A Realistic Dataset for Performance Evaluation of Document Layout Analysis

Antonacopoulos, Apostolos; Bridson, David; Papadopoulos, Christos; Pletschacher, Stefan

doi:10.1109/icdar.2009.271

Cited by 110 publications

(61 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It is used in high-profile applications such as evaluation datasets for layout analysis of contemporary documents [9], datasets and extensive evaluation infrastructure for historical documents (within the scope of the IMPACT project) as well as international competitions (ICDAR competition series) [10].…”

Section: Discussionmentioning

confidence: 99%

The PAGE (Page Analysis and Ground-Truth Elements) Format Framework

Pletschacher¹,

Antonacopoulos²

2010

2010 20th International Conference on Pattern Recognition

Self Cite

125

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

The PAGE (Page Analysis and Ground-Truth Elements) Format Framework

Pletschacher¹,

Antonacopoulos²

2010

2010 20th International Conference on Pattern Recognition

Self Cite

125

View full text Add to dashboard Cite

show abstract

“…The importance of the availability of realistic datasets for meaningful performance evaluation has been repeatedly discussed and the authors have addressed the issue for contemporary documents by creating the PRImA Layout Analysis dataset with ground truth [4] and making it available to all researchers. The overall dataset contains a wide selection of contemporary documents (with complex as well as simple layouts) together with comprehensive ground truth and extensive metadata.…”

Section: The Datasetmentioning

confidence: 99%

“…Since the 2009 edition of the ICDAR Page Segmentation competition a more extensive evaluation scheme has been used [3], allowing for higher level goal-oriented evaluation and much more detailed region comparison, going far beyond simple precision/recall metrics. In addition, the used datasets have been selected from curated repositories [4][5] containing realistic and representative documents. This edition (RDCL2017) is based on the same principles established and refined by the 2011, 2013, and 2015 competitions on historical and contemporary document layout analysis [6] but its focus is on documents with complex layouts.…”

Section: Introductionmentioning

confidence: 99%

ICDAR2017 Competition on Recognition of Documents with Complex Layouts - RDCL2017

Clausner

Antonacopoulos

Pletschacher

2017

2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

View full text Add to dashboard Cite

“…For instance, Antonacopoulos et al 26 proposed a dataset to evaluate techniques of document layout analysis. That dataset contains 1, 240 images from websites, newspaper pages, magazines pages.…”

Section: Evaluation Datasetsmentioning

confidence: 99%

Benchmark for license plate character segmentation

Gonçalves

Silva

Menotti

et al. 2016

J. Electron. Imaging

View full text Add to dashboard Cite

Abstract. Automatic License Plate Recognition (ALPR) has been the focus of many researches in the past years. In general, ALPR is divided into the following problems: detection of on-track vehicles, license plates detection, segmention of license plate characters and optical character recognition (OCR). Even though commercial solutions are available for controlled acquisition conditions, e.g., the entrance of a parking lot, ALPR is still an open problem when dealing with data acquired from uncontrolled environments, such as roads and highways when relying only on imaging sensors. Due to the multiple orientations and scales of the license plates captured by the camera, a very challenging task of the ALPR is the License Plate Character Segmentation (LPCS) step, which effectiveness is required to be (near) optimal to achieve a high recognition rate by the OCR. To tackle the LPCS problem, this work proposes a novel benchmark composed of a dataset designed to focus specifically on the character segmentation step of the ALPR within an evaluation protocol. Furthermore, we propose the Jaccard-Centroid coefficient, a new evaluation measure more suitable than the Jaccard coefficient regarding the location of the bounding box within the ground-truth annotation. The dataset is composed of 2,000 Brazilian license plates consisting of 14,000 alphanumeric symbols and their corresponding bounding box annotations. We also present a new straightforward approach to perform LPCS efficiently. Finally, we provide an experimental evaluation for the dataset based on five LPCS approaches and demonstrate the importance of character segmentation for achieving an accurate OCR.

show abstract

A Realistic Dataset for Performance Evaluation of Document Layout Analysis

Cited by 110 publications

References 2 publications

The PAGE (Page Analysis and Ground-Truth Elements) Format Framework

The PAGE (Page Analysis and Ground-Truth Elements) Format Framework

ICDAR2017 Competition on Recognition of Documents with Complex Layouts - RDCL2017

Benchmark for license plate character segmentation

Contact Info

Product

Resources

About