New Scientist - USA (2013-06-08)

20 | NewScientist | 8 June 2013

A CHAINSAW revs in a remote

swathe of the Indonesian

rainforest. Within minutes,

rangers appear as if from

nowhere, stopping illegal

loggers in their tracks and saving

countless trees. How did they

know? A tip off from a recycled

cellphone hanging hundreds

of metres away in the forest.

That’s the vision of Topher

White, founder of Rainforest

Connection, based in San

Francisco. The non-profit

organisation is launching a pilot

project this month in the forests

of Indonesia that uses modified

Android smartphones to record

and identify the sound-signatures

of chainsaws.

At first, Rainforest Connection

will use new phones donated for

the trial, though White ultimately

plans to use recycled handsets

that supporters contribute when

they upgrade to the latest model.

The phones are outfitted with

solar panels specifically designed

to take advantage of the brief

periods when light reaches the

but we find out after, so we

cannot trace when it happens,”

says Dwiati Novita Rini, who

works on reforestation of cleared

land in Sumatra with Birdlife

International. Conservation

groups can also pay police to

perform aerial surveys of areas

vulnerable to logging, but they are

too expensive to do frequently.

For its initial trial, Rainforest

Connection will work with the

conservation group Kalaweit to

place and test 15 phone rigs in the

25,000-hectare Air Tarusan reserve

in western Sumatra. White hopes

each phone will have a listening

radius of 0.5 kilometres, providing

a low-cost way to monitor remote

stretches of jungle.

Indonesia loses more than a

million hectares of forest a year,

according to an estimate by

Rainforest Action Network. The

country’s rainforest is the third

largest in the world, and home

to many unique native species

of plants and animals. But more

than half of it has been cleared

since the 1960s.

Eventually, White hopes to

simplify the technology so that

locals can plug a phone into a

box, nail it to a tree, and begin

tracking loggers right away.

“We’ll ultimately rely upon locals

to intervene when an ‘event’

is detected. Making it simple,

effective and accessible for them

is our first priority.” n

forest floor. Their microphones

stay on at all times, and software

listens for the telltale growl of a

chainsaw, which triggers an alert.

Initially, only rangers will be

notified, but White hopes to

release a free app that lets anyone

receive real-time alerts with the

audio that the phones pick up and

the location. “We want to make

people feel like they are taking

part in the dramatic events on

the front lines of environmental

protection,” he says.

Current efforts to stop loggers

in Indonesia are limited. “We can

find out how much forest has

been cut using satellite images,

• Not hard to hear–

Cat Ferguson

Trunk call to save forest

In the Indonesian rainforest, has-been cellphones could help stop illegal logging

“The system could help you

assess the BMI and state of

health of people you might

find on online dating sites”

Yusuf Ahm

Ad/reuT

ers

TeChNOLOGY

Photo of your face

is all it takes to

predict your BmI

THERE’S a lot you can tell from a face –

and now there’s one more thing: body

mass index. Software that can predict

your BMI from a snapshot of your face

could turn a simple headshot into a

revealing portrait of your build, and

even your risk of certain diseases.

BMI is a standard health metric

that’s equal to a person’s weight (in

kilograms) divided by the square of

their height (in metres). Someone

with a BMI over 30 is acknowledged as

obese and below 18.5 as underweight.

How we perceive someone’s weight

based on images of their face is

strongly correlated with actual body

weight and associated health risks

(Perception, doi.org/c67jd5). Building

on that knowledge, Guodong Guo and

colleagues at West Virginia University

in Morgantown have now produced an

algorithm that can analyse a mugshot

and predict that person’s BMI.

The software assesses seven

weight-related components in a

face image, including the ratios of

cheekbone width to jaw width, face

length to cheekbone width and the

average distance between eyebrow

and eye. They then ran the program

across images of 14,500 faces of

people with known BMIs. The

predicted BMIs were mostly within

two or three points of the person’s

actual BMI (Image and Vision

Computing, doi.org/mnz).

That’s just a first attempt, Guo

says. He says that tweaking the

software to analyse more facial

features should improve the results.

“This could be used in smart health

applications, relating face images to

BMI and associated health risks,” Guo

says. “Or on online dating sites, for

instance, it could help you assess the

BMI and state of health of people you

might date.”

Ioanna Tzoulaki, an epidemiologist

at Imperial College London, says such

analyses could be useful if fat in the

face can be correlated with patterns

of disease, but she worries about

discrimination. “What can this be

used for except dating sites? People

often have facial images in their CVs.

Is it right to assign people into obesity

groups from these?” Paul marks n

130608_N_TechSpread.indd 20 4/6/13 10:55:56