Uddin, Dr. JiaNakib, MohammadKhan, Rozin TanvirHasan, Md. Sakibul2017-07-302017-07-3020174/13/2017ID 13301044ID 13101117ID 13101145http://hdl.handle.net/10361/8373Cataloged from PDF version of thesis report.Includes bibliographical references (page 43-44).This thesis report is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2017.Crime scene prediction without human intervention can have outstanding impact on computer vision. In this paper, we present CNN in the use of detect knife, blood and gun in order to reach a prediction whether a crime has occurred in a particular image. We emphasized on the accuracy of detection so that it hardly gives us wrong alert to ensure efficient use of the system. This paper use Non linearity ReLu, Convolutional Neural Layer, Fully connected layer and dropout function of CNN to reach a result for the detection. We use Tensorflow open source platform to implement CNN to achieve our expected output. This system can achieve the test accuracy of 90.2 % for the data sets we have that is very much competitive with other systems for this particular task.44 pagesenBRAC University thesis are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.Crime sceneConvolutional neural networkThesis