Abstract
This paper introduces a novel deep learning architecture, named DIGITNET, and a large-scale digit dataset, named DIDA, to detect and recognize handwritten digits in historical document images written in the nineteen century. To generate the DIDA dataset, digit images are collected from 100,000 Swedish handwritten historical document images, which were written by different priests with different handwriting styles. This dataset contains three sub-datasets including single digit, large-scale bounding box annotated multi-digit, and digit string with 250,000, 25,000, and 200,000 samples in Red-Green-Blue (RGB) color spaces, respectively. Moreover, DIDA is used to train the DIGITNET network, which consists of two deep learning architectures, called DIGITNET-dect and DIGITNET-rec, respectively, to isolate digits and recognize digit strings in historical handwritten documents. In DIGITNET-dect architecture, to extract features from digits, three residual units where each residual unit has three convolution neural network structures are used and then a detection strategy based on You Look Only Once (YOLO) algorithm is employed to detect handwritten digits at two different scales. In DIGITNET-rec, the detected isolated digits are passed through 3 different designed Convolutional Neural Network (CNN) architectures and then the classification results of three different CNNs are combined using a voting scheme to recognize digit strings. The proposed model is also trained with various existing handwritten digit datasets and then validated over historical handwritten digit strings. The experimental results show that the proposed architecture trained with DIDA (publicly available from: https://didadataset.github.io/DIDA/) outperforms the state-of-the-art methods.
Highlights
In the last two decades, there has been tremendous escalation in digitalization of handwritten documents to preserve the valuable historical information [1]
To detect digits in historical documents using digit detection algorithms, the methods based on Self-Organizing Maps (SOM) [11], Connected Component (COC) [12], Features of Connected Component (FEC) [13], Skeleton (SKE) [17] are implemented
The dataset has been collected from the historical Swedish handwritten document images written between the year 1800 and 1940 and contains: 1) single digit images with original appearance, 2) multi-digit images in RGB color space and 3) image dataset for deep learning object detection algorithms
Summary
In the last two decades, there has been tremendous escalation in digitalization of handwritten documents to preserve the valuable historical information [1]. Tion framework must automatically extract textual (e.g. characters, words, and sentences) and/or numerical (single- and multi-digits) contents from handwritten document images. This is a very challenging problem due to large intra- and inter-intensity variations as well as inter-class similarities and intra-class disparities in images. In order to avoid time-consuming and inefficient search processes, it is a vital task to develop an automatic handwritten digit string detection and recognition system.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
