Object Classification and Color Sensing Using YOLO V3

  • K N V Suresh Varma et al.


The object tracking, recognition and detection is turning into the elemental facet of recent technologies.
Alongside the detection of objects, the very next facet typically comes the colour of the object detected
which is the main objective of this project. In this paper, we have combined the task of object classification
and colour sensing to be done at once which are usually performed separately. The neural network YOLO
V3(You Only Look Once, version3) [1] is used for object detection that has been proved to achieve excellent
results. In combination with this, for color detection [3], we built an application through which one can
automatically get the name of the color by double clicking on the respective area in the obtained output on
the screen. This combination would be able to give the output of the image or video with bounding boxes
around the objects detected which are classified and trained with their respective names and their
respective colours.