visual data, separating the red-edge region of
the electromagnetic spectrum: the visible red portion lying in
the range of 690-730 nm. At this range, the low vegetation reflectance in the red section is immediately followed by a sharp
increase to the high reflectance in the NIR. The high valuation
of these start-ups seems to suggest that agriculture applications
are considered some of the most important areas for space imagery utilization. Lastly, Figure 6 shows that hyperspectral data
(column "600" on the chart) is valued by venture capitalists
lower than ordinary multispectral data, coming from three to
six separate spectral channels.
We also considered SPR as a characteristic that affects the
final product and, therefore, the company's valuation during
the process of investment rounds. In Figure 7, we grouped the
EO missions in the sample according to generally accepted
mass and spatial classifications (Tables 4 and 5).
ieee Geoscience and remote sensing magazine

march 2019

IEEE Geoscience and Remote Sensing Magazine - March 2019

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - March 2019