VigilantAI: Real-time detection of anomalous activity from a video stream using deep learning
Bibliographic record
Abstract
In an era where artificial intelligence (AI) solutions are increasingly integrated into various sectors, we have utilized Artificial Intelligence for enhancing public safety through real-time detection of illegal activities such as robberies and threats at gunpoint using CCTV footage. With the proliferation of deep learning in object detection, the study focuses on deploying the YoloV5 model, trained on a custom dataset compiled from diverse CCTV sources and movies, to identify specific criminal actions. One of the major problems faced in this field is the availability of a large robust labeled dataset on which a deep learning model can be trained. For this purpose, we have created our own dataset by converting various CCTV footage and movies into images, and then labeling them with the correct class. In addition, we also augmented data by using various data augmentation techniques for the chosen images. This dataset, enriched through augmentation techniques and annotated with bounding boxes, allows for the precise detection of threats, achieving an accuracy rate of 85%. Our system stands out by not only spotting these activities but also by instantly alerting security personnel, facilitating a rapid response to potentially dangerous situations. This capability is important for law enforcement agencies worldwide, offering them an advanced tool to act swiftly and prevent crimes, thereby enhancing public security. The essence of our work demonstrates the practical application and significant impact of AI in bolstering security measures, providing a solid foundation for future enhancements in the field. Through this initiative, we aim to foster a safer environment in public spaces, reducing crime rates and increasing the general public’s sense of safety.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".