Scribd
Upload
73 views27 pages

A Knowledge-Driven Approach To Activity Recognition in Smart

The document discusses a knowledge-driven approach to activity recognition in smart homes. It proposes using ontological modeling and semantic reasoning to recognize activities of daily living based on data from multiple sensors. An ontology was developed to represent both sensor data and activities in a unified way. This allows using semantic reasoning for activity recognition at different levels of abstraction. It also allows the initial activity models to evolve through learning, combining traditional data-driven and knowledge-driven approaches. This makes the approach flexible, applicable and scalable for rapid development and deployment of assistive systems in smart homes.

Uploaded by

VPLAN INFOTECH
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
73 views27 pages

A Knowledge-Driven Approach To Activity Recognition in Smart

The document discusses a knowledge-driven approach to activity recognition in smart homes. It proposes using ontological modeling and semantic reasoning to recognize activities of daily living based on data from multiple sensors. An ontology was developed to represent both sensor data and activities in a unified way. This allows using semantic reasoning for activity recognition at different levels of abstraction. It also allows the initial activity models to evolve through learning, combining traditional data-driven and knowledge-driven approaches. This makes the approach flexible, applicable and scalable for rapid development and deployment of assistive systems in smart homes.

Uploaded by

VPLAN INFOTECH
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 27

A Knowledge-Driven Approach to Activity Recognition in Smart Homes

INTRODUCTION

• WITH the rising aging population and overstretched healthcare resources, technology-driven healthcare delivery to
support independent living has attracted increasing amounts of attention. Within this new paradigm, the concepts of
Smart Homes (SH) have recently emerged as a viable mainstream approach to achieving this goal. An SH is a
residential home setting augmented with a diversity of multimodal sensors, actuators, and devices along with
Information and Communication Technologies (ICT)-based services and systems. By monitoring environmental
changes and inhabitant’s activities, an assistive system in an SH can process perceived sensor data, make timely
decisions, and take appropriate actions to assist an inhabitant perform activities of daily living (ADL), thus
extending the period of time living independently within their own home environment. Currently, there are a
number of SH projects being developed for the purpose of proof-of-concept demonstration in addition to the
establishment of real living environments. There is a broad range of enabling technologies such as sensor networks,
data communications and devices, that provide fragments of the necessary functionality required for the SH .
HARDWARE AND SOFTWARE USED
HARDWARE REQUIREMENTS

1. Processor Type : i7

2. RAM : 4GB

3. Hard Disk Capacity : 1TB


SOFTWARE REQUIREMENTS

•1. Operating System : Microsoft windows XP

• 2. Programming Language : Visual Studio .NET 2008.


EXISTING SYSTEM:

 At present there are no systems which works according to the signals transmitted from the hand.

 Intel is now developing a game which we can play just by sitting and thinking of the moves but it
is not completed still.

 Blind cannot use the latest technologies such as touch screen.


PROPOSED SYSTEM:

 In the proposed system, we use hand signals for control the robotic actions.

 Our hand creates a certain electromagnetic signature corresponding to our thoughts.

 We can measure these EEG signals with an electroencephalogram.

 These EEG waves are commonly called Hand waves, and are classified into five types(delta,
theta, alpha, beta and gamma) based on the speed of their on-off activity.
MODULES:

 User Authentication

 Adding activity and Removing activity

 Edge Detection

 Feature extraction

 Enrolment Module

 Array matching Module


DIAGRAM
SFD/DFD
SYSTEM ARCHITECTURE
SYSTEM FLOW DIAGRAM:
DATA FLOW DIAGRAM:
Level 0

Input image Enrollment of Image Conversion


the image Using Hmm

Storing the Enrolled Images In The Database


For Further Proceedings
Level 1
COMPARISON OF
INPUT TO THE
DATABASE

Input Image IMAGE ACCURACY


CONVERSION OF MATCH
USING HMM IN %

RESULT
Use case model
MODULES DESCRIPTION
User Authentication:

• In this Module ,To be authenticated first have to be registered.

login:

• The Registered User Can be Allowed to view inner details for which Permitted.
Adding activity and Removing activity:

• You can manage the list of activities according to your needs. You can add the activities you need,
and remove them later, if necessary. By default, only walking and running are present in the
activity list when you switch on the time.
Edge Detection

• Edge detection is the name for a set of mathematical methods which aim at identifying points in a
digital image at which the image brightness changes sharply or, more formally, has discontinuities.
The points at which image brightness changes sharply are typically organized into a set of curved
line segments termed edges. The same problem of finding discontinuities in 1D signals is known
as step detection and the problem of finding signal discontinuities over time is known as
change detection. Edge detection is a fundamental tool in image processing, machine vision and
computer vision, particularly in the areas of feature detection and feature extraction
Feature Extraction

• In machine learning, pattern recognition and in image processing, feature extraction starts from an
initial set of measured data and builds derived values (features) intended to be informative, non-
redundant, facilitating the subsequent learning and generalization steps, in some cases leading to
better human interpretations. Feature extraction is related to dimensionality reduction.
Enrolment Module

• With the image analysis phase of the Hand Signal images are processed and enrolled with respect
to the person it belongs to. The further process becomes easy after enrolment of the Hand Signal
Image. The image analysis is the classification of Hand Signal with respect to the user’s database.
Array matching Module (Image Comparison and Verification)

• The comparison process is successfully completed by the help of array matching algorithm. The
algorithm aims at getting the values of the input image and the database image in two separate
multi-Dimensional arrays. Then each value of both arrays is verified to get the accuracy of
matching the Hand Signal
SAMPLE SCREENS
Starting window
Registration form
login
Main
CONCLUSION AND FUTURE SCOPE

• In this paper, we proposed a knowledge-driven approach to activity recognition based on ontological modeling and
semantic reasoning. We have analyzed the nature and characteristics of ADLs upon which we argue the necessity of
domain knowledge for pattern recognition using multisource sensor data. Following knowledge engineering
practices, we have developed context and ADL ontologies for the SH. We have conceived and designed an agent-
based integrated system architecture to illustrate the realization of the proposed approach. The compelling feature of
the system is the unified ontological modeling and representation for both sensor data and activities, which not only
facilitates domain knowledge reuse but also allows the exploitation of semantic reasoning for activity recognition. In
particular, we developed a novel activity recognition algorithm that allows continuous activity recognition at multiple
levels of abstraction, thus enabling course-grained and fine-grained activity assistance. The mechanism for evolving
initial ontological activity models through learning provides a way of marrying the strengths of traditional datadriven
approaches with knowledge-driven practices, making our approach flexible, applicable, and scalable in terms of rapid
system development and deployment.
REFERENCE
• [1] M. Chan, D. Este`ve, and C. Escriba, “A Review of Smart Homes—Present State and Future Challenges,” Computer
Methods and Programs in Biomedicine, vol. 91, no. 1, pp. 55-81, 2008.

• [2] D. Cook, H. Hagras, V. Callaghan, and A. Helal, “Making OurEnvironments Intelligent,” J. Pervasive and Mobile
Computing, vol. 5, pp. 556-557, 2009.

• [3] S. Helal, W. Mann, H. El-Zabadani, J. King, Y. Kaddoura, and E. Jansen, “The Gator Tech Smart House: A
Programmable Pervasive Space,” Computer, vol. 38, no. 3, pp. 50-60, Mar. 2005.

• [4] D.J. Cook and P. Rashidi, “Keeping the Resident in the Loop: Adapting the Smart Home to the User,” IEEE Trans.
Systems, Man, and Cybernetics J., Part A, vol. 39, no. 5, pp. 949-959, Sept. 2009.
THANKYOU

You might also like