Proceedings of The 2nd International Conference On Advance And Scientific Innovation, ICASI 2019, 18 July, Banda Aceh, Indonesia

Research Article

Implementation Of Neural Network And Canny Edge Detection To Recognize the Crime Through Surveillance Cameras

Download317 downloads
  • @INPROCEEDINGS{10.4108/eai.18-7-2019.2288590,
        author={Erwien Tjipta Wijaya and Aditya Galih Sulaksono},
        title={Implementation Of Neural Network And Canny Edge Detection To Recognize the Crime Through Surveillance Cameras},
        proceedings={Proceedings of The 2nd International Conference On Advance And Scientific Innovation, ICASI 2019, 18 July, Banda Aceh,  Indonesia},
        publisher={EAI},
        proceedings_a={ICASI},
        year={2019},
        month={11},
        keywords={smart cities surveillance cameras neural-network system image processing canny edge detection facial emotion body gestures recognition mape},
        doi={10.4108/eai.18-7-2019.2288590}
    }
    
  • Erwien Tjipta Wijaya
    Aditya Galih Sulaksono
    Year: 2019
    Implementation Of Neural Network And Canny Edge Detection To Recognize the Crime Through Surveillance Cameras
    ICASI
    EAI
    DOI: 10.4108/eai.18-7-2019.2288590
Erwien Tjipta Wijaya1,*, Aditya Galih Sulaksono1
  • 1: Departement of Information System, University of Merdeka, Terusan Dieng No 57-59, Malang, Indonesia
*Contact email: erwien@unmer.ac.id

Abstract

The development of the smart city era in various countries is still offset by the magnitude of crime rates in every corner and downtown. Delivery of slow information to find out early crime will affect the precautions that should be taken, so as to require intelligent system-based tools. This research object is based in Jakarta Smart City (JSC) as an integrated command center with 5000 more surveillance cameras that have been installed at all strategic points in Jakarta. Surveillance cameras that have been fitted with intelligent system tools will send information to the JSC command center so that the category of criminal activity is immediately detected and immediate prevention. Among the analytical methods used are computational intelligence using Neural-network System, through video surveillance camera performed image processing using facial recognition techniques and body motion with canny edge detection algorithm segmentation. Measurement testing using a method of absolute percentage error (MAPE). So, the result level to the accuracy is 65%. Difficulty in the process of recognition because it is difficult to get the position of taking the corner of the suspect image and the victim. Implementation of intelligent system-based technology installed in surveillance cameras is expected to reduce crime rates.