SIMULINK Model for Crime Intention Detection Using Convolutional Neural Network (CNN-Based)

Development and Implementation of Convolutional Neural Network (CNN-Based) Crime-Intention Detection System" this journal aim at developing a system that reliably detects and forecasts possible criminal activity in real-time. The approach overcomes CNN limitations by integrating contextual information and multimodal data. Because ethical factors like privacy and surveillance are taken into account, the technology is scalable for practical uses. In order to enhance timely crime detection and location monitoring, the crime-intention detection system was created utilizing a CNN rule-based system and a traditional SIMULINK architecture. The study contrasted the criminal detection rate of the CNN-enhanced system with that of a traditional system. Findings indicated that traditional problems with inadequate financing and resources led to poor crime detection at a rate of 35%; the CNN method improved this percentage by 5.07%, to 29.93%. Crime detection was impacted by inadequate training by 28%; however, with CNN integration, this decreased to 23.94%, indicating a 4.06% improvement. Furthermore, the total crime detection rate rose from 32% to 43.77%, indicating an 11.77% improvement, whereas data management problems decreased detection by 10% and improved to 8.55% using CNN (1.45% better). These results contribute to the development of AI-driven crime detection technology by demonstrating that CNN algorithms are more successful than traditional techniques at increasing crime detection rates.