Detection of weapon possession and fire in Public Safety surveillance cameras

Abstract

The employment of video surveillance cameras by public safety agencies enables incident detection in monitored cities by using object detection for scene description, enhancing the protection to the general public. Object detection has its drawbacks, such as false positives. Our work aims to enhance object detection and image classification by employing IoU (Intersection over Union) to minimize the false positives and identify weapon holders or fire in a frame, adding more information to the scene.

Ethics: What is the Research Scenario in the Brazilian Conference BRACIS?

“…As one possible explanation for this result, BRACIS publications emphasize technical aspects and contributions. Many resort to available online databases, artificial or not, for training models or other experiments, as in [Moura et al 2021, Nunes et al 2013]. Among the 42 results of the wide screening, 21 (50%) show involvement or participation of human beings.…”

“…USP figuring in three of the nine papers [dos Santos and Paraboni 2019, Pavan et al 2020, Gonc ¸alves and Cozman 2021. UFBA is involved in two, both with explicit and provocative social critical content [Ferreira et al 2021, Moura et al 2021].…”

“…Similar to the result of RQ8, there was a significant variation, considering the technological sub-domain (RQ1) and application context variation (RQ8). For example, the difference usage between images or videos [Moura et al 2021, Ferreira et al 2021, applying YOLO (You Only Look Once); and texts [dos Paraboni 2019, Pavan et al 2020], applying Support Vector Machine (SVM) or Random Forest (RF).…”

“…The inclusion step considered nine publications presenting relevant and pertinent ethical aspects for qualitative synthesis [Moura et al 2021, Broder and Berton 2021, Ferreira et al 2021, Gonc ¸alves and Cozman 2021, Pavan et al 2020, Viana and Alcântara 2020, dos Santos and Paraboni 2019, Carvalho et al 2014, Nunes et al 2013. Four researchers extracted data in order to answer the questions in Table 1.…”