ISSN (Online) 2581-9429

IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252

Design and Implementation of the Suspicious

Activity Detection System using Machine Learning
Anagha Jawalkar1, Divya Shirtode2, Chaitrali Pandit3, Divya Varute4, Dnyaneshwar Kengar5
Assistant Professor, Department of Information Technology1
UG Scholar, Department of Information Technology2,3,4,5
RMD Sinhgad School of Engineering, Pune, Maharashtra, India
Savitribai Phule Pune University, Pune, Maharashtra, India

Abstract: We intend to develop a real-time programme for detecting suspicious behaviour of persons in
public settings. Our tool may be used to monitor areas where there is a risk of robbery or a gun assault,
such as malls, airports, and train stations. To train our system, we will use deep learning and neural
networks. This model will then be implemented as a mobile and desktop app, taking real-time CCTV footage
as input and sending an alarm to the administrator's smartphone if a suspicious stance is detected. Human
suspicious behaviour is associated with the identification of human bodily parts and perhaps tracking their
travels. Its real-world applications range from gaming to AR/VR, healthcare, and gesture detection. In
comparison to the image data domain, there has been very limited research into using CNNs to video
categorization. This is due to the fact that videos are more complicated than photos since they have another
dimension - temporal. Unsupervised learning, which takes use of temporal connections between frames, has
proven effective for video analysis. Some techniques to suspicious behaviour employ CPU rather than GPU,
allowing suspicious activity to execute on low-cost hardware such as embedded systems and mobile phones.

Keywords: Suspicious behaviour, Machine Learning, CNN, Random Forest (RF) and KNN, CCTV.

I. INTRODUCTION
Human suspicious activity is one of the key problems in computer vision that has been studied for more than 15 years. It is
important because of the sheer number of applications which can benefit from suspicious activity. For example, human
suspicious activity is used in applications including video surveillance, animal tracking and behavior understanding, sign
language detection, advanced human-computer interaction, and marker less motion capturing. Low-cost depth sensors
have limitations like limited to indoor use, and their low resolution and noisy depth information make it difficult to
estimate human poses from depth images. Hence, we plan to use neural networks to overcome these problems. Suspicious
human activity recognition from surveillance video is an active research area of image processing and computer vision.
Through the visual surveillance, human activities can be monitored in sensitive and public areas such as bus stations,
railway stations, airports, banks, shopping malls, school and colleges, parking lots, roads, etc. to prevent terrorism, theft,
accidents and illegal parking, vandalism, fighting, chain snatching, crime and other suspicious activities. It is very difficult
to watch public places continuously, therefore an intelligent video surveillance is required that can monitor the human
activities in real-time and categorize them as usual and unusual activities; and can generate an alert. Mostly, of the
research being carried out is on images and not videos. Also, none of the papers published tries to use CNNs to detect
suspicious activities.

II. RELATED WORK

Mohammad Sabokrou ET AL. [1] suggested a system for real-time suspicious identification and localisation in congested
settings in this work. Each movie is specified as a series of non-overlapping cubic patches, with two local and global
descriptors. These descriptions capture video qualities from various angles. We can discriminate between normal and
abnormal activity in films by adding simple and low-cost Gaussian classifiers. Mahmudul Hasan et al. [2] proposed the
unclear notion of'meaningfulness,' as well as clutter in the environment, perceiving meaningful actions in a long video
sequence is a difficult task. We tackle this challenge by creating a generative model for regular motion patterns (also

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 823

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252
known as regularity) from many sources with very little supervision. Jefferson Ryan Medel et al. [3] proposed that
because of the vagueness in how such events are classified, automating the identification of anomalous occurrences within
extended video sequences is difficult. We address the challenge by training generative models to detect abnormalities in
films with little supervision. End-to-end trainable composite Convolutional Long Short-Term Memory (Conv-LSTM)
networks that can anticipate the evolution of a video sequence from a limited number of input frames are proposed. Yong
Shean Chong et al. [4] said We offer a fast approach for identifying video abnormalities. Recent convolutional neural
network applications have demonstrated the promise of convolutional layers for object identification and recognition,
particularly in photos. Convolutional neural networks, on the other hand, are supervised and require labels as learning
signals. S. L. Bangare et al. [5-11] have worked in the brain tumor detection. N. Shelke et al [12] given LRA-DNN
method. Suneet Gupta et al [13] worked for end user system. Gururaj Awate et al. [14] worked on Alzheimers Disease. P.
S. Bangare et al [15-17] worked on the object detection. Kalpana Thakare et al [18-23] have worked on various machine
learning algorithms. M. L. Bangare et al. [24-25] worked on the cloud platform. Rajesaheb R. Kadam et al [26] and
Sachindra K. Chavan et al. [27] have discussed security issues with cloud.

III. SYSTEM ANALYSIS

3.1 System Architecture

Fig.1 System Architecture

3.2 Proposed Work of Model

Data collection begins with the extraction of information for various Websites and Social Media apps depending on
certain factors.
1. Preprocessing: To prepare our dataset, we will do different pre-processing techniques such as noise removal,
resizing, binary conversion, and grey scaling.
2. Noise removal: Noise from the input video is eliminated. Filtering is the main step in image processing for
denoising; commonly, average filters, median filters, Wiener filters, and Kalman filters are used to minimise
noise.
3. Picture scaling is required when we need to increase or reduce the total number of pixels, whereas remapping is
used when we need to compensate for lens distortion or rotate an image.
4. Binary conversion: A binary picture is one that has pixels that can only have one of two colours, black and white.
Binary pictures are sometimes known as bi-level or two-level images. This means that each and every pixel is
stored as a single bit, with values ranging from 0 to 1.
5. Gray scaling is the process of converting a continuous-tone image into an image that a computer can easily alter.
6. Picture segmentation is an important technique that involves isolating a digital image into several parts, i.e. (sets
of pixels, also recognised as image objects).
Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 824
www.ijarsct.co.in
 ISSN (Online) 2581
2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252
7. Data Training: We collect artificial as
a well as real-time
time web news data and train any machine learning classifier.
8. Feature extraction: Feature extraction is a step in the dimensionality reduction technique that involves separating
and compacting an initial collection of raw data into more mana
manageable categories.
9. Classification is the technique of classifying and identifying groupings of pixels or vectors inside a picture using
certain rules and instructions.
10. Data Training: We collected simulated as well as realreal-time social media data and trainedd any machine learning
classifier.
11. Machine learning testing: We provide a testing dataset to the system and use a machine learning algorithm to
detect activity.

3.3 UML diagrams

Fig.2 Data Flow diagram

Fig.3 Use case Diagram

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 825

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252

Fig.4 Activity Diagram

Fig. 5 Sequence Diagram

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 826

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252

Fig.6 Class Diagram

IV. EXPERIMENTAL RESULTS

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 827

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252

Fig.7 Screens shots

V. CONCLUSION
A system that processes real-time CCTV data to detect any suspicious behaviour would assist to improve security and
reduce the need for human involvement. Great advancements have been achieved in the realm of human suspicious
activity, allowing us to better serve the numerous applications that it is capable of using CNN. Furthermore, research in
adjacent domains such as Activity Tracking can significantly improve its productive application in a variety of fields.

REFERENCES
[1] Sabokrou, Mohammad & Fathy, Mahmood & Mojtaba, H. & Klette, Reinhard,“Real-Time Anomaly Detection and
Localization in Crowded Scenes”.Conference: Computer Vision and Pattern Recognition (CVPR)At: Boston,
MassachusettsVolume: GROW Workshop (2015).
[2] Mahmudul Hasan, Jonghyun Choi, Jan Neumann, Amit K. Roy-Chowdhury, Larry S. Davis, “Learning Temporal
Regularity in Video Sequences”, https://doi.org/10.48550/arXiv.1604.04574
[3] Jefferson Ryan Medel, Andreas Savakis, “Anomaly Detection in Video Using Predictive Convolutional Long Short-
Term Memory Networks”, https://doi.org/10.48550/arXiv.1612.00390
[4] Yong Shean Chong, Yong Haur Tay, “Abnormal Event Detection in Videos using Spatiotemporal Autoencoder”,
https://doi.org/10.48550/arXiv.1701.01546

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 828

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252
[5]S. L. Bangare, G. Pradeepini, S. T. Patil, “Implementation for brain tumor detection and three dimensional
visualization model development for reconstruction”, ARPN Journal of Engineering and Applied Sciences (ARPN JEAS),
Vol.13, Issue.2, ISSN 1819-6608, pp.467-473. 20/1/2018 http://www.arpnjournals.org/jeas/research_papers
/rp_2018/jeas_0118_6691.pdf
[6]S. L. Bangare, S. T. Patil et al, “Reviewing Otsu’s Method for Image Thresholding.” International Journal of Applied
Engineering Research, ISSN 0973-4562, Volume 10, Number 9 (2015) pp. 21777-21783, © Research India Publications
https://dx.doi.org/10.37622/IJAER/10.9.2015.21777-21783
[7]S. L. Bangare, G. Pradeepini, S. T. Patil, “Regenerative pixel mode and tumor locus algorithm development for brain
tumor analysis: a new computational technique for precise medical imaging”, International Journal of Biomedical
Engineering and Technology, Inderscience, 2018, Vol.27 No.1/2. https://www.inderscienceonline.com/doi/pdf/
10.1504/IJBET.2018.093087
[8]S. L. Bangare, A. R. Khare, P. S. Bangare, “Quality measurement of modularized object oriented software using
metrics”, ICWET '11: Proceedings of the International Conference & Workshop on Emerging Trends in Technology,
February 2011, pp. 771–774. https://doi.org/10.1145/1980022.1980190.1.
[9]S. L. Bangare, G. Pradeepini and S. T. Patil, "Brain tumor classification using mixed method approach," 2017
International Conference on Information Communication and Embedded Systems (ICICES), 2017, pp. 1-4, doi:
10.1109/ICICES.2017.8070748.
[10]S. L. Bangare, S. Prakash, K. Gulati, B. Veeru, G. Dhiman and S. Jaiswal, "The Architecture, Classification, and
Unsolved Research Issues of Big Data extraction as well as decomposing the Internet of Vehicles (IoV)," 2021 6th
International Conference on Signal Processing, Computing and Control (ISPCC), 2021, pp. 566-571, doi:
10.1109/ISPCC53510.2021.9609451.
[11]S. L. Bangare, G. Pradeepini, S. T. Patil et al, “Neuroendoscopy Adapter Module Development for Better Brain
Tumor Image Visualization”, International Journal of Electrical and Computer Engineering (IJECE) Vol. 7, No. 6,
December 2017, pp. 3643~3654. http://ijece.iaescore.com/index.php/IJECE/article/view/8733/7392
[12]N. Shelke, S. Chaudhury, S. Chakrabarti, S. L. Bangare et al. “An efficient way of text-based emotion analysis from
social media using LRA-DNN”, Neuroscience Informatics, Volume 2, Issue 3, September 2022, 100048, ISSN 2772-
5286, https://doi.org/10.1016/j.neuri.2022.100048 .
[13]Suneet Gupta, Sumit Kumar, Sunil L. Bangare, Shibili Nuhmani, Arnold C. Alguno, Issah Abubakari Samori,
“Homogeneous Decision Community Extraction Based on End-User Mental Behavior on Social Media”, Computational
Intelligence and Neuroscience, vol. 2022, Article ID 3490860, 9 pages, 2022. https://doi.org/10.1155/2022/3490860.
[14]Gururaj Awate, S. L. Bangare, G. Pradeepini and S. T. Patil, “Detection of Alzheimers Disease from MRI using
Convolutional Neural Network with Tensorflow”,arXiv, https://doi.org/10.48550/arXiv.1806.10170
[15]P. S. Bangare, S. L. Bangare, R. U. Yawle and S. T. Patil, "Detection of human feature in abandoned object with
modern security alert system using Android Application," 2017 International Conference on Emerging Trends &
Innovation in ICT (ICEI), 2017, pp. 139-144, doi: 10.1109/ETIICT.2017.7977025.
[16] P. S. Bangare and S. L. Bangare. "The Campus Navigator: An Android Mobile Application." International Journal of
Advanced Research in Computer and Communication Engineering 3, no. 3 (2014): 5715-5717.
[17] P. S. Bangare, N. J. Uke, and S. L. Bangare, "An approach for detecting abandoned object from real time video."
International Journal of Engineering Research and Applications (IJERA) 2.3 (2012): 2646-2649.
[18]Kalpana S. Thakare, Viraj Varale, “Prediction of Heart Disease using Machine Learning Algorithm”, Bioscience
Biotechnology Research Communications (Special issue) Volume 13, Issue 12, 2020 (Dec 2020 issue).
[19]Kalpana S. Thakare, A. M. Rajurkar, “Shot Boundary Detection of MPEG Video using Biorthogonal Wavelet
Transform”, International Journal of Pure and Applied Mathematics, Volume 118, No. 7, pp. 405-413, ISSN: 1311-8080
(printed version); ISSN: 1314-3395 (on-line version), url: http://www.ijpam.eu
[20]Kalpana S. Thakare, A. M. Rajurkar, R. R. Manthalkar, “Video Partitioning and Secured Key frame Extraction of
MPEG Video”, Proceedia Computer Science Journal, Volume 78, pp 790-798, Elsevier, 2016. Scopus DOI:
htpp://10.1016/j.procs.2016.02.058, www.sciencedirect.com/science/article/pii/S1877050916000600

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 829

www.ijarsct.co.in
 ISSN (Online) 2581-9429
IJARSCT
International Journal of Advanced Research in Science, Communication and Technology (IJARSCT)

Volume 2, Issue 5, May 2022

Impact Factor: 6.252
[21]Kalpana S. Thakare, A. M. Rajurkar and R. R. Manthalkar, “Content based Video Retrieval using Latent Semantic
Indexing and Color, Motion and Edge Features”, International Journal of Computer Applications 54(12):42-48, September
2012, Published by Foundation of Computer Science, New York, USA. DOI: 10.5120/8621-2486
[22]Kalpana S. Thakare, Archana M. Rajurkar, R. R. Manthalkar, “A Comprehensive System Based on Spatiotemporal
Features Such as motion, Quantized Color and Edge Features”, International Journal of Wireless and Microwave
Technologies (IJWMT) ISSN 1449 (Print), ISSN: 2076-9539 (Online), Vol.1, No.3, June. 2011, DOI: 10.5815 /ijwmt
[23]Kalpana S. Thakare, Archana M. Rajurkar, Dr. R. R. Manthalkar, “An effective CBVR system based on Motion,
Quantized color and edge density features”, International Journal of Computer Science & Information Technology
(IJCSIT), ISSN 0975 – 3826, Vol 3, No 2, April 2011 DOI: 10.5121/ijcsit.2011.3206 78.
[24]M. L. Bangare, “Attribute Based Encryption And Data Integrity For Attack on Cloud Storage”, Journal of Analysis
and Computation (JAC), (An International Peer Reviewed Journal), www.ijaconline.com, ISSN 0973-2861, ICASETMP-
2019, pp.1-4. http://www.ijaconline.com/wp-content/uploads/2019/07/ICASETMP67.pdf
[25]M. L. Bangare, Sarang A. Joshi, “Kernel interpolation-based technique for privacy protection of pluggable data in
cloud computing”, International Journal of Cloud Computing, Volume 9, Issue 2-3, pp.355-374, Publisher Inderscience
Publishers (IEL).
[26]Rajesaheb R. Kadam and Manoj L. Bangare, “A survey on security issues and solutions in live virtual machine
migration”, International Journal of Advance Foundation and Research in Computer (IJAFRC), (December, 2012). ISSN
(2014), pp.2348-4853.
[27]Sachindra K. Chavan, Manoj L. Bangare, “Secure Data Storage in Cloud Service using RC5 Algorithm”, International
Journal of Recent Technology and Engineering (IJRTE), ISSN: 2277-3878, Volume-2, Issue-5 November 2013, pp.139-
144.

Copyright to IJARSCT DOI: 10.48175/IJARSCT-4143 830

www.ijarsct.co.in