Speed Sign Detection and Recognition by Convolutional Neural Networks
By Maurice PeemenAlgorithm mapping

Dataset for training and testing a speed sign detection and recognition application:

A fully trainable application is for speed sign detection and recognition from a video stream is developed in this work. We show that a fully trainable solution can perform reliable classification under varying circumstances (day and night).
When such a parallel neural network is mapped to a parallel platform such as a GPU; real-time detection is achieved with 35 fps.

Data set:

To train the application we have build a dataset that contains 3617 patterns from various sources. Most patterns are collected with Google image search and Street View. Also some selected patterns from a different public available dataset are added to this set [1]. All patterns are cropped and scaled to 32x32 pixel grayscale patterns. Also the labeling of the patterns is done by means of an output matrix. Each column contains an output vector corresponding to the pattern number.

[1] Y.-Y. Nguwi and S.-Y. Cho, "Emergent self-organizing feature map for recognizing road sign images", 2009