Remote imaging systems raise unprecedented challenges in artificial intelligence. The dataset provided (extracted from the SpaceNet 6 challenge) shows SAR images having distorted intensities (compared to the expected results, the latter being visible in the RGB and NIR images which are also provided) due to the geophysics of the SAR acquisition system and the geometries of ground objects. Can we teach an Artificial Intelligence to find the right re-projections for automatically correcting such distorted and compressed intensities ?


This data set is regarding the paper submitted to the IEEE Transactions on Molecular, Biological, and Multi-Scale Communications. The title of the paper is 'Molecular Signal Tracking and Detection Methods in Fluid Dynamic Channels' with the ID of TMBMC-TPS-19-0014.R2. The data are images taken from the particle image velocimetry (PIV) method and the Planar Laser-Induced fluorescence (PLIF) method. The images are being used to describe these two experimental methods for the molecular communication community.


The dataset consists of 60285 character image files which has been randomly divided into 54239 (90%) images as training set 6046 (10%) images as test set. The collection of data samples was carried out in two phases. The first phase consists of distributing a tabular form and asking people to write the characters five times each. Filled-in forms were collected from around 200 different individuals in the age group 12-23 years. The second phase was the collection of handwritten sheets such as answer sheets and classroom notes from students in the same age group.