28 September

Coimbatore
From
Mr. J.SAM ASHBAL RAJ
II M. Sc., Computer Science
Department of Software Systems
Sri Krishna Arts and Science College
Kuniamuthur, Coimbatore – 641 008, TN.
To
The MEMBER SECRETARY,
Tamilnadu State Council for Science and Technology,
DOTE Campus, Chennai - 600 025
Through
The PRINCIPAL
Sri Krishna Arts and Science College
Kuniamuthur, Coimbatore – 641 008, TN.

Respected Sir,

Sub.: Submission of student project proposal – reg.

Herewith, I submit two copies of my proposal under Student project scheme and the

proposal entitled “Mobile Object Detection using TensorFlow Lite” for your consideration

Thanking You

Yours sincerely

(J. SAM ASHBAL RAJ)

 A Research Proposal Submitted under the
STUDENT PROJECTS SCHEME to
TAMILNADU STATE COUNCIL FOR SCIENCE AND TECHNOLOGY

Entitled

Mobile Object Detection using TensorFlow Lite

By

Mr. J. Sam Ashbal Raj,

II M. Sc., Computer Science
Department of Software Systems
Sri Krishna Arts and Science College
Coimbatore – 641 008, Tamil Nadu

Affiliated to
Bharathiar University, Coimbatore

September 2023
 STUDENT PROJECT PROPOSAL submitted to TNSCST

1. Name of the Student (s) one : Mr. Sam Ashbal Raj J;

email ID samashbalrajj22mcs043@skasc.ac.in

2. Name of the Guide : Dr. B. Rajesh Kumar

Designation : Assistant Professor

Institutional Address : Sri Krishna Arts and Science College, Kuniamuthur,

Coimbatore – 641 008, Tamil Nadu

Phone No. & Mobile No : +91 6369409591; rajeshkumarb@skasc.ac.in

3. Project Title : Mobile Object Detection using TensorFlow Lite

4. Sector in which your Project : Computer Science and Engineering

proposal is to be considered

5. Project Details : Project proposal have been attached as Annexure I

6. Has a similar project been : No

carried out in your college /
elsewhere? If so furnish
details of the previous project
and highlight the
improvements suggested in
the present one
 Annexure I

Mobile Object Detection using TensorFlow Lite

1. INTRODUCTION:

Object detection has become an important research area due to its various practical
applications. In this paper, we propose a deep learning-based approach for mobile object
detection using TensorFlow Lite. We use a deep neural network with multiple convolutional
layers to extract features from the input image. The extracted features are then fed to a set of
detection heads to predict the bounding boxes and corresponding class labels of the objects in
the image. We use the MobileNetV2 architecture as our backbone network, and the SSD
(Single Shot Detector) algorithm as our detection framework. Our approach achieves high
accuracy and real-time performance on mobile devices. We evaluate our approach on several
popular object detection datasets, and demonstrate that it outperforms existing state-of-the-art
approaches. Our approach can be used in a wide range of applications, such as autonomous
driving, surveillance, and augmented reality.

2. AIM AND OBJECTIVES

Aim

To establish and maintain the Mobile Object Detection using TensorFlow Lite

Objectives

• Implement a deep neural network with multiple convolutional layers to extract features
from input images.
• Develop a set of detection heads to predict bounding boxes and corresponding class
labels of objects within images.
• Utilize TensorFlow Lite for model deployment on mobile devices.
• Achieve a balance between high accuracy and real-time performance, optimizing for
mobile hardware constraints.
3. METHODOLOGY

Deep Learning Model:

You're using a deep neural network with multiple convolutional layers. This suggests that
you're leveraging the power of deep learning to automatically learn and extract features from
input images, which is crucial for accurate object detection.

Feature Extraction:

Convolutional layers are employed to extract relevant features from the input image. Deep
learning models, especially convolutional neural networks (CNNs), have proven to be highly
effective in feature extraction tasks in computer vision.

Detection Framework:

You've chosen SSD (Single Shot Detector) as your detection framework. SSD is known for its
efficiency and ability to detect objects in real-time, making it suitable for mobile applications.

Backbone Network:

MobileNetV2 is used as the backbone network. MobileNetV2 is optimized for mobile devices,
balancing between accuracy and computational efficiency, making it ideal for real-time
applications on mobile platforms.

4. OUTLINE OF THE WORK:

Data integrity proofs in cloud storage encompass a multi-step process to ensure the security
and reliability of data stored remotely. Initially, data is divided into segments or blocks, and
cryptographic hash functions generate unique fingerprints for each block. These hash values
are stored alongside the data. Encryption is applied to protect data at rest and in transit.
Redundancy mechanisms, like replication or erasure coding, are implemented for fault
tolerance. Regular verification checks, challenge-response mechanisms, and incident response
strategies are used to maintain data integrity. Compliance requirements are met, and automated
alerts and monitoring ensure proactive security. This holistic process safeguards data against
unauthorized access, tampering, or loss while enhancing trust in cloud storage services.
 Data integrity proofs in cloud storage involve a multi-step process to ensure the security and
reliability of data stored remotely. Initially, data is segmented into blocks and subjected to
cryptographic hashing for unique identification. These blocks are encrypted to protect against
unauthorized access. Redundancy mechanisms, like replication or erasure coding, are
employed to mitigate data loss risks. Metadata management keeps track of data's integrity and
location. Regular integrity checks or on-demand verification are performed using
cryptographic methods, and incident response plans are in place for any breaches. Compliance
with relevant regulations is ensured, and reporting mechanisms are established. This holistic
approach guarantees that data remains unaltered, secure, and available in the cloud, enhancing
data reliability and user confidence.
REFERENCES

1. Ahmad, I., Ullah, I., Khan, W. U., Ur Rehman, A., Adrees, M. S., Saleem, M. Q., et al.
(2021). Efficient algorithms for E-healthcare to solve multi object fuse detection problem. J.
Healthc. Engin. 2021:9500304.

2. Ahmad, S., Ullah, T., Ahmad, I., Al-Sharabi, A., Ullah, K., Khan, R. A., et al. (2022).
A novel hybrid deep learning model for metastatic cancer detection. Comput. Intell. Neurosci.
2022:8141530. doi: 10.1155/2022/8141530

3. Asadi, K., Ramshankar, H., Pullagurla, H., Bhandare, A., Shanbhag, S., Mehta, ., et al.
(2018). Building an integrated mobile robotic system for real-time applications in construction.
arXiv (preprint) arXiv:1803.01745. doi: 10.3390/ s131217222

4. Bian, X., Chen, Y., Wang, S., Cheng, F., and Cao, H. (2021). “Medical Waste
Classification System Based on OpenCV and SSD-MobileNet for 5G,” in 2021 IEEE wireless
communications and networking conference workshops (WCNCW), (Nanjing: IEEE), 1–6.
doi: 10.1109/WCNCW49093.2021.9420036

5. Biswas, D., Su, H., Wang, C., and Stevanovic,

A. (2019). Speed estimation of multiple moving objects from a moving UAV platform. ISPRS
Int. J. Geo-Inf. 8:259. doi: 10.3390/ijgi8060259

6. Chandan, G., Jain, A., and Jain, H. (2018). “Real time object detection and tracking
using Deep Learning and OpenCV,” in 2018 international conference on inventive research in
computing applications (ICIRCA), (Coimbatore: IEEE), 1305–1308. doi:
10.1109/ICIRCA.2018.8597266

7. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2015). Region-based convolutional
networks for accurate object detection and segmentation. IEEE Trans. Pattern Anal. Machine
Intell. 38, 142–158. doi: 10.1109/TPAMI.2015.2437384

8. Han, K., Sun, M., Zhou, X., Zhang, G., Dang, H., and Liu, Z. (2017). “A new method
in wheel hub surface defect detection: Object detection algorithm based on deep learning,” in
2017 international conference on advanced mechatronic systems (ICAMechS), (Xiamen:
IEEE), 335–338.

9. Hoang Ngan Le, T., Zheng, Y., Zhu, C., Luu, K., and Savvides, M. (2016).“Multiple
scale faster-rcnn approach to driver’s cell-phone usage and hands on steering wheel detection,”
in Proceedings of the IEEE conference on computer vision and pattern recognition workshops,
(Las Vegas: IEEE), 46–53.
5. BUDGET REQUIRED

S. No Items Expenditure in Rs
1. Hardware and Software cost 1600
2. Training Object Detection Data sets 2000
3. Data Annotation Services 3000
4. Cloud Services 2000
5. Testing and Evaluation 1200
Total 9800

6. Any other information: NIL

CERTIFICATE

This is to certify that Mr. Sam Ashbal Raj J (22MCS043) is a bonafide final year
student of P.G. Science courses of our college and it is also certified that two copies of
utilization certificate and final report along with seminar paper will be sent to the Council after
completion of the project by the end of April 2024.

Signature of Guide Signature of the HOD Signature of the Principal/

Head of the Institution