I want to know how I can make my own images dataset like CIFAR-10 having different classes of my own choice. After making the dataset how I can use it in Tensorflow. I have downloaded images of different classes now I want to know how I can convert them into an format that can be used in Convolutional Neural Network in Tensorflow. Kindly help me with this problem.
You must be logged in to post. Please login or register an account.
You convert them to pixel arrays. The easiest thing to do is threshold them to basically 0s and 1s.
This is a series using KNN to do image analysis. The earlier videos are focused on converting images to arrays, as well as thresholding. You might want to check it out. I wouldn't be surprised if you could combine opencv nowadays to do these operations much more efficiently, however, then convert to array.
-Harrison 8 years ago
You must be logged in to post. Please login or register an account.
Thank you for replying. I will try to do, will come back if got any problem.
-Muhammad Hannan 8 years ago
You must be logged in to post. Please login or register an account.