Description
A device for determining moisture content of an object and a
method thereof
The present invention relates to a device and a method for
determining moisture content of an object, particularly by
using audio signals.
Moisture may be defined as a relatively small quantity of
diffused water. All materials contain at least a diminutive
volume of moisture as part of their molecular makeup.
However, the relative percentage of moisture in a substance
is dynamic.
Any given material will absorb moisture relative to ambient
temperature and humidity conditions through hygroscopic
action. Moisture content is a critical component of material
quality and its measurement is essentially a function of
quality control in most production and laboratory facilities.
In an entire spectrum of organizations ranging from
pharmaceutical manufacturers to food producers, moisture
content greatly influences the physical properties and
product quality at all stages of processing and even to the
final product existence or preservation.
Weight, thermal expansion, amalgamation, and electrical
conductivity are some of the properties of a material that
are influenced by a change in the moisture content. A
processed material's moisture content will define, for
example, the shelf life, the reactivity of chemical compounds
in inventory, or the binding properties of bulk materials. As
a result, the ability to accurately identify moisture content

levels during initial selection and processing procedures is
paramount to the quality of the final product.
The typical scenario faced in the food industry, for example
cashew industry is described here. The raw cashews are
separated from cashew apples and are taken to the factory for
further processing. The cashew comprises of a kernel covered
by a thick, hard, but porous shell. Due to porous nature of
outer shell, cashew absorbs moisture from surroundings. The
typical moisture content of a well suited cashew feed ranges
from 6% to 10%. Cashew having moisture content greater than
10% is considered as bad quality yield and is unaccepted for
production.
The measurement of moisture content using thermo gravimetric
analysis defines moisture as the loss of mass of a substance
when heated, by the process of water vaporisation. The
difference in mass is continually calculated and recorded by
a precision balance. Sample substance mass is measured before
and after the drying process for final moisture determination
on percentage basis. Methods for heating include oven drying,
microwave drying processes, halogen and infrared drying. The
disadvantages of the technique are its destructive nature and
the time consumed.
It is an object of the present invention to provide non-
destructive way of measuring moisture content.
The said object is achieved by a method for determining
moisture content of an object according to claim 1 and by a
device for determining moisture content of an object
according to claim 6 and by a system according to claim 10.

The underlying idea is to determine the moisture content of
an object using the audio signals. The method involves
obtaining the said audio signal of a sound, generated by one
of the objects colliding with another one of the objects
using a signal capturing module. Then the method involves,
comparing the obtained audio signal with a plurality of
reference audio signals corresponding to objects of known
moisture content and finally determining the moisture content
of the object based on the comparison of the audio signal
with the plurality of the reference audio signals using a
signal processing module. Since the moisture content of the
reference audio signals are known, the comparison or matching
of the audio signal of unknown moisture content with the
reference audio signal of known moisture content will give a
relative information on the moisture content of the samples
under examination. Since the method wholly depends on the
processing of the acquired sound information corresponding to
the collision of the objects, the objects still remain intact
and it is a non destructive procedure. For example in a
practical scenario, in a food processing industry like cashew
industry, to assess the quality of the raw cashews, the audio
signals processed could be the sound captured during the
collision of the raw cashews.
In a preferred embodiment, comparing the audio signal with a
reference audio signal gives a value indicating the
difference between the audio signal and the reference audio
signal. This enables to identify how close the audio signal
matches the reference audio signal.
In a further preferred embodiment, comparing the audio signal
with the plurality of reference audio signals further
comprising: sub-dividing the audio signal into audio samples
of a pre-determined time interval, comparing individual audio

samples with the plurality of reference audio signals, for
each of the individual audio samples, finding the value that
indicates the minimum difference between the individual audio
sample and the plurality of reference audio signals ; and
finding an average of the values obtained for all the
individual audio samples. Sampling of the audio signals helps
in faster processing and thereby aids in minimizing the error
in the matching. Hence this will give a more accurate level
of moisture content in the object.
In an alternative embodiment, the method further comprises
the step of comparing the average of the values with a
plurality of threshold values corresponding to known moisture
contents of the objects. The threshold values are obtained by
comparing audio signals corresponding to objects having known
moisture content. This further enables the accuracy of the
detection. Since the threshold values correspond to known
moisture content, the comparison of the average value with
said threshold will give the closeness of the average value
with the threshold, so that the exact moisture content could
be arrived at, based on the closest threshold.
In an alternative embodiment, the device further comprises an
indicator module to indicate the moisture content of the
object. This enables the user to know the moisture content of
the object and take actions based on the information.
In another alternative embodiment, the device further
comprises a holding device adapted to hold the plurality of
objects. The device further comprises an actuating module
adapted to move the holding device enabling the collision of
the objects inside the holding device. This enables the
generation of sound signals for analysis.

The present invention is further described hereinafter with
reference to illustrated embodiments shown in the
accompanying drawings, in which:
FIG 1 illustrates the block diagram of the device for
determining moisture content of a plurality of objects
according to an embodiment of the invention,
FIG 2 illustrates a spectrogram plot of sound profiles of dry
samples of cashews according to an embodiment of the
invention,
FIG 3 illustrates a spectrogram plot of sound profiles of wet
samples of cashews according to an embodiment of the
invention,
FIG 4 illustrates the comparison of two audio signal
sequences using a grid according to an embodiment of the
invention,
FIG 5 illustrates a block diagram, representing a system in
an industrial setup where a control unit is used to monitor
multiple devices used for determining moisture content of
objects, and
Fig 6 illustrates the arrangement of an actuating module with
respect to a holding device, according to an embodiment of
the invention.
The present invention will be described by taking a typical
scenario faced in the cashew industry, but the invention is
not limited to this application rather could be extended for
any other objects, for which the moisture content need to be
determined. The traditional way of finding the moisture

content is described below. Typically a hand full of cashews
is taken in hand. The cashews are made to jiggle forcing them
to collide with one another thereby creating characteristic
sound. This is done by lateral movement of cashew filled
palms. Based on the rattling sound of cashews, judgment is
made about moisture content of cashew. This method is purely
based on the experience of individual and cannot
differentiate marginal differences in moisture content. Also
it is unscientific in its approach.
FIG 1 illustrates the block diagram of the device 100,
according to one embodiment of the invention for determining
moisture content of a plurality of objects 102. The device
100 comprises a signal capturing module 104 for obtaining an
audio signal of a sound generated by one of the objects 102
colliding with another one of the objects 102. In this
typical scenario, the objects for example are the cashews.
Hence the signal capturing module 104, which could be a
sensitive microphone or a piezo sensor or any other acoustic
sensor, obtains the audio signals 106 corresponding to the
sound generated by the collision. The device 100 further has
a signal processing module 108 for comparing the audio signal
106 with a plurality of reference audio signals 110
corresponding to the same objects 102, but of known moisture
content. The comparison finally helps in determining the
moisture content of the object 102.
The signal processing module 108 could further comprise a
signal conditioning module 112, which could include an audio
signal filter and/or an amplifier. The signal conditioning
module 112 ensures that the audio signals 106 are properly
conditioned to remove the noise signals and other
abnormalities prior to converting the audio signals to
digital signals. The conversion of analog signal to the

digital signal is done using an analog-to-digital convertor
(ADC) 114.
The signal processing module 108, further has a processor
116, which is the heart of the signal processing module 108,
and is adapted to determine the moisture content of the
object 102 by performing a comparison between the digitized
audio signal 106 and the reference audio signal 110. The
reference audio signals 110 for the comparison could be pre-
stored in a memory 118 or could be provided at the time of
comparison, using an interface. The processor used could be
but not restricted to a digital signal processor (DSP), a
field-programmable gate array (FPGA), a micro processor or a
micro controller. The processor will control the sampling
rate of the ADC depending on the conditions and further it
will run a signal processing algorithm. The signal processing
module 108, also has a wired interface 120 and wireless
interface 122 for communicating to any external environment.
The device 100 further comprises an indicator module 124 to
indicate the moisture content of the object. The indication
could be a display or it could be an audio indication or even
both.
The method by which the moisture content of the objects is
determined using the device 100 will be explained below. When
taking the scenario of cashews, wet samples and dry samples
produce a distinct characteristic sound, when the cashews
strike one another. Higher the moisture content of the raw
cashew, damper will be the sound emitted. The invention makes
use of pattern matching techniques, using the audio signals
for example matching technique like Dynamic Time Warping
(DTW). When the group of raw cashews (dry or wet) strikes
each other, it produces distinct sound patterns. The acoustic

signals produced by the above impact have been recorded using
sensitive microphones.
FIG 2 illustrates a spectrogram plot 200 of sound profiles of
dry samples of cashews according to an embodiment of the
invention. FIG 3 illustrates a spectrogram plot 300 of sound
profiles of wet samples of cashews according to an embodiment
of the invention. From the spectrograms it could be observed
that spectral characteristic of dry samples are spread in
wide range of frequency bands and of wet samples fall in
specific frequency bands and are narrow band signals.
The first column 202 and the first column 302 in FIG 2 and
FIG 3 respectively show the time domain representation of the
signal captured. The second column 204 and second column 304
in FIG 2 and FIG 3 respectively shows the frequency
spectrogram representation. The X-axis of all the columns
represent time and the Y-axis for the column 202 and column
302 represent amplitude and Y-axis of column 204 and column
304 represent frequency. One of the pattern matching
algorithms that could be used here is dynamic time warping
(DTW) technique to classify wet and dry samples respectively
by determining the moisture content.
Dynamic time warping is an algorithm for measuring similarity
between two sequences which may vary in time or speed. In
general, DTW is a method that allows a computer to find an
optimal match between two given sequences. The sequences are
"warped" non-linearly in the time dimension to determine a
measure of their similarity. A frequency domain calculation
is more preferred here because the audio samples in the
frequency domain are very much distinguishable irrespective
of the size of cashews colliding each other.

FIG 4 illustrates the comparison of two audio signal
sequences using a grid 400 according to an embodiment of the
invention. The example shows the two sequences of the audio
signals arranged on the sides of a grid 400 with the unknown
sequence on the bottom (eight observations in the example)
i.e along the X-axis and the reference audio samples of the
signal of known moisture content up the left hand side (six
observations)i.e along the Y-axis. Both sequences start on
the bottom left of the grid 400.
Both the signals are sampled generally at equal intervals, so
that the comparison between the samples is more easy and
accurate. Comparing the audio signal with a reference audio
signal gives a value indicating the differences between the
audio signal and the reference audio signal.
To enable the comparison, the audio signal 106 is sub-divided
into audio samples of a pre-determined time interval. When
using DTW, the reference audio signals, having known moisture
content also is sampled at the same interval. Then each
individual audio sample in the X-axis is compared with the
plurality of reference audio signals in the Y-axis. For each
of the individual audio samples, the value that indicates the
minimum difference between the individual audio sample and
the plurality of reference audio signals is determined in the
grid. For example the first audio sample 1 in the grid 4 00 is
compared against each of the six reference audio samples. In
the present example, the fist audio sample matches closely
with the first reference audio signal 1 in the grid. Hence a
minimum value 402, corresponding to the two matching values
is indicated in the grid. For each of the further audio
signal samples, the minimum values for example 404, 406 etc
is determined. Then an average value is obtained from all the
values obtained for all the individual audio samples. Then

the average of the value is compared with a plurality of
threshold values corresponding to known moisture contents of
the objects. The threshold values are obtained by comparing
audio signals corresponding to objects having known moisture
content.
FIG 5 illustrates a block diagram, representing a system 500
in an industrial setup where a control station 508 is used to
monitor multiple devices used for determining moisture
content of objects. In FIG 5, the first device 502, the
second device 504 and the third device 506 could be used to
determine the moisture content of the same type of objects or
even different type of objects. The number of devices used in
the arrangement can vary based on the requirement. The
control station further comprises or connected to an
indication means 510 to indicate the moisture content of the
objects. This can again be a visual or audio indication. If
each of the devices discussed are enabled with some wireless
interface then even remote controlling of the said devices
from a remote location is possible. Basically the control
station 508 could communicate to the devices using a
communication channel 512. This communication channel 512
could be wired or wireless. The control station 508 can even
reside on a remote location, far from the actual devices.
FIG 6 illustrates the arrangement 600 of an actuating module
604 with respect to a holding device 602, according to an
embodiment of the invention. Practically to run the device
and to capture the audio signals of the objects colliding
together, the objects 102 are put in a holding device 602.
The said holding device 602 is then moved in such a way that
the objects collide. Then using a sensitive microphone for
example the audio signals are captured and then forwarded for
analysis . The actuating module 604 is used to enable the

movement of the holding device 602 for the collision of the
objects 102. The actuating module 604 could be a DC motor,
which drives the holding device 602. There can be different
ways in which the collision of the objects could be realized.
In the specific embodiment, the inner walls of such a holding
device 602 are covered by a soft material so as to absorb the
sounds from the striking of the raw cashews with the wall.
The objects, whose moisture content need to be determined
could be of a different types not restricted to cashews. The
invention can be extended to find the moisture content of any
object which absorbs moisture using this acoustic method.
Although the invention has been described with reference to
specific embodiments, this description is not meant to be
construed in a limiting sense. Various modifications of the
disclosed embodiments, as well as alternate embodiments of
the invention, will become apparent to persons skilled in the
art upon reference to the description of the invention. It is
therefore contemplated that such modifications can be made
without departing from the spirit or scope of the present
invention as defined.

We claim:
1. A method of determining moisture content of a plurality
of objects (102), comprising:
obtaining an audio signal (106) of a sound generated by
one of the objects (102) colliding with another one of
the objects (102);
comparing the audio signal (106) with a plurality of
reference audio signals (110) corresponding to objects
(102) of a known moisture content; and
determining the moisture content of the objects (102)
based on the comparison of the audio signal (106) with
the plurality of the reference audio signals (110).
2. The method according to claim 1, wherein comparing the
audio signal (106) with a reference audio signal (110) gives
a value indicating the difference between the audio signal
(106) and the reference audio signal (110) .
3. The method according to claim 1 or 2, wherein comparing
the audio signal (106) with the plurality of reference audio
signals (110) further comprising:
sub-dividing the audio signal (106) into audio samples
of a pre determined time interval,
comparing individual audio samples with the plurality of
reference audio signals (110),
for each of the individual audio samples, finding the
value that indicates a minimum difference between the
individual audio sample and the plurality of reference
audio signals (110) ; and
finding an average of the values obtained for all the
individual audio samples.

4. The method according to claim 3, further comprising the
step of comparing the average of the values with a plurality
of threshold values corresponding to known moisture contents
of the objects (102).
5. The method according to any of the preceding claims,
wherein the threshold values are obtained by comparing
reference audio signals (110) corresponding to objects (102)
having known moisture content.
6. A device (100) for determining moisture content of a
plurality of objects (102), comprising:
signal capturing module (104) for obtaining an audio
signal (106) of a sound generated by one of the objects
(102) colliding with another one of the objects (102);
and
signal processing module (108) for comparing the audio
signal (106) with a plurality of reference audio signals
(110) corresponding to objects (102) of a known moisture
content and determining the moisture content of the
object (102) based on the comparison.
7. The device (100) according to claim 6, further comprises
an indicator module (124) to indicate the moisture content of
the objects (102).
8. The device (100) according to claim 6 or 7, further
comprises a holding device (602) adapted to hold the
plurality of objects (102).
9. The device (100) according to any of the claims 6 to 8,
further comprises an actuating module (604) adapted to move
the holding device (602) enabling the collision of the
objects (102) in the holding device (602).

10. A system (500) for determining moisture content of a
plurality of objects (102), comprising:
a plurality of devices (502, 504, 506), said device as
claimed in any of the claims 6 to 9;
a control station (508), said control station (508)
adapted to receive the information associated with the
determination of the moisture content of the objects
(102) from the plurality of devices (502, 504, 506); and
a communication channel (512) to transfer the
information associated with the determination of the
moisture content of the objects (102) between the
plurality of devices (502, 504, 506) and the control
station (508).
11. The system (500) according to claim 10, wherein the
communication channel (512) is wired or wireless.
12. The device (500) according to claim 10 or 11, wherein the
control station (508) further comprise an indication means
(510) to indicate the moisture content of the objects (102).

The present invention provides a non-destructive way of
measuring moisture content of an object (102). The method
first involves obtaining an audio signal (106) of a sound
generated by one of the objects (102) colliding with another
one of the objects (102). Then the method involves comparing
the audio signal (106) with a plurality of reference audio
signals (110) corresponding to objects (102) of known
moisture content; and then determining the moisture content
of the object (102) based on the comparison of the audio
signal (106) with the plurality of the reference audio
signals (110).