Search

In the fields of Optical Character Recognition (OCR) and Natural Language Processing (NLP), integrating multilingual capabilities remains a critical challenge, especially when considering languages with complex scripts such as Arabic. This paper introduces the Comprehensive Post-OCR Parsing and Receipt Understanding Dataset (CORU), a novel dataset specifically designed to enhance OCR and information extraction from receipts in multilingual contexts involving Arabic and English. CORU consists of over 20,000 annotated receipts from diverse retail settings, including supermarkets and clothing stores, alongside 30,000 annotated images for OCR that were utilized to recognize each detected line, and 10,000 items annotated for detailed information extraction. These annotations capture essential details such as merchant names, item descriptions, total prices, receipt numbers, and dates. They are structured to support three primary computational tasks: object detection, OCR, and information extraction. We establish the baseline performance for a range of models on CORU to evaluate the effectiveness of traditional methods, like Tesseract OCR, and more advanced neural network-based approaches. These baselines are crucial for processing the complex and noisy document layouts typical of real-world receipts and for advancing the state of automated multilingual document processing. Our datasets are publicly accessible (https://github.com/Update-For-Integrated-Business-AI/CORU).

In this paper, we address the significant gap in Arabic natural language processing (NLP) resources by introducing ArabicaQA, the first large-scale dataset for machine reading comprehension and open-domain question answering in Arabic. This comprehensive dataset, consisting of 89,095 answerable and 3,701 unanswerable questions created by crowdworkers to look similar to answerable ones, along with additional labels of open-domain questions marks a crucial advancement in Arabic NLP resources. We also present AraDPR, the first dense passage retrieval model trained on the Arabic Wikipedia corpus, specifically designed to tackle the unique challenges of Arabic text retrieval. Furthermore, our study includes extensive benchmarking of large language models (LLMs) for Arabic question answering, critically evaluating their performance in the Arabic language context. In conclusion, ArabicaQA, AraDPR, and the benchmarking of LLMs in Arabic question answering offer significant advancements in the field of Arabic NLP. The dataset and code are publicly accessible for further research https://github.com/DataScienceUIBK/ArabicaQA., Comment: Accepted at SIGIR 2024

The extraction of key information from receipts is a complex task that involves the recognition and extraction of text from scanned receipts. This process is crucial as it enables the retrieval of essential content and organizing it into structured documents for easy access and analysis. In this paper, we present AMuRD, a novel multilingual human-annotated dataset specifically designed for information extraction from receipts. This dataset comprises $47,720$ samples and addresses the key challenges in information extraction and item classification - the two critical aspects of data analysis in the retail industry. Each sample includes annotations for item names and attributes such as price, brand, and more. This detailed annotation facilitates a comprehensive understanding of each item on the receipt. Furthermore, the dataset provides classification into $44$ distinct product categories. This classification feature allows for a more organized and efficient analysis of the items, enhancing the usability of the dataset for various applications. In our study, we evaluated various language model architectures, e.g., by fine-tuning LLaMA models on the AMuRD dataset. Our approach yielded exceptional results, with an F1 score of 97.43\% and accuracy of 94.99\% in information extraction and classification, and an even higher F1 score of 98.51\% and accuracy of 97.06\% observed in specific tasks. The dataset and code are publicly accessible for further researchhttps://github.com/Update-For-Integrated-Business-AI/AMuRD.

Tables are everywhere, from scientific journals, papers, websites, and newspapers all the way to items we buy at the supermarket. Detecting them is thus of utmost importance to automatically understanding the content of a document. The performance of table detection has substantially increased thanks to the rapid development of deep learning networks. The goals of this survey are to provide a profound comprehension of the major developments in the field of Table Detection, offer insight into the different methodologies, and provide a systematic taxonomy of the different approaches. Furthermore, we provide an analysis of both classic and new applications in the field. Lastly, the datasets and source code of the existing models are organized to provide the reader with a compass on this vast literature. Finally, we go over the architecture of utilizing various object detection and table structure recognition methods to create an effective and efficient system, as well as a set of development trends to keep up with state-of-the-art algorithms and future research. We have also set up a public GitHub repository where we will be updating the most recent publications, open data, and source code. The GitHub repository is available at https://github.com/abdoelsayed2016/table-detection-structure-recognition.

Arabic is one of the languages that present special challenges to Optical character recognition (OCR). The main challenge in Arabic is that it is mostly cursive. Therefore, a segmentation process must be carried out to determine where the character begins and where it ends. This step is essential for character recognition. This paper presents Arabic character segmentation algorithm. The proposed algorithm uses the projection-based approach concepts to separate lines, words, and characters. This is done using profile's amplitude filter and simple edge tool to find characters separations. Our algorithm shows promising performance when applied on different machine printed documents with different Arabic fonts.

This work introduces an economic solution for the problems of sound insulation of recording studios. Sound insulation at wall resonance frequency is weak. Instead of acoustical treatment, a digital filter is used to eliminate the effects of wall resonance and coincidence phenomena on recording of speech. Sound insulation of studio is measured to calculate the wall resonance frequency and the coincidence frequency. Pole /zero placement technique is used to calculate the IIR filter coefficients. The digital filter is designed, simulated and implemented. The proposed system is used to treat these problems and it is shown to be effective in recording the noisy speech. In this work digital signal processing is used instead of the acoustic treatment to eliminate the effect of noise at the studio wall resonance. This technique is cheap and effective in canceling the noise at the desired frequencies. Field Programmable Gate Array (FPGA) is used for hardware implementation of the proposed filter structure which provides fast and cheap solution for processing real time audio signals. The implementation is carried out using Spartan chip from Xinlinx achieving higher performance than commercially available software solutions., Comment: Submitted to Journal of Telecommunications, see http://sites.google.com/site/journaloftelecommunications/volume-2-issue-2-may-2010

To improve the performance of speaker identification systems, an effective and robust method is proposed to extract speech features, capable of operating in noisy environment. Based on the time-frequency multi-resolution property of wavelet transform, the input speech signal is decomposed into various frequency channels. For capturing the characteristic of the signal, the Mel-Frequency Cepstral Coefficients (MFCCs) of the wavelet channels are calculated. Hidden Markov Models (HMMs) were used for the recognition stage as they give better recognition for the speaker's features than Dynamic Time Warping (DTW). Comparison of the proposed approach with the MFCCs conventional feature extraction method shows that the proposed method not only effectively reduces the influence of noise, but also improves recognition. A recognition rate of 99.3% was obtained using the proposed feature extraction technique compared to 98.7% using the MFCCs. When the test patterns were corrupted by additive white Gaussian noise with 20 dB S/N ratio, the recognition rate was 97.3% using the proposed method compared to 93.3% using the MFCCs.