COCO-10

Description: COCO-10 contains 5700 colour images of 150 colour barcodes printed on paper with two different densities and acquired by several smartphone cameras under different illuminations. For each image, a mask specifying the position of the barcodes is provided. 

The 150 color barcodes, acquired under the different conditions listed above, have been generated by coloring the lines of 150 standard black & white barcodes. The line colors have been randomly selected from a palette of 10 colors. The 150 color barcodes are released along with a set of 150 synthetic colour images, where colour barcodes have been super-imposed over cluttered, real-world backgrounds. The total number of images in COCO-10, including masks, is 11700.

The name COCO-10 is an acronym for COlor BarCOde and the number 10 refers to the fact that barcode colours are picked up from a palette of 10 colors. 

Aim: COCO-10 would like to be a testbench to develop and/or test algorithms for device- and illuminant- invariant colour barcode detection and decoding. It can be used also to develop and test algorithms for gamut and tone mapping, colour correction, machine colour constancy. 

Structure: COCO-10 is organized in three folders, as sketched in the picture below: COLOUR-BARCODES (150 images), COLOUR-BARCODES-ON-WHITE-PAPERS (10800 images) and COLOUR-BARCODES-ON-CLUTTERED-BACKGROUND (750 images).

More details can be found in https://zenodo.org/records/10065746.

Publications:

M. Lecca, P.  Lecca. A Dataset for Illuminant- and Device- invariant Colour Barcode Decoding with Cameras. Data in Brief, Vol. 52, Feb. 2024 [doi]