ICDAR 2021 Competition on Historical Map Segmentation

Abstract

This paper presents the final results of the ICDAR 2021 Competition on Historical Map Segmentation (MapSeg), encouraging research on a series of historical atlases of Paris, France, drawn at 1/5000 scale between 1894 and 1937. The competition featured three tasks, awarded separately. Task 1 consists in detecting building blocks and was won by the L3IRIS team using a DenseNet-121 network trained in a weakly supervised fashion. This task is evaluated on 3 large images containing hundreds of shapes to detect. Task 2 consists in segmenting map content from the larger map sheet, and was won by the UWB team using a U-Net-like FCN combined with a binarization method to increase detection edge accuracy. Task 3 consists in locating intersection points of geo-referencing lines, and was also won by the UWB team who used a dedicated pipeline combining binarization, line detection with Hough transform, candidate filtering, and template matching for intersection refinement. Tasks 2 and 3 are evaluated on 95 map sheets with complex content. Dataset, evaluation tools and results are available under permissive licensing at https://icdar21-mapseg.github.io/.

Robust Grid Detection in Historical Map Images

“…The CMM [1] approach uses morphological processing and the Radon transform to detect the graticule lines. The lines correspond to the Radon transform maxima at corresponding and orthogonal angle.…”

“…The L3IRIS [1] approach uses the U-Net [6] as intersection point detector. There are several noisy point predictions in the output so the Hough transform is used to detect candidate lines.…”

“…We use the same scenario, provided evaluation tools and dataset [12] as in MapSeg competition [1]. The source codes and other related materials are freely available for non-commercial purposes at https://gitlab.kiv.zcu.cz/balounj/21_icdar_ mapseg_competition.…”

“…Important features that can be used for geo-referencing the maps are graticule lines indicating the North/South/East/West major coordinates and their intersections. [1] Another use case is assembling the single map sheets into a larger seamless map. This task was also the part of the ICDAR 2021 Competition on Historical Map Segmentation 1 [1] (MapSeg).…”

“…[1] Another use case is assembling the single map sheets into a larger seamless map. This task was also the part of the ICDAR 2021 Competition on Historical Map Segmentation 1 [1] (MapSeg). In this work we concentrate on the graticule line intersections detection according to the Task 3 of the competition and present the winning UWB method.…”