Académique Documents
Professionnel Documents
Culture Documents
Geralyn P. Masuhay
Allvel P. Gorne
A Capstone Project
AUGUST 2019
CHAPTER 1
INTRODUCTION
1.1 Introduction
marketplace, today pre-schools students are being educated. Largely, for jobs
that have not yet been invented. Technology is being created and updated at a
frenetic pace, and growing more pervasive and useful with each stride.
future for our youth, but we can build our youth for the future. As schools face
dwindling resources and higher stakes than ever, does mobile technology hold
students. Mobile Technology offers plethora features and benefits that enables
new technology.
Guide based on the Republic Act 8980 known as the Early Childhood Care
system for early childhood care and development, providing funds therefore
and for the other purposes. As children grow into early childhood, their world
will begin to open up. They will become more independent and begin to focus
more on adults and children outside of the family. They will want to explore and
ask about the things around them even more. Preschool curricula take into
account the length of the preschool day. Many preschools run for only a few
hours a day, while others (especially in public school settings) run for much
longer days. Some even run longer than a typical school day to cover all the
core products of content, technology, and services. The emphasis is upon the
importance of integrated learning paths that vary according to the student, the
targets, offering the learner a resource that is seen as appealing, valuable and
The goal of object detection is to detect all instances of objects from a known
class, such as people, cars or faces in an image. Typically, only a small number
of instances of the object are present in the image, but there is a very large
number of possible locations and scales at which they can occur and that need
learning.This study aimed to provide a better way to teach the young ones to
identify the basic objects that they need to recognize and in this way the
sensitive for their future academic and social life. Based on the survey
the proponent conducted all of the teacher said that the students in
preschool are very active in object identifying. The teacher also uses
foster the knowledge and the experiences for this age and the support
last decade (2005-2015), which focus on the skills that are explored in
android phone that has many applications or games that installed to it. It
age 3-5 years old in preschool, it will catch their attention quickly and the
teacher can use that as the perfect time on teaching the things that they
have to learn and the things they have to familiarize. There are few steps
to use the application, the user will simply face the camera to the object
around the classroom or study room and it will automatically recognize
the object it detected and the application will read the label of the object
children at the age of 3-5 years old who are on the stage of learning, by this
they can easily familiarized the object they encounter inside their study room or
classroom. The application can assist the teachers in educating the children in
a current way.
General Objective
students.
Specific Objective
around the classroom based on the theme or the quarter of the class
need to recognize.
o Functionality;
o Reliability;
o Efficiency and;
o Usability.
1.5.1 Scope
The application can detect multiple objects within an image and read the
1.5.2 Limitation
recognize it.
1. https://www.researchgate.net/publication/309446775_Mobile_Learning_For_Presc
hool_Education#targetText=Mobile%20Learning%20For%20Preschool%20Education
&targetText=Nowadays%20preschool%20education%20is%20considered,future%20
academic%20and%20social%20life.
2. https://www.verywellfamily.com/all-about-preschool-curriculum-2764977
3. https://www.cdc.gov/ncbddd/childdevelopment/positiveparenting/preschoolers.ht
ml
CHAPTER 2
chapter helps in familiarizing information that was relevant and similar to the
present study.
(2018)
Detecting human skin color is a very difficult function and these are the
researchers in this field tried to resolve those problems by using thresholds that
implementation, it does not provide sufficient information for recognizing all skin
tones of humans.
to improve the accuracy of human skin detection. Three experiments have been
(UCI) machine learning repository, RGB (Red, Green, Blue), HSV (Hue,
This said study was similar to the current study in the way that they
are detecting human skin color. The system of the said study also shows that
the proposed system can achieve preferable accuracy in skin detection 98%,
97% and 97% using RGB dataset, HSV dataset and YCbCr dataset
respectively.
associated with that image. Usually Image Processing system includes treating
Object Detection Using Image Processing and the current study of the
proponents are just the same for it talks about Object Detection. The system of
the said study can also analyze and manipulating the image which includes
data compression and image enhancement and spotting patterns that are not
2.1.3 A real-time system for object detection and location reminding with
system that can automatically capture the last scene where the user-defined
input and has high detection rate in messy indoor environments. Additionally,
we build a user-friendly using flow on object online learning and detection which
Chung-Yu Chi, Szu-Lu Hsu, Liang-Gee Chen, “An integrated system for object
tracking detection and online learning with real-time RGB-D video”, Acoustics
time detection that also detects object and process or identities through the
object that has a high percentage of having similar characteristics with accurate
The study presents and evaluate a system, which allows a mobile robot
Markus Vincze (2017) While other systems have demonstrated one of these
elements, to our knowledge, they present the first system, which is capable of
doing all of these things, all without human interaction, in normal indoor scenes.
The system detects objects to learn by modeling the static part of the
view plan around a dynamic element to gather additional views for learning.
This said study was similar to the current study in the way that it also
talks about detecting and recognizing objects. The system of the said study can
identify the real-world object that mobile device's camera aim at. In this paper,
they present a location-based object identification algorithm that has been used
setting.
The said system was similar to the current study based on the
algorithm, discuss issues in using the algorithm, and present the algorithm
to Loc N. Huynh, Youngki Lee, Rajesh Krishna Balan such applications often
high accuracy, use of deep learning algorithms in mobile devices raises critical
This said study was similar to the current study in the way that it also
talks about Mobile Learning. The system of the said study can also classify an
Galaxy S7, taking an important step towards continuous vision without imposing
2.1.7 Real Time Object Scanning Using a Mobile Phone and Cloud-based
access to visual information in the world around them, but their efficacy is
capturing the information necessary for these applications to work because they
cannot see what they are taking a picture of. According to Yu Zhong, Pierre J.
Garrigues, Jeffrey P. Bigham (2017) The study presents Scan Search, a mobile
application that offers a new way for blind people to take high-quality photos to
identifies the most significant object inside the picture. This way, blind users
can scan for objects of interest and hear potential results in real time. The study
also present a study exploring the tradeoffs in how many photos are sent, and
conduct a user study with 8 blind participants that compare Scan Search with
a standard photo-snapping interface. Our results show that Scan Search allows
mobile learning for the 21st century. With AR, any real-world environment can
devices such as smartphones and tablets is also introducing AR into the mobile
domain.
different learning experience for learners. This AR mobile application will not
only provide learners with supplementary information but will also assist
science at the tertiary level that are at times difficult for learners to understand
developed, the learners are able to see what is happening and experience a
different form of learning where the focus is more on “learning by doing” and on
the ability of visualizing the complete set of steps involved for a particular
Nirmit V. Desai, Mooi Choo Chuah (2016) the study shows that recognition
for adaptation from in-situ mobile photos using two major steps: (i) a distributed
algorithm which identifies qualifying images stored in each mobile device for
training, (ii) a user labeling process for recognizable objects identified from
model.
Using a newly collected dataset with smartphone images collected from
different locations, time of a day, and device types, we show that DeepCham
CNN have reduced the running time of these detection networks, exposing
with the detection network, thus enabling nearly cost-free region proposals. An
and objectness scores at each position. RPNs are trained end to end to
generate high-quality region proposals, which are used by Fast R-CNN for
detection. With a simple alternating optimization, RPN and Fast R-CNN can be
trained to share convolutional features. For the very deep VGG-16 model, our
detection system has a frame rate of 5fps (including all steps) on a GPU, while
(73.2% mAP) and 2012 (70.4% mAP) using 300 proposals per image. Code is
available at https://github.com/ShaoqingRen/faster_rcnn.
This study is related to the proponents study due to the fact that it also
application. The difference is that it has an added feature that is the region
proposal networks.
been made in the field of object detection. In this article, the deep learning
algorithm is applied to the detection of daily objects, and some progress has
methods, the daily objects detection method based on deep learning is faster
and more accurate. The main research work of this article: 1. Collect a small
data set of daily objects; 2. In the Tensor flow framework to build different
models of object detection, and use this data set training model; 3. the training
process and effect of the model are improved by fine-tuning the model
Neural Network.
framework for machine learning and has the capability to enhance the clarity of
detecting a particular object in real-time. With the new research about deep
neural network, advance algorithm may also adapted by the proponents study
2.2.3 You Only Look Once: Unified, Real-Time Object Detection (2015)
Divvala, Ross Girshick, Ali Farhadi (2015) YOLO Prior work on object detection
class probabilities. A single neural network predicts bounding boxes and class
probabilities directly from full images in one evaluation. Since the whole
on detection performance.
the network, Fast YOLO, processes an astounding 155 frames per second
This study has a similarity to the current study in the way that it talks
about processes images in real-time. The system of the said study also makes
more localization errors but is far less likely to predict false detections where
experts and instructors are crucial to ensure its validity, value and sustainability.
for CAML is developed based on meaningful learning theory using the analytic
different CAML learning activities are tested, and some experts are invited to
evaluate and compare them. As a result, the strengths and weaknesses of each
learning activity are obtained. With the aid of the evaluation model proposed in
this work, CAML developers can progressively enhance the value of learning
This study has a similarity to the current study in the way that it talks
about mobile learning fields. The system of the said study also makes more
popular issue in educational research. It can also verify the effectiveness of the
model.
2.2.5 Object recognition and detection with deep learning for autonomous
different object detection algorithms have been proposed, not all are robust
Aysegul Ucar, Yakup Demir, Cuneyt Guzelis (2017) the study propose a novel
Networks (CNNs) and Support Vector Machines (SVMs) due to their powerful
The proposed system has the whole image into local regions and employ
multiple CNNs to learn local object features. The system select discriminative
SVMs applying both empirical and structural risk minimization instead of using
Finally, we fuse SVM outputs. In addition, the system pre-trained AlexNet and
Object detection and tracking is one of the critical areas of research due
smoothen the video sequence. On the other hand, few methods use the prior
available information about object shape, color, texture and so on. Tracking
and analyzed in this research. The goal of this paper is to analyze and review
the previous approach towards object tracking and detection using video
This study has a similarity to the current study in the way that it talks
about tracking and object detection. The system of the said study also
identify the gap and suggest a new approach to improve the tracking of
The use of digital technology in the learning process and teaching practices
Gabriel Pedro, Claudia Marina Monica de Oliveira Barbosa and Carlos Manuel
das Neves (2018) Although several studies value the importance of
technologies in our media-enriched world and the learn anytime and anywhere
motto associated with mobile learning, the argue that the classroom dynamics
are becoming more and more at risk with the addictive dimension brought about
by the ubiquitous presence of digital devices and social media in students’ lives.
The study of the said system make a critical review of the literature
learning facilitator. Finally, the system discuss the use of technology in the
learning process.
for machines object detection is an issue. Thus, Neural Networks have been
also called as ‘Artificial Neural Networks’ [13]. Artificial Neural Networks are
Neural Networks such as ANN, KNN, FASTER R-CNN, 3D-CNN, RNN etc. with
their accuracies. From the study of various research papers, the accuracies of
different Neural Networks are discussed and compared and it can be concluded
that in the given test cases, the ANN gives the best accuracy for the object
detection.
This study has a similarity to the current study in the way that it talks
about object detection. The system of the said study have an Artificial Neural
a human brain works. Rather than using a digital model, in which all
neurons.
Learning (2018)
With the advancement in deep learning in the past few years, the
system are able to create complex machine learning models for detecting
viability of using deep learning models for object detection in real-time video
algorithms.
This study has a similarity to the current study in the way that it talks
about Mobile Object Detection. The system of the said study shows a
algorithms with the use of transfer learning on neural networks adapted for
mobile use.
Learning (2018)
its localization and classification. It has wide spread applications and is a critical
and Kuntal Dey (2018) the system seeks to perform a rigorous survey of
modern object detection algorithms that use deep learning. As part of the
speed/size tradeoffs and training methodologies. This paper focuses on the two
types of object detection algorithms- the SSD class of single step detectors and
the Faster R-CNN class of two step detectors. Ultimately, a rigorous review of
the strengths and weaknesses of each detector leads us to the present state of
the art.
This study has a similarity to the current study in the way that it talks
Modern Object Detection. The system of the said study have a techniques to
construct detectors that are portable and fast on low powered devices are also
CHAPTER 3
the data for user input, the process of the system, the output and also the
Project
eTudlo: A
Mobile
Learning Object
School in
Object
Identification
using Open
CV
chair Description
Attribute
The figure 2 shows how the users access the application through using a mobile
phone.
used.
library.
user groups:
R Read
S Scan
3.2.3.1 Reliability
3.2.3.2 Operability
3.2.3.3 Maintainability
3.2.3.4 Scalability
The application has only 1 user types.
3.2.3.5 Availability
3.2.3.6 Delivery
Hardware Specification
Android Phone 2 GB
Windows OS 7, 8 or 10
Figure 3. The phase of the System Development Life Cycle RAD (Rapid
Application Development)
modules are developed in parallel as prototypes and are integrated to make the
the application, then the designing and program coding is the phase which is
the application’s front-end and back-end will be develop, after the designing
and program coding is the testing which the application will be tested if it meets
gather enough data from the respondents and make notable information that is
discussed to the respondent for verification purpose. The analysis that the
3.4.2 Design
Object
using Open CV
Object Information
In the figure above, there is only one user. The user needs to scan
the object to identify and analyze by the application which is the eTudlo: A
Mobile Leaning Tool for Pre-school in Object identification using Open CV and
In the figure above, shows what the general movement is and the
way in which the application behaves when it is used. The first process after
the application has been opened; the object such as chair, board, table and etc.
automatically.
3.4.3 Implementation
During the implementation, the app function were develop. The syntaxes
and the queries needed to our app were created and build through the aid of
3.4.4 Testing
After the implementation phase, proponents will now recognize the result
predictable result. Please see section 3.9 for the Verification, Validation and
Testing Plans.
3.4.5 Maintenance
The last phase is the maintenance. This phase can be work on after the
completion of the project. The proponents decided to monitor the system every
month. These are the types of maintenance that the proponents would possibly
go through.
environment.
3.5 Software Development Tools
android platforms.
Activities 1 2 3 1 2 3 4 1 2 3 4 1 2 3 4
Planning
Designing
Coding
Testing
Maintenance
The figure shows the timeline for the process and development of the
project.
Team
Name
or group of aligned
assigned to program
other functionality of
the system.
Diane Researcher/ Behave honestly and Analysts/
technique to write
application codes.
project deliverable
meet business
requirements.
2 Months of
Manager
System/Analysis 18,000 36,000
Total P21,738
3.8.3 Facilities
Development)
Total P198,600.00
Table 8 shows the facility cost that the proponent has been used during the
development.
3.9 Verification, Validation and Testing Plan
References
Available: https://doi.org/10.1108/EUM0000000005485
https://www.jstor.org/stable/jeductechsoci.17.4.352
https://www.jstor.org/stable/jeductechsoci.12.2.163
Language Learning & Technology s October 2013, Volume 17, Number
3 pp. 52–73
1. 1.
Talukder, Z., Basak, A., Shoyaib, M.: Human skin detection. J. Comput. Sci.
Show Context View Article Full Text: PDF (2020KB) Google Scholar
https://search.proquest.com
https://search.proquest.com/openview/a348827d4fc6aaf650f26a36d2e
92e28/1?pq-origsite=gscholar&cbl=18750&diss=y
object recognition and tracking", Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst.
Show Context View Article Full Text: PDF (4142KB) Google Scholar
Tan, Qing; Chang, William; Kinshuk
Pages 82-95
Article No. 20
2 doi>10.1145/2513383.2513443
146. https://doi.org/10.1108/ITSE-07-2015-0017
[online] Available:
https://github.com/BVLC/caffe/tree/master/models/bvic_reference_caffen
et/.
[online] Available:
https://github.com/BVLC/caffe/tree/master/models/bvic_reference_caffen
et/.
https://doi.org/10.1111/bjet.12147
, http://www.mobileye.com/en-uk/mobileye-features/pedestrian-collision-
Google Scholar
Research Scholar, Department of Electronics & Communication,