• Palivela Madhu Bharathi
• Dr.Manjula Devarakonda Venkata
• D.Uma

Without human assistance, crime scene prediction can have a significant impact on computer vision. The usage of CNN (Convolutional Neural Network) to detect knife, blood, and gun from an image is presented in this research. Detecting these threatening objects in an image can help us determine whether or not a crime has occurred, as well as where the image was captured. We focused on detecting accuracy so that we don't get false alerts and can make the most of the system. To arrive at a detection result, this model employs the Convolutional Layer, Fully Connected Layer, and CNN Dropout Function. To obtain our desired results, we develop CNN using Tensorflow, an open source platform. For the dataset that was examined, the proposed model had 90.2 percent accuracy.

Crime Prediction, CNN, Tensorflow, Deep Learning, Keras, Surveillance Systems.

"CRIME PREDICTION AND PREVENTION USING DEEP LEARNING AND COMPUTER VISION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 9, page no.a719-a725, September-2022, Available :http://www.jetir.org/papers/JETIR2209076.pdf

"CRIME PREDICTION AND PREVENTION USING DEEP LEARNING AND COMPUTER VISION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 9, page no. ppa719-a725, September-2022, Available at : http://www.jetir.org/papers/JETIR2209076.pdf