Performance evaluation for document image analysis and understanding is a recurring problem. Many ground- truthed document image databases are now used to evaluate algorithms, but these databases are less useful for the design of a complete system in a precise context. This paper proposes an approach for the automatic generation of synthesised document images and associated ground-truth information based on a derivation of publishing tools. An implementation of this approach illustrates the richness of the produced information.

International audienceRecent progress in the digitization of heterogeneous collections of ancient documents has rekindled new challenges in information retrieval in digital libraries and document layout analysis. Therefore, in order to control the quality of historical document image digitization and to meet the need of a characterization of their content using intermediate level metadata (between image and document structure), we propose a fast automatic layout segmentation of old document images based on five descriptors. Those descriptors, based on the autocorrelation function, are obtained by multiresolution analysis and used afterwards in a specific clustering method. The method proposed in this article has the advantage that it is performed without any hypothesis on the document structure, either about the document model (physical structure), or the typographical parameters (logical structure). It is also parameter-free since it automatically adapts to the image content. In this paper, firstly, we detail our proposal to characterize the content of old documents by extracting the autocorrelation features in the different areas of a page and at several resolutions. Then, we show that is possible to automatically find the homogeneous regions defined by similar indices of autocorrelation without knowledge about the number of clusters using adapted hierarchical ascendant classification and consensus clustering approaches. To assess our method, we apply our algorithm on 316 old document images, which encompass six centuries (1200-1900) of French history, in order to demonstrate the performance of our proposal in terms of segmentation and characterization of heterogeneous corpus content. Moreover, we define a new evaluation metric, the homogeneity measure, which aims at evaluating the segmentation and characterization accuracy of our methodology. We find a 85% of mean homogeneity accuracy. Those results help to represent a document by a hierarchy of layout structure and content, and to define one or more signatures for each page, on the basis of a hierarchical representation of homogeneous blocks and their topology

In this paper, we present an evaluative study of pixel-labeling methods using the HBA 1.0 dataset for historical book analysis. This study is held in the context of the 2nd historical book analysis (HBA2019) competition and in conjunction with the 15th IAPR international conference on document analysis and recognition (ICDAR2019). The HBA2019 competition provides a large experimental corpus and a thorough evaluation protocol to ensure an objective performance benchmarking of pixel-labeling document image methods. Two nested challenges are evaluated in the HBA2019 competition: Challenge 1 and Challenge 2. Challenge 1 evaluates how image analysis methods could discriminate the textual content from the graphical ones at pixel level. Challenge 2 assesses the capabilities of pixel-labeling methods to separate the textual content according to different text fonts (e.g. lowercase, uppercase, italic, etc.) at pixel level. During the competition, we received 52 and 38 different teams' registrations for Challenge 1 and Challenge 2, respectively and finally 5 of them submitted their results in each challenge. Qualitative and numerical results of the participating methods in both challenges are reported and discussed in this paper in order to provide a baseline for future evaluation studies in historical document image analysis. The evaluation shows that the method submitted by the NLPR-CASIA team achieves the highest performance in both challenges.

Recently, texture features have been widely used for historical document image analysis. However, few studies have focused exclusively on feature selection algorithms for historical document image analysis. Indeed, an important need has emerged to use a feature selection algorithm in data mining and machine learning tasks, since it helps to reduce the data dimensionality and to increase the algorithm performance such as a pixel classification algorithm. Therefore, in this paper we propose a comparative study of two conventional feature selection algorithms, genetic algorithm and ReliefF algorithm, using a classical pixel-labeling scheme based on analyzing and selecting texture features. The two assessed feature selection algorithms in this study have been applied on a training set of the HBR dataset in order to deduce the most selected texture features of each analyzed texture-based feature set. The evaluated feature sets in this study consist of numerous state-of-the-art texture features (Tamura, local binary patterns, gray-level run-length matrix, auto-correlation function, gray-level co-occurrence matrix, Gabor filters, Three-level Haar wavelet transform, three-level wavelet transform using 3-tap Daubechies filter and three-level wavelet transform using 4-tap Daubechies filter). In our experiments, a public corpus of historical document images provided in the context of the historical book recognition contest (HBR2013 dataset: PRImA, Salford, UK) has been used. Qualitative and numerical experiments are given in this study in order to provide a set of comprehensive guidelines on the strengths and the weaknesses of each assessed feature selection algorithm according to the used texture feature set.