Citrus Fruits

Today I learned about a cirtus fruit dataset called CitDet which was introduced in this paper. To quote the paper:

The dataset consists of over 32,000 bounding box annotations for fruit instances contained in 579 high-resolution images.

One feature that I thought was particularily interesting was that the dataset makes a distinction between fruits that are on the plant and those that have fallen on to the ground.

Figure from the paper that shows different algorithms applied to the dataset, as well as the distinction of fallen fruit.

It's another dataset that might be interesting if you're into computer vision for agriculture. The application of this dataset is also pretty cool because the pictures of the trees are from a region that suceptible to a disease. To quote the article:

... we present a new dataset to advance the state of the art in detecting citrus fruit and accurately estimate yield on trees affected by the Huanglongbing (HLB) disease in orchard environments via imaging. Despite the fact that significant progress has been made in solving the fruit detection problem, the lack of publicly available datasets has complicated direct comparison of results

Pretty neat!