CONCLUSION
The main aim of the work was to realise the detection and recognition of road
signs. To achieve this aim, the performer was assigned a number of tasks related to
the study of existing methods for selecting objects in the image, searching for the
area of the sign in the images, and recognizing traffic signs.
The existing methods of searching for objects in images were studied, and a
relevant algorithm was implemented to solve this problem.
Methods for recognizing road signs were studied.
This is implemented by
using the OpenCV and Keras libraries. The final software system is able to recognize
the most common types of road signs.
Naturally, implementation has places that
can be optimized for better
performance. It is also possible to implement the
recognition of non-trivial
(combined)
characters, the type of which does not allow
recognizing them at this
stage.
All steps that are mandatory and necessary for correct
operation in various
conditions: аbility to configure filters, automatic formatting of various types of input
and output data in accordance with the program
requirements have been
implemented in the program.