For our flower classification example, we will be using the University of Oxford's Visual Geometry Group (VGG) image dataset collection. The collection can be accessed at http://www.robots.ox.ac.uk/~vgg/data/.
The VGG is the same department that won previous ImageNet competitions. The pretrained models, such as VGG14 and VGG16, were built by this department and they won in 2014 and 2016, respectively. These datasets are used by the VGG to train and evaluate the models that they build.
The flower dataset can be found in the Fine-Grain Recognition Datasets section of the page, along with textures and pet datasets. Click on Flower Category Datasets, or use the following link to access the flower datasets from VGG, http://www.robots.ox.ac.uk/~vgg/data/flowers/.
Here, you can find two datasets, one with 17 different species of flowers, and the other with 102 different species of flowers. You can choose either one based on their ease of use for the tutorial, or based on the kind of processing that is available at your disposal.
Here is a subset of the images you will find here. As you will see, the folder names match up identically with those we will use a bit later on in this chapter:
Aside from the images we talked about above, here are several additional links that you can use to get image data for similar classification use cases should you ever desire to use them:
- CVonline datasets: http://homepages.inf.ed.ac.uk/rbf/CVonline/Imagedbase.htm
- CVpapers datasets: http://www.cvpapers.com/datasets.html
- Image datasets: http://wiki.fast.ai/index.php/Image_Datasets
- Deep learning datasets: http://deeplearning.net/datasets/
- COCO datsets: http://cocodataset.org/#home
- ImageNet datasets: http://www.image-net.org/
- Open Images datasets: https://storage.googleapis.com/openimages/web/index.html
- Kaggle datasets: https://www.kaggle.com/datasets?sortBy=relevance&group=featured&search=image
- Open datasets: https://skymind.ai/wiki/open-datasets
- Wikipedia: https://en.wikipedia.org/wiki/List_of_datasets_for_machine_learning_research#Object_detection_and_recognition