After all steps we need to combine trained RNN to real application which will recognize traffic signs from video stream. Lets’s do it. Python program: #!/usr/bin/env python ''' traffic signs detection with trained RNN traffic_ru_v1.pb USAGE: ts_detect_2.py ''' # Python 2/3 compatibility from __future__ import print_function from __future__ import division # for...
There is more simple architecture of RNN with tensorflow but result on russian traffic dataset is bad (about 70% of recognized images with 100000 epoch – training time – about 15 hours). Anyway I will show python program for information. """ modification of traffic_ru dataset with image preprocession (make YUV mormalized images) from http://jokla.me/robotics/traffic-signs/...
We have prepared all necessary files after implementation of part 1 and part 2 works. Lets start procedure of learning with tensorflow. TF model part taken at this link (Jiovanny Claudio). Jiovanny gave very good explanation of Tensorflow model which have used for deep learning. I have made some corrections in model for correct...
I have described bad attempts to train RNN with Russian Traffic dataset before – here for example. To solve this problem first I have sorted all images into classes to it’s own directory and found that quality of images is very bad: 1. there is a lot of images which can’t recognize by human...