Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks
Surveillance videos record malicious events in a locality utilizing various machine learning algorithms for detection. Deep-learning algorithms being the most prominent AI algorithms are data-hungry as well as computationally expensive. These algorithms perform better when trained over a diverse and...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universidad Internacional de La Rioja (UNIR)
2023-06-01
|
| Series: | International Journal of Interactive Multimedia and Artificial Intelligence |
| Subjects: | |
| Online Access: | https://www.ijimai.org/journal/bibcite/reference/3039 |
| _version_ | 1797810727185022976 |
|---|---|
| author | Atif Jan Gul Muhammad Khan |
| author_facet | Atif Jan Gul Muhammad Khan |
| author_sort | Atif Jan |
| collection | DOAJ |
| description | Surveillance videos record malicious events in a locality utilizing various machine learning algorithms for detection. Deep-learning algorithms being the most prominent AI algorithms are data-hungry as well as computationally expensive. These algorithms perform better when trained over a diverse and huge set of examples. These modern AI methods have a dire need of utilizing human intelligence to pamper the problem in such a way as to reduce the ultimate effort in terms of computational cost. In this research work, a novel methodology termed Bag of Focus (BoF) based training methodology has been proposed. BoF is based on the concept of selecting motion-intensive blocks in a long video, for training different deep neural networks (DNN's). The methodology reduced the computational overhead by 90% (ten times) in comparison to when full-length videos are entertained. It has been observed that training networks using BoF are equally effective in terms of performance for the same network trained over the full-length dataset. In this research work, firstly, a fine-grained annotated dataset including instance and activity information has been developed for real-world volume crimes. Secondly, a BoF-based methodology has been introduced for effective training of the state-of-the-art 3D, and 2D Convolutional Neural Networks (CNNs). Lastly, a comparison between the state-of-the-art networks have been presented for malicious event recognition in videos. It has been observed that 2D CNN even with lesser parameters achieved a promising classification accuracy of 98.7% and Area under the curve (AUC) of 99.7%. |
| first_indexed | 2024-03-13T07:12:06Z |
| format | Article |
| id | doaj.art-ad73644c09424667a87f574a9cfd6f9f |
| institution | Directory Open Access Journal |
| issn | 1989-1660 |
| language | English |
| last_indexed | 2024-03-13T07:12:06Z |
| publishDate | 2023-06-01 |
| publisher | Universidad Internacional de La Rioja (UNIR) |
| record_format | Article |
| series | International Journal of Interactive Multimedia and Artificial Intelligence |
| spelling | doaj.art-ad73644c09424667a87f574a9cfd6f9f2023-06-05T20:31:39ZengUniversidad Internacional de La Rioja (UNIR)International Journal of Interactive Multimedia and Artificial Intelligence1989-16602023-06-018215816710.9781/ijimai.2021.10.010ijimai.2021.10.010Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural NetworksAtif JanGul Muhammad KhanSurveillance videos record malicious events in a locality utilizing various machine learning algorithms for detection. Deep-learning algorithms being the most prominent AI algorithms are data-hungry as well as computationally expensive. These algorithms perform better when trained over a diverse and huge set of examples. These modern AI methods have a dire need of utilizing human intelligence to pamper the problem in such a way as to reduce the ultimate effort in terms of computational cost. In this research work, a novel methodology termed Bag of Focus (BoF) based training methodology has been proposed. BoF is based on the concept of selecting motion-intensive blocks in a long video, for training different deep neural networks (DNN's). The methodology reduced the computational overhead by 90% (ten times) in comparison to when full-length videos are entertained. It has been observed that training networks using BoF are equally effective in terms of performance for the same network trained over the full-length dataset. In this research work, firstly, a fine-grained annotated dataset including instance and activity information has been developed for real-world volume crimes. Secondly, a BoF-based methodology has been introduced for effective training of the state-of-the-art 3D, and 2D Convolutional Neural Networks (CNNs). Lastly, a comparison between the state-of-the-art networks have been presented for malicious event recognition in videos. It has been observed that 2D CNN even with lesser parameters achieved a promising classification accuracy of 98.7% and Area under the curve (AUC) of 99.7%.https://www.ijimai.org/journal/bibcite/reference/3039volume crime classificationcrime detectionmalicious activity detectiondeep learning |
| spellingShingle | Atif Jan Gul Muhammad Khan Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks International Journal of Interactive Multimedia and Artificial Intelligence volume crime classification crime detection malicious activity detection deep learning |
| title | Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks |
| title_full | Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks |
| title_fullStr | Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks |
| title_full_unstemmed | Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks |
| title_short | Real World Anomalous Scene Detection and Classification using Multilayer Deep Neural Networks |
| title_sort | real world anomalous scene detection and classification using multilayer deep neural networks |
| topic | volume crime classification crime detection malicious activity detection deep learning |
| url | https://www.ijimai.org/journal/bibcite/reference/3039 |
| work_keys_str_mv | AT atifjan realworldanomalousscenedetectionandclassificationusingmultilayerdeepneuralnetworks AT gulmuhammadkhan realworldanomalousscenedetectionandclassificationusingmultilayerdeepneuralnetworks |