THE 3
rd
INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS
dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
122
Figure 1.
Convolutional Neural Network system
Firstly, the image is divided to pixels and the pixels are sent as inputs to
system, all inputs has a line connected to hidden layer by which are multiplied
to certain coefficient (between 0-1). This term is again and again will be
applied to hidden layers and at the end last hidden layers which are
connected to output will present a final coefficient. By the use of final
coefficient system decide that there is an accident or not.
Experimental results
The structure covered in Research Methodology has been applied to
practice by the use of Deep Learning Ensemble in Python. In real time video
is represented to program to analyze if there is an accident or not. Initially,
the video splits to frames (figure 2), then processed to homogenize the
images to grayscale and downscale the width to 640 pixels. Then the trained
CNN model generate a preliminary prediction.
Dostları ilə paylaş: