I am trying to put together a public agricultural image database of corn and soybeans, to train convolutional neural networks. The main method of image collection will be through taking pictures of various fields in the growing season. The images will be uploaded to a public data sharing site which will be accessible by many.
However, I could get much more images compiled if I were to take some off of, say, Google Images. If there anything wrong with this? Would there be any issues with copywright infringements if I find the images on a publicly-available search engine? I need a lot of images, so I thought this would be a good method of increasing my image numbers.