GridVids
Sign in with YouTube

Neural Network Shape Detection Using C#

John Dodd May 19, 2019 20:10
Create an account to hide videos, mark them watched, or save them to Don't Miss.
Don't Miss
Create an account to hide videos, mark them watched, or save them to Don't Miss.
Mark watched
Create an account to hide videos, mark them watched, or save them to Don't Miss.
Hide
Open in popup Open on YouTube
Share video

Share video

Copy the GridVids link or iframe code.

Video link
Iframe code
Iframe content

Find out more at: johntravisdodd.wordpress.com This is a video to showcase a program I wrote which was originally intended for text detection. The user would draw a letter or number and the program would tell you which number or letter was drawn. In this example I use shapes, since the training time for 3 different shapes is much shorter than letters or numbers. This application uses C# forms, and a custom neural network classes I wrote to decipher which shape/letter/number is being entered. It can be trained on any data given to it, and theoretically has no limit on the number of classifications which can be handed to it. For example it could be trained to recognize drawn emojis, or abstract original symbols. I wrote this to show off my knowledge of neural networks and present a real-world type application of their usage. Thanks for watching, feel free to leave a comment if you have any questions.
People & Blogs
#programming #neural network #text detection #drawing #shape detection #c# #demonstration of neural net