Generally, the invention relates to thermal imaging systems. More
specifically, the invention relates to a method and system for performing
thermal screening.
BACKGROUND OF INVENTION
[002] In situation of epidemics, such as Ebola, Influenza, Severe Acute
Respiratory Syndrome (SARS), and now, novel Coronavirus-19 (nCOVID19), transmission risk is high specially at crowded places. The recent
outbreak of nCovid-19 has jeopardized the millions of lives across the world.
To prevent the transmission of such contagious diseases, usually, a variety
of Personal Protective Equipment (PPE) are recommended. The PPE may
include a head protector, an eye protector, a respiratory protector, a hand
protector, and a foot protector. Since fever and high temperature are
common symptoms of such contagious diseases, therefore, in addition to
screening people for PPE compliance, thermal screening is conducted as
preventive safety measure at places like airports, shopping malls, etc.
[003] Conventional techniques of screening a large crowd for PPE
compliance and for elevated body temperature are time consuming and
difficult due to two fundamental reasons. Firstly, the staff screening the
crowd for long durations may have to deal with physical strain and fatigue.
Secondly, it is not feasible to deploy a large workforce to conduct such
screenings at all the crowded places. Additionally, some of the convention
techniques employ two separate modalities – one based on an infrared
sensor to detect elevated body temperature (EBT) and other based on a
visible-light sensor to detect presence of the PPE. However, it is difficult to
perform EBT evaluation over PPE. Thus, the conventional techniques still
need to depend on manual supervision and are limited in their scope and
utility. In conclusion, conventional techniques are expensive, operationally
Docket No: IIP-HCL-P0031IN1
-3-
inefficient (e.g., require greater computational power, consume higher
power, require higher maintenance, etc.), large in size and weight, thereby
limiting the deployment options in real world.
SUMMARY OF INVENTION
[004] In one embodiment, a method of performing thermal screening of a
subject is disclosed. The method may include determining a presence and
a category of each of one or more personal protective equipment (PPE)
worn by the subject based on a thermal image of the subject, using an image
classification model. The method may further include determining a viability
of performing elevated body temperature (EBT) evaluation of the subject
based on the presence and the category of each of the one or more PPE. It
should be noted that the viability may be a positive viability when it is
possible to perform a substantially accurate EBT evaluation. The method
may further include performing the EBT evaluation of the subject based on
the thermal image of the subject, upon predicting the positive viability.
[005] In another embodiment, a system for performing thermal screening
of a subject is disclosed. The system may include a processor and a
memory communicatively coupled to the processor. The memory may store
processor-executable instructions, which, on execution, may causes the
processor to determine a presence and a category of each of one or more
personal protective equipment (PPE) worn by the subject based on a
thermal image of the subject, using an image classification model. The
processor-executable instructions, on execution, may further cause the
processor to determine a viability of performing elevated body temperature
(EBT) evaluation of the subject based on the presence and the category of
each of the one or more PPE. It should be noted that the viability may be a
positive viability when it is possible to perform a substantially accurate EBT
evaluation. The processor-executable instructions, on execution, may
further cause the processor to perform the EBT evaluation of the subject
Docket No: IIP-HCL-P0031IN1
-4-
based on the thermal image of the subject, upon predicting the positive
viability.
BRIEF DESCRIPTION OF THE DRAWINGS
[006] The present application can be best understood by reference to the
following description taken in conjunction with the accompanying drawing
figures, in which like parts may be referred to by like numerals
[007] FIG. 1 is a block diagram of an exemplary system for performing
thermal screening of a subject, in accordance with some embodiments of
the present disclosure.
[008] FIG. 2 is a functional block diagram of an exemplary thermal
screening system configured to perform thermal screening of a subject, in
accordance with some embodiments of the present disclosure.
[009] FIG. 3 is a block diagram of an exemplary deep learning image
classification model, in accordance with some embodiments of the present
disclosure.
[010] FIG. 4 is a flow diagram of an exemplary process for performing
thermal screening of a subject, in accordance with some embodiments of
the present disclosure.
[011] FIG. 5 are flow diagram of an exemplary process for determining a
presence and a category of a Personal Protective Equipment (PPE), in
accordance with some embodiments of the present disclosure.
[012] FIG 6 is a flow diagram of a detailed exemplary process for
performing thermal screening of a subject, in accordance with some
embodiments of the present invention.
DETAILED DESCRIPTION OF THE DRAWINGS
[013] The following description is presented to enable a person of ordinary
skill in the art to make and use the invention and is provided in the context
of particular applications and their requirements. Various modifications to
the embodiments will be readily apparent to those skilled in the art, and the
Docket No: IIP-HCL-P0031IN1
-5-
generic principles defined herein may be applied to other embodiments and
applications without departing from the spirit and scope of the invention.
Moreover, in the following description, numerous details are set forth for the
purpose of explanation. However, one of ordinary skill in the art will realize
that the invention might be practiced without the use of these specific
details. In other instances, well-known structures and devices are shown in
block diagram form in order not to obscure the description of the invention
with unnecessary detail. Thus, the present invention is not intended to be
limited to the embodiments shown, but is to be accorded the widest scope
consistent with the principles and features disclosed herein.
[014] While the invention is described in terms of particular examples and
illustrative figures, those of ordinary skill in the art will recognize that the
invention is not limited to the examples or figures described. Those skilled
in the art will recognize that the operations of the various embodiments may
be implemented using hardware, software, firmware, or combinations
thereof, as appropriate. For example, some processes can be carried out
using processors or other digital circuitry under the control of software,
firmware, or hard-wired logic. (The term “logic” herein refers to fixed
hardware, programmable logic and/or an appropriate combination thereof,
as would be recognized by one skilled in the art to carry out the recited
functions.) Software and firmware can be stored on computer-readable
storage media. Some other processes can be implemented using analog
circuitry, as is well known to one of ordinary skill in the art. Additionally,
memory or other storage, as well as communication components, may be
employed in embodiments of the invention.
[015] Referring now to FIG. 1, a block diagram of a system 100 for
performing thermal screening of a subject is illustrated, in accordance with
some embodiments of the present disclosure. In an embodiment, the
thermal screening system 102 may be used to resolve aforementioned
problems by eliminating the requirement of a visible-light sensor to detect
the presence and a type/category of a Personal Protective Equipment (PPE)
Docket No: IIP-HCL-P0031IN1
-6-
worn by the subject. Thus, the thermal screening system 102 may use only
thermal images not only to perform an accurate Elevated Body Temperature
(EBT) evaluation of the subject, but also to detect the presence and
category of the PPE worn by the subject.
[016] The thermal screening system 102 may include a memory 104, a
processor 106, an input/output (I/O) device 108, and a thermal imagining
camera 110. Examples of the I/O device 108 may include a speaker, and a
display. The speaker may be used to instruct the user (e.g., an
administrator, a staff, or the subject) via an audio/voice note. Similarly, the
display may show a message corresponding to the audio/ voice note. By
way of an example, the speaker and display may instruct the subject to
come inside a Field of View (FoV) of the thermal imaging camera 110 (e.g.,
come closer, shift right, shift left, and the like.). By way of another example,
the speaker and the display may also be used to instruct the subject to
remove a specific type of PPE (e.g., an eyewear or a mask), to notify the
user about completion of EBT evaluation, and to provide directions to the
user to perform further check.
[017] In some embodiments, the I/O device 108 may further include a user
interface 112. The user may interact with the thermal screening system 102
and vice versa through the user interface 112. By way of an example, the
user interface 112 may be used to provide results of analysis performed by
the thermal screening system 102, to the user. By way of another example,
the user interface 112 may be used by the user (e.g., administrator) to
provide inputs to the thermal screening system 102.
[018] As will be described in greater detail herein below, in order to perform
thermal screening, the thermal screening system 102 may acquire thermal
images of the subject by employing the thermal imaging camera 110.
Additionally, in some embodiments, the thermal screening system 102 may
extract thermal images of subjects (with or without one or more PPE) from
a server 114, which is further communicatively coupled to a database 116.
Docket No: IIP-HCL-P0031IN1
-7-
As will be appreciated, such pre-stored thermal images may be required for
training or retraining the thermal screening system 102 from time to time.
[019] The memory 104 and the processor 106 of the thermal screening
system 102 may perform various functions including acquiring the thermal
image, determining the presence and the category of the PPE, determining
a viability of performing the EBT evaluation, performing the EBT evaluation,
and notifying the subject. The memory 104 may store instructions that, when
executed by the processors 106, cause the processors 106 to perform
thermal screening of the subject, in accordance with some embodiments of
the present invention. The memory 104 may also store various data (e.g.,
thermal images, image classification model, etc.) that may be captured,
processed, generated, and/or required by the thermal screening system
102. The memory 104 may be a non-volatile memory (e.g., flash memory,
Read Only Memory (ROM), Programmable ROM (PROM), Erasable PROM
(EPROM), Electrically EPROM (EEPROM) memory, etc.) or a volatile
memory (e.g., Dynamic Random Access Memory (DRAM), Static RandomAccess memory (SRAM), etc.).
[020] In some embodiments, the thermal screening system 102 may
interact with the user (e.g., administrator, security staff, etc.) via external
devices 118 over a communication network 120. In such embodiments, the
thermal screening system 102 may render the results to the user via the
user interface 112 over the external devices 118. For example, the security
staff may get a notification over the external devices 118 to take a required
action for EBT evaluation. The one or more external devices 118 may
include, but may not be limited to, a desktop, a laptop, a notebook, a
netbook, a tablet, a smartphone, a remote server, a mobile phone, or
another computing system/device. The communication network 120 may be
any wired or wireless communication network and the examples may
include, but may be not limited to, the Internet, Wireless Local Area Network
(WLAN), Wi-Fi, Long Term Evolution (LTE), Worldwide Interoperability for
Microwave Access (WiMAX), and General Packet Radio Service (GPRS).
Docket No: IIP-HCL-P0031IN1
-8-
[021] Further, the thermal screening system 102 may interact with the
external devices 118 and/or the server 114 for sending/receiving various
data. The server 114 is further communicatively coupled to a database 116,
which may store the images sought or acquired by the thermal screening
system 102. Similarly, for example, in some embodiments, the thermal
screening system 102 may interact with one or more external devices 118
for sending and receiving various data.
[022] Referring now to FIG. 2, a functional block diagram of an exemplary
thermal screening system 200 (analogous to the thermal screening system
102) configured to perform thermal screening of a subject is illustrated, in
accordance with some embodiments of the present disclosure. The thermal
screening system 200 may acquire a thermal image 202 of the subject in
order to perform thermal screening of the subject. The thermal image 202
of the subject may be acquired using a thermal imaging camera (similar to
the thermal imaging camera 110). The thermal image 202 may be captured
from a pre-defined distance and may be focused, for example, on face of
the subject. For example, in some embodiments, at least seventy percent
of the thermal image should cover a face portion of the subject. In particular,
the thermal screening system 200 may organize various functions under
various modules including, but not limited to, a PPE presence and category
determination module 204, a viability determination module 206, an EBT
evaluation module 208, and a notification generation module 210.
[023] The PPE presence and category determination module 204 may be
configured to receive the thermal image 202 of the subject from the thermal
imaging camera. Further, the presence and category determination module
204, based on the thermal image, may determine a presence and a
category of each of one or more PPE worn by the subject. The one or more
PPE may include, but may not be limited to, a headgear, an eyewear, a face
mask, a respiratory mask, a hand glove, and an earwear. In order to perform
its function, the PPE presence and category determination module 204 may
include an Image Classification Model 212. The image classification model
Docket No: IIP-HCL-P0031IN1
-9-
212 may include at least one of image processing based image
classification model, machine learning image classification model, and deep
learning image classification model. By way of an example, the deep
learning image classification model may include an artificial neural network
(ANN) based multilevel image classifier. An internal architecture of such a
deep learning image classification model is explained in greater detail in
conjunction with FIG. 3. Classification of a plurality of PPE in different
categories may be explained in conjunction to FIG. 5. Further, the PPE
presence and category determination module 204 may be communicatively
connected to the viability determination module 206 in order to process the
determined results.
[024] The viability determination module 206 may be configured to
determine a viability of performing EBT evaluation of the subject. The
viability may be determined based on the presence and the category of each
of the one or more PPE. The viability may be a positive viability or a negative
viability. The viability is the positive viability when it is possible to perform a
substantially accurate EBT evaluation. Otherwise, for the negative viability,
results after the EBT evaluation may be inaccurate. Therefore, the EBT
evaluation is only performed in case of positive viability prediction. When
the negative viability is predicted, the subject needs to remove at least one
of the one or more PPE to get accurate EBT evaluation results. Further, the
viability determination module 206 may be communicatively coupled to the
EBT evaluation module 208 and a notification generation module 210. In
case of positive viability, a signal of positive viability prediction may be
transmitted to the EBT evaluation module 208. And, for the negative
viability, a signal may be transmitted to the notification generation module
210.
[025] The EBT evaluation module 208 may be configured to perform the
EBT evaluation of the subject. The EBT evaluation may be performed based
on the thermal image of the subject, upon prediction of the positive viability.
Further, the notification generation module 210 may be configured to
Docket No: IIP-HCL-P0031IN1
-10-
generate a notification, upon prediction of the negative viability. Also, the
notification generation module 210 may be configured to notify the subject
to remove at least one of a plurality of PPE based on the viability of
performing the EBT evaluation and based on the presence and the category
of each of the one or more PPE. The notification generation module 210
may also notify the security staff when the subject doesn’t take a required
action or remove the at least one of a plurality of PPE, which is required to
be removed for the accurate EBT evaluation.
[026] It should be noted that the thermal screening system 102, 200 may
be implemented in programmable hardware devices such as programmable
gate arrays, programmable array logic, programmable logic devices, or the
like. Alternatively, the thermal screening system 102, 200 may be
implemented in software for execution by various types of processors. An
identified engine/module of executable code may, for instance, include one
or more physical or logical blocks of computer instructions which may, for
instance, be organized as a component, procedure, function, or other
construct. Nevertheless, the executables of an identified engine/module
need not be physically located together but may include disparate
instructions stored in different locations which, when joined logically
together, comprise the identified engine/module and achieve the stated
purpose of the identified engine/module. Indeed, an engine or a module of
executable code may be a single instruction, or many instructions, and may
even be distributed over several different code segments, among different
applications, and across several memory devices.
[027] Referring now to FIG. 3, a block diagram of an exemplary deep
learning image classification model 300 is illustrated, in accordance with
some embodiments of the present disclosure. As stated above, the deep
learning image classification model comprises an artificial neural network
(ANN) based multilevel image classifier. As illustrated, the deep learning
image classification model 300 may include an input layer 302, an output
layer 304, and a set of hidden layers. For example, in the illustrated
Docket No: IIP-HCL-P0031IN1
-11-
embodiment, the deep learning image classification model 300 may include
a total of six hidden layers i.e., five convolution layers 306a-306e (for
example, a first convolution layer 306a, a second convolution layer 306b, a
third convolution layer 306c, a fourth convolution layer 306d, and a fifth
convolution layer 306e) and a dense layer 312. The input layer 302 may be
fed with a thermal image of a subject. The input layer 302 may process the
thermal image in grayscale with suitable resizing of the thermal image to the
first convolution layer 306a. It should be noted that each of the convolution
layers 306a to 306e may filter an output generated by its preceding layer.
[028] Further the deep learning image classification model 300 may also
include a set of pooling layers 308a-308e (i.e., a first pooling layer 308a, a
second pooling layer 308b, a third pooling layer 308c, a fourth pooling layer
308d, and a fifth pooling layer 308e) preceded by the corresponding set of
convolution layers 306a-306e. For example, the first convolution layer 306
is followed by the first pooling layer 308a, the second convolution layer 306
is followed by the second pooling layer 308b, and so forth. Thus, the second
convolution layer 306b may filter the output generated by the first pooling
layer 308a. Further, the pooling layers 308a-308e may be employed to
reduce the dimensions of the input feature maps thereby helps in generating
a noise-free or less noisy output. The convolution layers and pooling layers
are repeated many times (say, five times) ensuring that the visual
information is extracted at multiple levels (i.e., from global details to finer
details). The dense layer 312 may be a fully connected layer that is trained
to provide a multi-class (quaternary) label.
[029] In some embodiments, the image classification model 212, such as
the deep learning image classification model 300, may be trained with a
plurality of thermal images associated with a plurality of subjects. The
plurality of thermal images may include the plurality of subjects with one or
more PPEs as well as the plurality of subjects without the PPE. It should be
noted that the plurality of PPE may include, but may not be limited to, a
Docket No: IIP-HCL-P0031IN1
-12-
headgear, an eyewear, a face mask, a respiratory mask, a hand glove, and
an earwear.
[030] As will be appreciated by one skilled in the art, a variety of processes
may be employed for performing thermal screening of a subject. For
example, the exemplary system 100 and associated thermal screening
system 102, 200 may perform thermal screening of the subject, by the
process discussed herein. In particular, as will be appreciated by those of
ordinary skill in the art, control logic and/or automated routines for
performing the techniques and steps described herein may be implemented
by the system 100 and the associated thermal screening system 102, 200
either by hardware, software, or combinations of hardware and software.
For example, suitable code may be accessed and executed by the one or
more processors on the system 100 to perform some or all of the techniques
described herein. Similarly, application specific integrated circuits (ASICs)
configured to perform some or all the processes described herein may be
included in the one or more processors on the system 100.
[031] Referring now to FIG. 4, an exemplary process 400 for performing
thermal screening of a subject is depicted via a flowchart, in accordance
with some embodiments of the present disclosure. Each of the steps of
process 400 may be performed by a thermal screening system, such as the
thermal screening system 102, 200.
[032] At step 402, a presence and a type (i.e., category) of each of one or
more PPE worn by the subject may be determined from a thermal image of
the subject. In some embodiments, the thermal image of the subject may
be acquired by a thermal imaging camera, such as the thermal imaging
camera 110. It should be noted that the thermal imaging camera may
acquire thermal image of a particular subject within a FoV of the thermal
imaging camera. The thermal imaging camera may be calibrated focusing
more on the subject’s face. Further, an image classification model, such as
the image classification model 212 may be used to determine the presence
and the category of each of one or more PPE worn by the subject.
Docket No: IIP-HCL-P0031IN1
-13-
[033] In some embodiments, the image classification model may include at
least one of image processing based image classification model, machine
learning image classification model, and deep learning image classification
model. Additionally, in some embodiments, the deep learning image
classification model may include an artificial neural network (ANN) based
multilevel image classifier. An exemplary deep learning image classification
model has already been explained in detail in conjunction with Fig. 3. Thus,
for example, the multilevel image classifier may include a set of convolution
layers and a fully connected dense layer. Each of the set of convolution
layers is followed by a pooling layer. Each of the set of convolution layers is
configured for filtering the output of a previous layer, while each of the set
of pooling layers is configured for reducing dimensions of an input feature
map. The image classification model, such as the deep learning image
classification model, may be trained with a plurality of thermal images
associated with a plurality of subjects. With regards to the plurality of thermal
images, in some embodiments, the plurality of subjects may be wearing a
plurality of PPE. Additionally, in some embodiments, the plurality of subjects
may not be wearing any PPE. It should be noted that the plurality of PPE
may include at least one of a headgear, an eyewear, a face mask, a
respiratory mask, a hand glove, and an earwear.
[034] At step 404, a viability of performing EBT evaluation of the subject
may be determined. The viability may be determined based on the presence
and the category of each of the one or more PPE. It may be noted that the
viability is a positive viability when it is possible to perform a substantially
accurate EBT evaluation. In some embodiments, to determine the viability
of performing the EBT evaluation, an overall presence of the PPE over face
of the subject may be ascertained, based on the presence and the category
of each of the one or more PPE. It should also be noted that, in some
embodiments, the overall presence may be based on a weighted average
of the presence of each of the one or more PPE based on the category of
each of the one or more PPE. For example, an eyewear may be assigned
Docket No: IIP-HCL-P0031IN1
-14-
with a highest weight, while an earwear may be assigned with a second
highest weight, as they affect most the EBT evaluation. Similarly, a weight
may be assigned to each of the other PPE. The overall presence score may
then be compared with a pre-defined threshold value to determine the
viability of performing EBT evaluation of the subject.
[035] By way of an example, consider a scenario where a subject is
wearing the eyewear, the face mask, and the hand glove at a same time.
Further, by way of example, the headgear, the eyewear, the face mask, the
respiratory mask, the hand glove, and the earwear are assigned with
weights ‘0.3’. ‘0.7’, ‘0.4’, ‘0.4’, ‘0’, and ‘0.6’. In that situation the thermal
imaging camera may capture a thermal image of the subject wearing the
eyewear, the face mask, and the hand glove and process the image using
the image classification model. The image classification model may
determine the weighted average for the eyewear, the face mask, and the
hand glove based on their respective weights. Thereafter, the weighted
average value may be compared with a pre-defined threshold value in order
to determine viability of EBT evaluation. For example, if the weighted
average value is less than the threshold value, the viability is the positive
viability.
[036] Thereafter, at step 406, the EBT evaluation of the subject may be
performed. It should be noted that the EBT evaluation of the subject is also
performed based on the thermal image of the subject. Also, it should be
noted that the EBT evaluation may be performed only upon predicting the
positive viability of EBT evaluation.
[037] In some embodiments, the process 400 may further notify the subject
to remove at least one of the one or more PPE based on the viability of
performing the EBT evaluation. For example, when a subject is wearing an
eyewear and it is determined that the viability to perform a substantially
accurate EBT evaluation is negative, then the subject may be notified to
remove the eyewear. The process 400 may further perform the EBT
Docket No: IIP-HCL-P0031IN1
-15-
evaluation of the subject upon removal of the one or more PPE (e.g., the
eyewear) by the subject.
[038] Referring now to FIG. 5, an exemplary process 500 for determining
a presence and a category of a PPE is depicted via a flowchart, in
accordance with some embodiments of the present disclosure. It should be
noted that each step of the process 500 may be performed by the PPE
presence and category determination module 204. At step 502, the PPE
presence and category determination module 204 may receive a thermal
image of a subject (whose thermal screening is being performed) from a
thermal imaging camera, such as the thermal imaging camera 110.
[039] At step 504, the PPE presence and category determination module
204 may feed the thermal image to an image classification model, such as
the image classification model 212. In some embodiments, the image
classification model may be a deep learning image classification model,
such as the deep learning image classification model 300. Further, as
described above in conjunction to FIG. 3, the deep learning image
classification model may include an ANN based multilevel image classifier.
At step 506, the image classification model may detect a presence and
determine a category of the PPE based on the thermal image. By way of an
example, the category may include, but may not be limited to, a headgear
506a, an eyewear 506b, a face mask 506c, a respiratory mask 506d, and
no face covering 506e. The headgear 506a may include a plurality of
elements worn on head and made of different materials. Examples of the
headgear 506a may include, but may not be limited to, a helmet, a hat, and
a cap. The eyewear 506b may include items worn over the eyes for either
protection or enhancement of visual acuity. Examples of the eyewear 506b
may include, but may not be limited to, a spectacle, a sunglass, and a
goggle. The face mask 506c and the respiratory mask 506d may refer to a
protective equipment worn over face that primarily covers the nose and
mouth of a person. Examples of the face mask 506c and the respiratory
mask 506d may include, but may not be limited to, a surgical mask, a dust-
Docket No: IIP-HCL-P0031IN1
-16-
mask, and a N95 respirator. Further, the category ‘no face covering’ 506e
may refer to an absence of any PPE over face of the subject. This
represents a bare/naked skin of the face that is free from all obstructions.
[040] At step 508, a presence of the PPE may be determined based on
the thermal image. It should be noted that the PPE is present, when the
category of the PPE belongs to at least one of a prespecified category, such
as the headgears 506a, the eyewear 506b, the face mask 506c, and the
respiratory mask 506d. Otherwise the PPE is not present. For example,
consider a scenario where the subject is wearing a hand glove and a foot
protector. In that case, they may be considered in the ‘no face covering’
category 506e as they may not affect thermal screening of the subject.
[041] Referring now to 6, an exemplary process 600 for performing thermal
screening of a subject is depicted in greater detail via a flowchart, in
accordance with some embodiments of the present disclosure. Each step
of the flow diagram 600 may be performed by various modules 204-210 of
the thermal screening system 102, 200.
[042] At step 602, a thermal imaging camera may acquire a thermal image
of the subject within FoV of the thermal imaging camera. At step 604, the
PPE presence and category determination module 204 may determine a
presence and a category of each of one or more PPE worn by the subject,
based on a thermal image of the subject, using an image classification
model. This has been already explained in conjunction with FIG. 5. In some
embodiments, the PPE presence and category determination module 204
may ascertain an overall presence of the PPE over face of the subject,
based on the presence and the category of each of the one or more PPE.
At step 606, the viability determination module 206 may check a presence
of an obstructing PPE (i.e., a PPE that obstructs an accurate EBT
evaluation). For example, in some embodiments, the viability determination
module 206 may check a presence of an eyewear over face of the subject.
As stated above, the eyewear may be assigned with a highest weight for
Docket No: IIP-HCL-P0031IN1
-17-
predicting a viability of performing EBT evaluation. The eyewear may
include sunglasses, spectacles, and a similar eye protecting equipment.
[043] At step 608, the EBT evaluation module 208 may perform the EBT
evaluation of the subject based on the thermal image of the subject, when
the eyewear is not present. However, at step 610, the notification generation
module 210 may generate a notification, if the eyewear is present. Further,
the notification may be transmitted to a user or the subject for removing the
eyewear, in order to perform an accurate EBT evaluation.
[044] Thus, the present disclosure may help in eliminating limitation of
conventional systems as discussed earlier. The disclosure introduces a
thermal screening system that uses only thermal images to determine
presence type of a Personal Protective Equipment (PPE) over face of a
subject and to perform an accurate EBT evaluation of the subject. Further,
the disclosure offers various advantages such as eliminating privacy
concerns (since visible-light image or fused thermal and visible-light image
is not employed), increasing accuracy of EBT evaluation (EBT evaluation is
performed only when substantially accurate EBT valuation is possible),
providing for illumination independency (some of the conventional system
rely on fluorescent films pasted over the different PPE material for the
visible-light sensors to detect bright-coloured area, to overcome
environmental factors (e.g., smoke, dust, and fog), increasing computational
efficiency, and lowering deployment cost. The use of a single thermal
imaging camera results in less power consumption, enhanced
compactness, greater deployment options, and cheaper production and
deployment cost for end-customers. Moreover, the disclosed system
requires less interfaces between the sensors and processor, and between
the sensor and display.
[045] It will be appreciated that, for clarity purposes, the above description
has described embodiments of the invention with reference to different
functional units and processors. However, it will be apparent that any
suitable distribution of functionality between different functional units,
Docket No: IIP-HCL-P0031IN1
-18-
processors or domains may be used without detracting from the invention.
For example, functionality illustrated to be performed by separate
processors or controllers may be performed by the same processor or
controller. Hence, references to specific functional units are only to be seen
as references to suitable means for providing the described functionality,
rather than indicative of a strict logical or physical structure or organization.
[046] Although the present invention has been described in connection
with some embodiments, it is not intended to be limited to the specific form
set forth herein. Rather, the scope of the present invention is limited only by
the claims. Additionally, although a feature may appear to be described in
connection with particular embodiments, one skilled in the art would
recognize that various features of the described embodiments may be
combined in accordance with the invention.
[047] Furthermore, although individually listed, a plurality of means,
elements or process steps may be implemented by, for example, a single
unit or processor. Additionally, although individual features may be included
in different claims, these may possibly be advantageously combined, and
the inclusion in different claims does not imply that a combination of features
is not feasible and/or advantageous. Also, the inclusion of a feature in one
category of claims does not imply a limitation to this category, but rather the
feature may be equally applicable to other claim categories, as appropriate.

CLAIMS
What is claimed is:
1. A method (400) of performing thermal screening of a subject, the method
(400) comprising:
determining (402), by a thermal screening system (200), a presence
and a category of each of one or more personal protective equipment (PPE)
worn by the subject based on a thermal image of the subject, using an image
classification model (212);
determining (404), by the thermal screening system (200), a viability
of performing elevated body temperature (EBT) evaluation of the subject
based on the presence and the category of each of the one or more PPE,
wherein the viability is a positive viability when it is possible to perform a
substantially accurate EBT evaluation; and
performing (406), by the thermal screening system (200), the EBT
evaluation of the subject based on the thermal image of the subject, upon
predicting the positive viability.
2. The method (400) as claimed in claim 1, comprising:
acquiring, by the thermal screening system (200), the thermal image
of the subject within a field of view (FoV) of a thermal imaging camera (110)
employed to acquire the thermal image.
3. The method (400) as claimed in claim 1, wherein the image classification
model (212) comprises at least one of an image processing based image
classification model, a machine learning image classification model, and a
deep learning image classification model (300).
4. The method (400) as claimed in claim 3, wherein the deep learning
image classification model (300) comprises an artificial neural network
(ANN) based multilevel image classifier, wherein the multilevel image
Docket No: IIP-HCL-P0031IN1
-20-
classifier comprises a set of convolution layers (306) and a fully connected
dense layer (312), wherein each of the set of convolution layers (306) is
followed by a pooling layer, wherein each of the set of convolution layers
(306) is configured for filtering the output of a previous layer, and wherein
the pooling layer (308) is configured for reducing dimensions of an input
feature map.
5. The method (400) as claimed in claim 1, comprising training the image
classification model (212) with a plurality of thermal images associated with
a plurality of subjects wearing a plurality of PPE, and wherein the plurality
of PPE comprises at least one of a headgear, an eyewear, a face mask, a
respiratory mask, and an earwear.
6. The method (400) as claimed in claim 1, wherein determining the viability
(404) of performing the EBT evaluation comprises:
ascertaining, by the thermal screening system (200), an overall
presence of the PPE over face of the subject based on the presence and
the category of each of the one or more PPE, wherein the overall presence
is based on a weighted average of the presence of each of the one or more
PPE, and wherein a weight is pre-defined based on the category of each of
the one or more PPE.
7. The method (400) as claimed in claim 1, comprising:
notifying, by the thermal screening system (200), the subject to
remove at least one of the one or more PPE based on the viability of
performing the EBT evaluation; and
performing, by the thermal screening system (200), the EBT
evaluation of the subject upon removal of the one or more PPE by the
subject.
Docket No: IIP-HCL-P0031IN1
-21-
8. A system (100) for performing thermal screening of a subject, the
system (100) comprising:
a processor (106); and
a memory (104) communicatively coupled to the processor (106),
wherein the memory (104) stores processor-executable instructions, which,
on execution, causes the processor (106) to:
determine (402) a presence and a category of each of one
or more personal protective equipment (PPE) worn by the subject
based on a thermal image of the subject, using an image
classification model (212);
determine (404) a viability of performing elevated body
temperature (EBT) evaluation of the subject based on the presence
and the category of each of the one or more PPE, wherein the
viability is a positive viability when it is possible to perform a
substantially accurate EBT evaluation; and
perform (406) the EBT evaluation of the subject based on the
thermal image of the subject, upon predicting the positive viability.
9. The system (100) as claimed in claim 8, comprising a thermal imaging
camera (110) to acquire the thermal image of the subject within a field of
view (FoV) of the thermal imaging camera (110).
10. The system (100) as claimed in claim 8, wherein the image
classification model (212) comprises at least one of an image processing
based image classification model, a machine learning image classification
model, and a deep learning image classification model (300).
11. The system (100) as claimed in claim 10, wherein the deep learning
image classification model (300) comprises an artificial neural network
(ANN) based multilevel image classifier, wherein the multilevel image
classifier comprises a set of convolution layers (306) and a fully connected
Docket No: IIP-HCL-P0031IN1
-22-
dense layer (312), wherein each of the set of convolution layers (306) is
followed by a pooling layer, wherein each of the set of convolution layers
(306) is configured for filtering the output of a previous layer, and wherein
the pooling layer (308) is configured for reducing dimensions of an input
feature map.
12. The system (100) as claimed in claim 8, wherein the processorexecutable instructions cause the processor (106) to train the image
classification model (212) with a plurality of thermal images associated with
a plurality of subjects wearing a plurality of PPE, and wherein the plurality
of PPE comprises at least one of a headgear, an eyewear, a face mask, a
respiratory mask, and an earwear.
13. The system (100) as claimed in claim 8, wherein the processorexecutable instructions cause the processor (106) to determine (404) the
viability of performing the EBT evaluation by ascertaining an overall
presence of the PPE over face of the subject based on the presence and
the category of each of the one or more PPE, wherein the overall presence
is based on a weighted average of the presence of each of the one or more
PPE, and wherein a weight is pre-defined based on the category of each of
the one or more PPE.
14. The system (100) as claimed in claim 8, wherein the processorexecutable instructions cause the processor (106) to:
notify the subject to remove at least one of the one or more PPE
based on the viability of performing the EBT evaluation; and
perform the EBT evaluation of the subject upon removal of the one or more PPE by the subject.