Helsinki Deblur Challenge 2021 open photographic dataset

This dataset was primarily designed and captured to be used for the Helsinki Deblur Challenge 2021 but it can be used for any testing and benchmarking purposes of image deblurring algorithms.

The dataset contains photographs of random strings of text with varying levels of blur caused by misfocusing the camera. Each photo has both a blurred and sharp version. Why strings of text as targets? This is to enable quantitative measurement of deconvolution quality using an Optical Characted Recognition (OCR) algorithm in the Helsinki Deblur Challenge 2021.

The images are split into 20 separate zip files according to the level of blur. Each one of the zip files contains two folders, one for each of the different fonts used (Verdana and Times). In both font folders you find one folder named CAM1 (Camera 1) with the sharp images and one named CAM2 (Camera 2) with the blurred images. Each one of the CAM folders holds 100 images of the text targets and 3 images of technical targets for estimating the point spread function (PSF). Each one of the text target images is accompanied by a text file (same file name but .txt extension) containing the correct transcription of that particular text target.

An example image with one sharp-blurred pair of each one of the focus steps is also provided here for preview.

A more detailed description of the dataset can be found here: http://arxiv.org/abs/2105.10233

And the test dataset used for the HDC2021 can be found here: https://doi.org/10.5281/zenodo.5713637

Here is a link to the official webpage of the Helsinki Deblur Challenge 2021: https://www.fips.fi/HDC2021.php