International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

RESEARCH ARTICLE OPEN ACCESS

Enhance the Security for Internet of Things Information Using

Block Chain
Elayaraja P
Software Engineer-Chennai.

ABSTRACT
In recent decades, the Internet of Things (IoT) is transforming into an attractive system to drive a substantive hop on
stock and enterprises through physical, computerized, and social spaces. For enhancing the security level of IoT
multimedia data, a block-based security model is proposed. After forming the clusters, a hash function with
blockchain technique is utilized to the secure the IoT information. Initially, the data is changed over into a number of
blocks then applies a hash function to each block for an end to end blockchain model. Based on this function the
information’s are encrypted and decrypted stored in the cloud by an authenticated person. From the implementation
results of data security, the execution time and security levels are analyzed.
Keywords: Internet of things, security, optimization, clustering and blockchain

I. INTRODUCTION
Internet of Things (IoT) is a type of network that is the execution of computationally costly encryption
being utilized by wireless sensor associations and radio algorithms [36-38].
frequency identification (RFID) via network topology
[1] to accomplish high reliability in transmission as II. LITERATURE REVIEW
well as intelligent processing [2-10]. IOT comprises
three layers: the sensing layer, transport layer, and A blockchain is a database that stores every processed
application layer. IOT has made a tremendous change transaction – or information – in the sequential request,
in different areas such as business, agriculture, in an arrangement of PC recollections that are carefully
pharmacy, also in nuclear reactors [11-16]. To receive designed to foes. These exchanges are then shared by
the IoT innovation it is important to assemble the all users by DanielMinoli et al in 2018 [39], More
certainty among the clients about its security and significantly, we talk about, how blockchain, which is
privacy that it won't make any risk to their data the basic innovation for bitcoin, can be a key
integrity and authority [17-20]. In secure systems, the empowering agent to tackle numerous IoT security
secrecy of the information is kept up and it is ensured issues by Minhaj Ahmad Khan and Khaled Salah in
that at the processing of message exchange the 2018 [40]. It's additionally distinguished open research
information holds its inventiveness and no modification issues and difficulties for IoT security. One of the real
by the system [21-23]. In spite of the fact that, the IoT issues of a clustering protocol is choosing an optimal
can encourage the digitization of the data itself, the group of sensor nodes as the group heads to isolate the
dependability of such data is as yet a key challenge by network by Khalil Bennani et al. 2012 [41-42]. In any
Bitcoin [6]. It's upheld by a protocol that points of case, optimum clustering is an NP-Hard issue and
interest the infrastructure responsible for guaranteeing solving it includes searches through tremendous spaces
that the data remains unchanging after some time [24- of conceivable solutions. Two fundamental periods of
26]. Benefiting from blockchains power and versatility, optimization, exploration, and exploitation,, are
in this work, propose an effective decentralized structured by the social interaction of dragonflies in
verification system [27-30]. The principle motivation exploring, hunting foods, and keeping away from foes
behind network security data insurance is to accomplish while swarming powerfully or factually by Seyedali
secrecy as well as integrity. Security issues are of Mirjalili in 2015[43-45]. In Emanuel Ferreira Jesus et al
extraordinary significance in amplifying the size of in 2018 [46] the ideas about the structure and task of
network and gadgets [31-35]. Different cryptographic Blockchain and, mostly, investigate how the utilization
algorithms have been produced that tends to the said of this innovation can be utilized to give security and
issue, however, their use in IoT is questionable as the privacy in IoT.
equipment we deal in the IoT is not appropriate for

ISSN: 2347-8578 www.ijcstjournal.org Page 109

International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

III. SECURITY ISSUES IN IOT investigation of IoT situations, we focus on a genuine

MULTIMEDIA INFORMATION IoT smart home condition sent in our tested
concentrating on the interactions among the diverse IoT
Security approaches that depend greatly on encryption parts.
are not a solid match for these constrained gadgets
since they are not equipped for performing complex 4.2 IoT information Generation
encryption and decryption rapidly enough to have the The new invention of IoT and multimedia data
capacity to transmit information safely in a progressive applications is necessary to address explicit business
manner. Some security challenges in IoT security are solutions which require needs, for example, predictive
tag attack, Sybil attack, wormhole attack and, etc. maintenance, loss prevention, asset utilization,
Regardless, with traditional encryption procedures, inventory tracking, disaster planning, and recovery,
before dealing with some sensitive data from downtime minimization, energy usage optimization,
customers, the third party administration (cloud) would device performance effectiveness. These datasets are
decrypt this information and after that find that data particular in information structures, volume, get to
[47]. Data that is very sensitive to bank account details, procedures, and some extraordinary perspectives; they
usernames, passwords need to encrypt with at least two- can scarcely be stored and gotten as well, its shows in
factor authentication procedures to guarantee security. figure 1. IoT applications have quite certain qualities;
For enhancing the security in IoT data, BC is utilized. they produce extensive volumes of data and require
Now, the header turns out to be a piece of a network and power for significant time periods [57].
cryptographic riddle which must be comprehended by
the block chain's network of clients via a trial and error
procedure, from trillions of opportunities – before it is
included to the block chain [48--52].

IV. METHODOLOGY FOR IOT

INFORMATION SECURITY
The IoT visualizes a completely associated world,
where things can convey estimated information and
connect with one another. For enhancing the security
dimension of the multimedia data's in IoT, optimal
Block Chain (BC) security model is utilized, Its behinds
the bit coin concept a permanent open record of data
secured by a network of distributed members, its
strategy that to enables exchanges to be confirmed by a
gathering of untrustworthy on-screen characters. The
fundamental of this proposed strategy is, enhance the
secrecy as well as the reliability of the IoT data,
Moreover, this BC, every block contains the number of Fig 1: IoT Components
transactions. CH has coordinate proof about CH in the
event that it confirmed a block created by different Pre-processing
squares of data [53-56] This underlying preprocessing IoT data are separated to
choose the sensitive or quality information for security,
4.1 Security and privacy Analysis in IoT the purpose behind preprocessing is diminishing the
computational time. Moreover, the goal is to ensure
Security in IoT is difficult because of low asset abilities successful security and improve the performance of the
of the vast majority of devices, huge scale, model, we secure sensitive data only. IoT data
heterogeneity among the gadgets, and absence of clustering using optimization Information is clustered
standardization. Besides, a considerable lot of these IoT by utilizing irregular clustering model with
gadgets gather and offer a lot of information from our optimization. Each data is assessed by the weights of
own spaces, in this way opening up noteworthy privacy its cluster associations. Finally, among various nodes
concerns. Security and privacy risk analysis for a and as indicated by their weights a node is elected as
commonplace shrewd home engineering that depends the cluster head, for optimal cluster head selection
on existing and promptly accessible market IoT gadgets dragonfly optimization is proposed. Finally, among
and stages. As opposed to existing security and threat

ISSN: 2347-8578 www.ijcstjournal.org Page 110

International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

various nodes and as shown by their information a node accuracy by a gathering of creative network nodes
is elected as the cluster head. called "Miners". Specifically, rather than mining a
individuals to that of different individuals in the area single transaction, the miners package various
and a1 , a 2 , a3 , a 4 and a5 coefficient parameters. transactions that are waiting for the network to get
processed in a single unit called “block".
After the alignment procedure of individuals among the
dragonflies, cohesion process is performed which Encryption
implies the inclination of individuals towards the focal
point of the mass of the area. The new position Encryption system function as given a message and a
refreshed by utilizing the following condition key, it creates a ciphered message to be transmitted
Ct +1 = Ct + Levy(dis tan ce) * Ct (1) over unprotected channels, without any risk being
Food source and the enemy is selected over best and the comprehended by other people who don't have the
most exceedingly terrible solutions obtained in the decryption key. For the security purpose, the key
whole swarm at any minute. In light of above process generation dependent on the two sets, one public and
select optimal cluster head for IoT data clustering one private. The first to encrypt and the second to
model. After that, the clustered data are secured by decrypt and the vice versa; this is conceivable because
utilizing the BC technique. of the utilization of some numerical functions that have
the property of being irreversible.
For encrypted and decrypted information
Encryption Decryption Memory Execution
File
size size (byte) Time Enc = hash (inf o group, hash, publickey, IP)
size (2)
(ms)
10 23 10 1242488 94523 Each block of data encrypted by
E  m | mod inf o | (3)
k
20 34 20 374528 105481

30 44 30 312458 98450 V. RESULTS AND ANALYSIS

40 49 40 412141 112345 Our proposed IoT information security model
50 56 50 423412 112482 is executed in the Java programming language with the
JDK 1.7.0 in a windows machine enclosing the
4.2 Security model: A Block chain Model configurations such as the Intel (R) Core i3 processor,
1.6 GHz, 4 GB RAM, and the operating system
In a BC, each transaction in the set that contains a block platform is Microsoft Window7 Professional. This
is hashed to produce a hash value. Hashes are combined proposed security analysis is compared with other
into a Merkle Tree. Generally, the BC indicates a techniques.
consistently maintained and controlled database
considering developing variables and gathered Table 1: Proposed (blockchain) Security analysis results
information test sets. The key components of BC are a
member made transactions and the recorder blocks of Table 1 shows the result of proposed parameters which
such exchanges. Here, the block checks whether obtains in the study. Depends on file size, we find
transaction details were sustained in the correct encryption size, decryption size, memory and execution
grouping or not. This does not permit any altering of time. The result depicts that encryption size and
the data accessible. decryption increases if the file size increased, the
Bitcoin execution time also increased. But compared to other
Bitcoin is cryptographic money and a digital payment techniques proposed model secure the IoT data in a
system, in view of a public BC, each block of the high manner.
Bitcoin blockchain. In Bitcoin, transactions are
processed to check their integrity, authenticity, and

ISSN: 2347-8578 www.ijcstjournal.org Page 111

International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

Fig 4: Number of blocks Vs hash value

Fig 5: Security Level comparative analysis

Figure 4 shows the graph of security level based on a VI. CONCLUSION

number of blocks. Here, we compare the security In this chapter, the IoT multimedia information’s
result with the proposed method to existing security model with help hash function based
techniques such as bitcoin and ECC. The graph blockchain was discussed. Thusly, the advantages of
depicts that block chain reaches optimal security in applying BC to the IoT ought to be examined
the range of 82 to 91.23% compared to the other two. precisely and taken with caution. Also, this chapter
Figure 5 shows the security level based on the provided an analysis of the main difficulties that
database size. The hash blockchain performs an blockchain and IoT must address in order for them to
optimal security level for every database size. The effectively cooperate. This blockchain technology
security level reaches a maximum at 90% in hash can help to improve IoT applications and also this
block chain function.

ISSN: 2347-8578 www.ijcstjournal.org Page 112

International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

data clustering cluster head selection by DOA, its method for medical image watermarking.
give better accuracy. However, it is still in the In International Symposium on Computing,
beginning periods of creating block chains, and these Communication, and Control (ISCCC
obstructions will be defeated, opening the best 2009) (Vol. 1, pp. 31-34).
approach to numerous potential outcomes. One of the [10] Sampson, J., & Velmurugan, S. P. (2020,
principle concerns about blockchain, and especially March). Analysis of GAA SNTFT with
crypto currencies, resides in its volatility which has Different Dielectric Materials. In 2020 5th
also been exploited by individuals to take unfair International Conference on Devices, Circuits
advantage of this situation. The incorporation of the and Systems (ICDCS) (pp. 283-285). IEEE.
IoT and blockchain will extraordinarily increase the [11] Elhoseny, M., Bian, G. B., Lakshmanaprabu,
security level. S. K., Shankar, K., Singh, A. K., & Wu, W.
(2019). Effective features to classify ovarian
REFERENCE cancer data in internet of medical
things. Computer Networks, 159, 147-156.
[1] Chen, Y. T., Chen, C. H., Wu, S., & Lo, C. C. [12] Gochhayat, S. P., Kaliyar, P., Conti, M.,
(2019). A two-step approach for classifying Tiwari, P., Prasath, V. B. S., Gupta, D., &
music genre on the strength of AHP weighted Khanna, A. (2019). LISA: Lightweight
musical features. Mathematics, 7(1), 19. context-aware IoT service architecture. Journal
[2] Elhoseny, M., Shankar, K., &Uthayakumar, J. of cleaner production, 212, 1345-1356.
(2019). Intelligent diagnostic prediction and [13] Dutta, A. K., Elhoseny, M., Dahiya, V., &
classification system for chronic kidney Shankar, K. (2020). An efficient hierarchical
disease. Scientific reports, 9(1), 1-14. clustering protocol for multihop Internet of
[3] Sivakumar, P., Velmurugan, S. P., & vehicles communication. Transactions on
Sampson, J. (2020). Implementation of Emerging Telecommunications
differential evolution algorithm to perform Technologies, 31(5), e3690.
image fusion for identifying brain tumor. [14] Anand Nayyar, Vikram Puri,
[4] Khamparia, A., Gupta, D., Nguyen, N. G., Nhu Gia Nguyen, BioSenHealth 1.0: A Novel
Khanna, A., Pandey, B., & Tiwari, P. (2019). Internet of Medical Things (IoMT) Based
Sound classification using convolutional Patient Health Monitoring System, Lecture
neural network and tensor deep stacking Notes in Networks and Systems. Springer,
network. IEEE Access, 7, 7717-7727. 2019
[5] Jansirani, A., Rajesh, R., Balasubramanian, R., [15] Shankar, K., Lakshmanaprabu, S. K., Khanna,
& Eswaran, P. (2011). Hi-tech authentication A., Tanwar, S., Rodrigues, J. J., & Roy, N. R.
for pslette images using digital signature and (2019). Alzheimer detection using Group Grey
data hiding. Int. Arab J. Inf. Technol., 8(2), Wolf Optimization based features with
117-123. convolutional classifier. Computers &
[6] Jain, R., Gupta, D., & Khanna, A. (2019). Electrical Engineering, 77, 230-243.
Usability feature optimization using MWOA. [16] Paramathma, M. K., Pravin, A. C., Rajarajan,
In International conference on innovative R., & Velmurugan, S. P. (2019, April).
computing and communications (pp. 453-462). Development and Implementation of Efficient
Springer, Singapore. Water and Energy Management System for
[7] Shankar, K., & Lakshmanaprabu, S. K. (2018). Indian Villages. In 2019 IEEE International
Optimal key based homomorphic encryption Conference on Intelligent Techniques in
for color image security aid of ant lion Control, Optimization and Signal Processing
optimization algorithm. International Journal (INCOS) (pp. 1-4). IEEE.
of Engineering & Technology, 7(9), 22-27. [17] Chen, C. H., Song, F., Hwang, F. J., & Wu, L.
[8] Lyu, L., & Chen, C. H. (2020, July). (2020). A probability density function
Differentially Private Knowledge Distillation generator based on neural networks. Physica
for Mobile Analytics. In Proceedings of the A: Statistical Mechanics and its
43rd International ACM SIGIR Conference on Applications, 541, 123344.
Research and Development in Information [18] Kathiresan, S., Sait, A. R. W., Gupta, D.,
Retrieval (pp. 1809-1812). Lakshmanaprabu, S. K., Khanna, A., &
[9] Poonkuntran, S., Rajesh, R. S., & Eswaran, P. Pandey, H. M. (2020). Automated detection
(2011). Analysis of difference expanding and classification of fundus diabetic

ISSN: 2347-8578 www.ijcstjournal.org Page 113

International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

retinopathy images using synergic deep classification of mammograms using SVM,

learning model. Pattern Recognition Letters. Advances in Intelligent Systems and
[19] Gupta, D., & Ahlawat, A. K. (2016). Usability Computing, Volume 672, 2018, Pages 787-
determination using multistage fuzzy 795.
system. Procedia Comput Sci, 78, 263-270. [29] Khamparia, A., Saini, G., Gupta, D., Khanna,
[20] Amira S. Ashour, Samsad Beagum, Nilanjan A., Tiwari, S., & de Albuquerque, V. H. C.
Dey, Ahmed S. Ashour, Dimitra Sifaki (2020). Seasonal crops disease prediction and
Pistolla, Gia Nhu Nguyen, Dac-Nhuong Le, classification using deep convolutional
Fuqian Shi (2018), Light Microscopy Image encoder network. Circuits, Systems, and
De-noising using Optimized LPA-ICI Filter, Signal Processing, 39(2), 818-836.
Neural Computing and Applications, [30] Uthayakumar, J., Elhoseny, M., & Shankar, K.
Vol.29(12), pp 1517–1533, Springer, ISSN: (2020). Highly Reliable and Low-Complexity
0941-0643. (SCIE IF 4.664, Q1) Image Compression Scheme Using
[21] Pan, M., Liu, Y., Cao, J., Li, Y., Li, C., & Neighborhood Correlation Sequence
Chen, C. H. (2020). Visual Recognition Based Algorithm in WSN. IEEE Transactions on
on Deep Learning for Navigation Mark Reliability.
Classification. IEEE Access, 8, 32767-32775. [31] Huyen, D.T.T., Binh, N.T., Tuan, T.M.,
[22] Chen, C. H., Hwang, F. J., & Kung, H. Y. Nguyen, G.N, Dey, N., Son, L.H, Analyzing
(2019). Travel time prediction system based trends in hospital-cost payments of patients
on data clustering for waste collection using ARIMA and GIS: Case study at the
vehicles. IEICE TRANSACTIONS on Hanoi Medical University Hospital, Vietnam,
Information and Systems, 102(7), 1374-1383. Journal of Medical Imaging and Health
[23] Shankar, K., & Eswaran, P. (2015). ECC Informatics, 7(2), pp. 421-429.
based image encryption scheme with aid of [32] Van, V.N., Chi, L.M., Long, N.Q., Nguyen,
optimization technique using differential G.N., Le, D.-N, A performance analysis of
evolution algorithm. Int J Appl Eng openstack open-source solution for IaaS cloud
Res, 10(55), 1841-5. computing, Advances in Intelligent Systems
[24] Anand Nayyar, Vikram Puri, and Computing, 380, pp. 141-150.
Nhu Gia Nguyen, Dac Nhuong Le, Smart [33] Shankar, K., & Eswaran, P. (2016, January). A
Surveillance Robot for the Real Time new k out of n secret image sharing scheme in
Monitoring and Control System in visual cryptography. In 2016 10th
Environment and Industrial Applications, International Conference on Intelligent
Advances in Intelligent System and Systems and Control (ISCO) (pp. 1-6). IEEE.
Computing, pp 229-243, Springer [34] Dey, N., Ashour, A.S., Chakraborty, S., Le,
[25] Le Nguyen Bao, Dac-Nhuong Le, Gia Nhu D.-N., Nguyen, G.N, Healthy and unhealthy
Nguyen, Vikrant Bhateja, Suresh Chandra rat hippocampus cells classification: A neural
Satapathy (2017), Optimizing Feature based automated system for Alzheimer disease
Selection in Video-based Recognition using classification, Journal of Advanced
Max-Min Ant System for the Online Video Microscopy Research, 11(1), pp. 1-10
Contextual Advertisement User-Oriented [35] Velmurugan, S. P., & Rajasekaran, P. S. M. P.
System, Journal of Computational Science, (2017). CLASSIFICATION OF BRAIN
Elsevier ISSN: 1877-7503. Vol.21, pp.361- TUMOR USING MULTIMODAL FUSED
370. (SCIE IF 2.502, Q1) IMAGES AND PNN. International Journal of
[26] Chakchai So-In, Tri Gia Nguyen, Gia Nhu Pure and Applied Mathematics, 115(6), 447-
Nguyen: Barrier Coverage Deployment 457.
Algorithms for Mobile Sensor Networks. [36] Shankar, K., Elhoseny, M., Perumal, E.,
Journal of Internet Technology 12/2017; Ilayaraja, M., & Kumar, K. S. (2019). An
18(7):1689-1699. Efficient Image Encryption Scheme Based on
[27] Le, D.-N.a, Kumar, R.b, Nguyen, G.N., Signcryption Technique with Adaptive
Chatterjee, J.M.d, Cloud Computing and Elephant Herding Optimization.
Virtualization, DOI: 10.1002/9781119488149, In Cybersecurity and Secure Information
Wiley. Systems (pp. 31-42). Springer, Cham.
[28] Bhateja, V., Gautam, A., Tiwari, A., Nhu,
N.G., Le, D.-N, Haralick features-based

ISSN: 2347-8578 www.ijcstjournal.org Page 114

International Journal of Computer Science Trends and Technology (IJCST) – Volume 8 Issue 4, Jul-Aug 2020

[37] Wu, L., Chen, C. H., & Zhang, Q. (2019). A [47] Parvathy, V. S., & Pothiraj, S. (2019). Multi-
mobile positioning method based on deep modality medical image fusion using
learning techniques. Electronics, 8(1), 59. hybridization of binary crow search
[38] Lydia, E. L., Kumar, P. K., Shankar, K., optimization. Health Care Management
Lakshmanaprabu, S. K., Vidhyavathi, R. M., Science, 1-9.
& Maseleno, A. (2020). Charismatic document [48] Velmurugan, S. P., Sivakumar, P., &
clustering through novel K-Means non- Rajasekaran, M. P. (2018). Multimodality
negative matrix factorization (KNMF) image fusion using centre-based genetic
algorithm using key phrase algorithm and fuzzy logic. International
extraction. International Journal of Parallel Journal of Biomedical Engineering and
Programming, 48(3), 496-514. Technology, 28(4), 322-348.
[39] Sujitha, B., Parvathy, V. S., Lydia, E. L., Rani, [49] Chen, C. H. (2018). An arrival time prediction
P., Polkowski, Z., & Shankar, K. (2020). method for bus system. IEEE Internet of
Optimal deep learning based image Things Journal, 5(5), 4231-4232.
compression technique for data transmission [50] Shankar, K., Perumal, E., & Vidhyavathi, R.
on industrial Internet of things M. (2020). Deep neural network with moth
applications. Transactions on Emerging search optimization algorithm based detection
Telecommunications Technologies, e3976. and classification of diabetic retinopathy
[40] Lo, C. L., Chen, C. H., Hu, J. L., Lo, K. R., & images. SN Applied Sciences, 2(4), 1-10.
Cho, H. J. (2019). A fuel-efficient route plan [51] Mohanty, S. N., Ramya, K. C., Rani, S. S.,
method based on game theory. Journal of Gupta, D., Shankar, K., Lakshmanaprabu, S.
Internet Technology, 20(3), 925-932. K., & Khanna, A. (2020). An efficient
[41] Kung, H. Y., Chen, C. H., Lin, M. H., & Wu, Lightweight integrated Blockchain (ELIB)
T. Y. (2019). Design of Seamless Handoff model for IoT security and privacy. Future
Control Based on Vehicular Streaming Generation Computer Systems, 102, 1027-
Communications. Journal of Internet 1037.
Technology, 20(7), 2083-2097. [52] Elhoseny, M., & Shankar, K. (2020). Energy
[42] Elhoseny, M., & Shankar, K. (2019). Reliable efficient optimal routing for communication in
data transmission model for mobile ad hoc VANETs via clustering model. In Emerging
network using signcryption technique. IEEE Technologies for Connected Internet of
Transactions on Reliability. Vehicles and Intelligent Transportation System
[43] Shanmugam, P., Rajesh, R. S., & Perumal, E. Networks (pp. 1-14). Springer, Cham.
(2008, May). A reversible watermarking with [53] Chen, C. H. (2020). A cell probe-based
low warping: an application to digital fundus method for vehicle speed estimation. IEICE
image. In 2008 International Conference on Transactions on Fundamentals of Electronics,
Computer and Communication Communications and Computer
Engineering (pp. 472-477). IEEE. Sciences, 103(1), 265-267.
[44] Shankar, K., & Elhoseny, M. (2019). Trust [54] Khamparia, A., Singh, A., Anand, D., Gupta,
Based Cluster Head Election of Secure D., Khanna, A., Kumar, N. A., & Tan, J.
Message Transmission in MANET Using (2018). A novel deep learning-based multi-
Multi Secure Protocol with TDES. J. model ensemble method for the prediction of
UCS, 25(10), 1221-1239. neuromuscular disorders. Neural computing
[45] Parvathy, V. S., Pothiraj, S., & Sampson, J. and applications, 1-13.
(2020). Optimal Deep Neural Network model [55] Shankar, K., Zhang, Y., Liu, Y., Wu, L., &
based multimodality fused medical image Chen, C. H. (2020). Hyperparameter tuning
classification. Physical Communication, deep learning for diabetic retinopathy fundus
101119. image classification. IEEE Access, 8, 118164-
[46] Subbiah Parvathy, V., Pothiraj, S., & 118173.
Sampson, J. (2020). A novel approach in [56] Jesus, E.F., Chicarino, V.R., de Albuquerque,
multimodality medical image fusion using C.V. and Rocha, A.A.D.A., 2018. A Survey of
optimal shearlet and deep How to Use Blockchain to Secure Internet of
learning. International Journal of Imaging Things and the Stalker Attack. Security and
Systems and Technology. Communication Networks, 2018.

ISSN: 2347-8578 www.ijcstjournal.org Page 115