Detecting Violent Robberies in CCTV Videos Using Deep Learning

Giorgio Morales; Itamar Salazar-Reque; Joel Telles; Daniel Díaz

doi:10.1007/978-3-030-19823-7_23

Detecting Violent Robberies in CCTV Videos Using Deep Learning

Giorgio Morales^*
, Itamar Salazar-Reque
, Joel Telles
, Daniel Díaz

^*Corresponding author for this work

National University of Engineering

Research output: Chapter in Book/Published conference output › Conference publication

24 Citations (SciVal)

Abstract

Video surveillance through security cameras has become difficult due to the fact that many systems require manual human inspection for identifying violent or suspicious scenarios, which is practically inefficient. Therefore, the contribution of this paper is twofold: the presentation of a video dataset called UNI-Crime, and the proposal of a violent robbery detection method in CCTV videos using a deep-learning sequence model. Each of the 30 frames of our videos passes through a pre-trained VGG-16 feature extractor; then, all the sequence of features is processed by two convolutional long-short term memory (convLSTM) layers; finally, the last hidden state passes through a series of fully-connected layers in order to obtain a single classification result. The method is able to detect a variety of violent robberies (i.e., armed robberies involving firearms or knives, or robberies showing different level of aggressiveness) with an accuracy of 96.69%.

Original language	English
Title of host publication	Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings
Editors	Elias Pimenidis, Ilias Maglogiannis, Lazaros Iliadis, John MacIntyre
Publisher	Springer
Pages	282-291
Number of pages	10
ISBN (Print)	9783030198220
DOIs	https://doi.org/10.1007/978-3-030-19823-7_23
Publication status	Published - 12 May 2019
Event	15th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2019 - Hersonissos, Greece Duration: 24 May 2019 → 26 May 2019

Publication series

Name	IFIP Advances in Information and Communication Technology
Volume	559
ISSN (Print)	1868-4238

Conference

Conference	15th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2019
Country/Territory	Greece
City	Hersonissos
Period	24/05/19 → 26/05/19

Bibliographical note

Publisher Copyright:
© 2019, IFIP International Federation for Information Processing.

Keywords

Action recognition
convLSTM
Robbery detection

Access to Document

10.1007/978-3-030-19823-7_23

Cite this

Morales, G., Salazar-Reque, I., Telles, J., & Díaz, D. (2019). Detecting Violent Robberies in CCTV Videos Using Deep Learning. In E. Pimenidis, I. Maglogiannis, L. Iliadis, & J. MacIntyre (Eds.), Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings (pp. 282-291). (IFIP Advances in Information and Communication Technology; Vol. 559). Springer. https://doi.org/10.1007/978-3-030-19823-7_23

Morales, Giorgio ; Salazar-Reque, Itamar ; Telles, Joel et al. / Detecting Violent Robberies in CCTV Videos Using Deep Learning. Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings. editor / Elias Pimenidis ; Ilias Maglogiannis ; Lazaros Iliadis ; John MacIntyre. Springer, 2019. pp. 282-291 (IFIP Advances in Information and Communication Technology).

@inproceedings{e8f2cca0aba84a7d893015a5d3b787ef,

title = "Detecting Violent Robberies in CCTV Videos Using Deep Learning",

abstract = "Video surveillance through security cameras has become difficult due to the fact that many systems require manual human inspection for identifying violent or suspicious scenarios, which is practically inefficient. Therefore, the contribution of this paper is twofold: the presentation of a video dataset called UNI-Crime, and the proposal of a violent robbery detection method in CCTV videos using a deep-learning sequence model. Each of the 30 frames of our videos passes through a pre-trained VGG-16 feature extractor; then, all the sequence of features is processed by two convolutional long-short term memory (convLSTM) layers; finally, the last hidden state passes through a series of fully-connected layers in order to obtain a single classification result. The method is able to detect a variety of violent robberies (i.e., armed robberies involving firearms or knives, or robberies showing different level of aggressiveness) with an accuracy of 96.69\%.",

keywords = "Action recognition, convLSTM, Robbery detection",

author = "Giorgio Morales and Itamar Salazar-Reque and Joel Telles and Daniel D{\'i}az",

note = "Publisher Copyright: {\textcopyright} 2019, IFIP International Federation for Information Processing.; 15th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2019 ; Conference date: 24-05-2019 Through 26-05-2019",

year = "2019",

month = may,

day = "12",

doi = "10.1007/978-3-030-19823-7\_23",

language = "English",

isbn = "9783030198220",

series = "IFIP Advances in Information and Communication Technology",

publisher = "Springer",

pages = "282--291",

editor = "Elias Pimenidis and Ilias Maglogiannis and Lazaros Iliadis and John MacIntyre",

booktitle = "Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings",

address = "Germany",

}

Morales, G, Salazar-Reque, I, Telles, J & Díaz, D 2019, Detecting Violent Robberies in CCTV Videos Using Deep Learning. in E Pimenidis, I Maglogiannis, L Iliadis & J MacIntyre (eds), Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings. IFIP Advances in Information and Communication Technology, vol. 559, Springer, pp. 282-291, 15th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2019, Hersonissos, Greece, 24/05/19. https://doi.org/10.1007/978-3-030-19823-7_23

Detecting Violent Robberies in CCTV Videos Using Deep Learning. / Morales, Giorgio; Salazar-Reque, Itamar; Telles, Joel et al.
Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings. ed. / Elias Pimenidis; Ilias Maglogiannis; Lazaros Iliadis; John MacIntyre. Springer, 2019. p. 282-291 (IFIP Advances in Information and Communication Technology; Vol. 559).

Research output: Chapter in Book/Published conference output › Conference publication

TY - GEN

T1 - Detecting Violent Robberies in CCTV Videos Using Deep Learning

AU - Morales, Giorgio

AU - Salazar-Reque, Itamar

AU - Telles, Joel

AU - Díaz, Daniel

PY - 2019/5/12

Y1 - 2019/5/12

N2 - Video surveillance through security cameras has become difficult due to the fact that many systems require manual human inspection for identifying violent or suspicious scenarios, which is practically inefficient. Therefore, the contribution of this paper is twofold: the presentation of a video dataset called UNI-Crime, and the proposal of a violent robbery detection method in CCTV videos using a deep-learning sequence model. Each of the 30 frames of our videos passes through a pre-trained VGG-16 feature extractor; then, all the sequence of features is processed by two convolutional long-short term memory (convLSTM) layers; finally, the last hidden state passes through a series of fully-connected layers in order to obtain a single classification result. The method is able to detect a variety of violent robberies (i.e., armed robberies involving firearms or knives, or robberies showing different level of aggressiveness) with an accuracy of 96.69%.

AB - Video surveillance through security cameras has become difficult due to the fact that many systems require manual human inspection for identifying violent or suspicious scenarios, which is practically inefficient. Therefore, the contribution of this paper is twofold: the presentation of a video dataset called UNI-Crime, and the proposal of a violent robbery detection method in CCTV videos using a deep-learning sequence model. Each of the 30 frames of our videos passes through a pre-trained VGG-16 feature extractor; then, all the sequence of features is processed by two convolutional long-short term memory (convLSTM) layers; finally, the last hidden state passes through a series of fully-connected layers in order to obtain a single classification result. The method is able to detect a variety of violent robberies (i.e., armed robberies involving firearms or knives, or robberies showing different level of aggressiveness) with an accuracy of 96.69%.

KW - Action recognition

KW - convLSTM

KW - Robbery detection

UR - https://www.scopus.com/pages/publications/85065913796

UR - https://link.springer.com/chapter/10.1007/978-3-030-19823-7_23

U2 - 10.1007/978-3-030-19823-7_23

DO - 10.1007/978-3-030-19823-7_23

M3 - Conference publication

AN - SCOPUS:85065913796

SN - 9783030198220

T3 - IFIP Advances in Information and Communication Technology

SP - 282

EP - 291

BT - Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings

A2 - Pimenidis, Elias

A2 - Maglogiannis, Ilias

A2 - Iliadis, Lazaros

A2 - MacIntyre, John

PB - Springer

T2 - 15th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2019

Y2 - 24 May 2019 through 26 May 2019

ER -

Morales G, Salazar-Reque I, Telles J, Díaz D. Detecting Violent Robberies in CCTV Videos Using Deep Learning. In Pimenidis E, Maglogiannis I, Iliadis L, MacIntyre J, editors, Artificial Intelligence Applications and Innovations - 15th IFIP WG 12.5 International Conference, AIAI 2019, Proceedings. Springer. 2019. p. 282-291. (IFIP Advances in Information and Communication Technology). doi: 10.1007/978-3-030-19823-7_23

Detecting Violent Robberies in CCTV Videos Using Deep Learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this