Efficient and Accurate Date Extraction from Invoices: A Comprehensive Three-Step Methodology Integrating Custom Object Detection, OCR, and Refined Regular Expressions

Öz

In the realm of contemporary document processing, the challenge of extracting crucial information from diverse invoices necessitates innovative solutions. This article presents a comprehensive three-step methodology to address the complexity of date extraction from invoices. Leveraging LabelStudio, Python, and OpenCV, we constitute a dataset and train a custom object detection model using Ultralytics YOLOv8. Optical Character Recognition (OCR) provides us to convert the image data to string data that is enable to be processed. Regular expressions refine the extracted text, achieving precise date formats. The developed system significantly enhance the time efficiency, marking a noteworthy advancement in date extraction from invoices.

Anahtar Kelimeler

• Custom Object Detection
• OCR
• Invoice Processing
• YOLOv8

Kaynakça

1. Open Source Data Labeling | Label Studio. (n.d.). Label Studio. https://labelstud.io/
2. Ultralytics. (n.d.). GitHub - ultralytics/ultralytics: NEW - YOLOv8 in PyTorch > ONNX > OpenVINO > CoreML > TFLite. GitHub. https://github.com/ultralytics/ultralytics
3. PaddlePaddle. (n.d.). GitHub - PaddlePaddle/PaddleOCR: Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices). GitHub. https://github.com/PaddlePaddle/PaddleOCR
4. M. S. Satav, T. Varade, D. Kothavale, S. Thombare and P. Lokhande, "Data Extraction From Invoices Using Computer Vision," 2020 IEEE 15th International Conference on Industrial and Information Systems (ICIIS), RUPNAGAR, India, 2020, pp. 316-320, doi: 10.1109/ICIIS51140.2020.9342722.
5. D. A. Kosiba and R. Kasturi, "Automatic invoice interpretation: invoice structure analysis," Proceedings of 13th International Conference on Pattern Recognition, Vienna, Austria, 1996, pp. 721-725 vol.3, doi: 10.1109/ICPR.1996.547263.
6. H. Sidhwa, S. Kulshrestha, S. Malhotra and S. Virmani, "Text Extraction from Bills and Invoices," 2018 International Conference on Advances in Computing, Communication Control and Networking (ICACCCN), Greater Noida, India, 2018, pp. 564-568, doi: 10.1109/ICACCCN.2018.8748309.
7. Tesseract-Ocr. (n.d.). GitHub - tesseract-ocr/tesseract: Tesseract Open Source OCR Engine (main repository). GitHub. https://github.com/tesseract-ocr/tesseract
8. R. Smith, "An Overview of the Tesseract OCR Engine," Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), Curitiba, Brazil, 2007, pp. 629-633, doi: 10.1109/ICDAR.2007.4376991.

9. Ar, I., Karsligil, M.E. (2007). Text Area Detection in Digital Documents Images Using Textural Features. In: Kropatsch, W.G., Kampel, M., Hanbury, A. (eds) Computer Analysis of Images and Patterns. CAIP 2007. Lecture Notes in Computer Science, vol 4673. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74272-2_69
10. Zhang, X., Duan, L., Ma, L., Wu, J. (2017). Text Extraction for Historical Tibetan Document Images Based on Connected Component Analysis and Corner Point Detection. In: Yang, J., et al. Computer Vision. CCCV 2017. Communications in Computer and Information Science, vol 772. Springer, Singapore. https://doi.org/10.1007/978-981-10-7302-1_45
11. Dachengwang,. (2011). ANALYSIS OF FORM IMAGES. International Journal of Pattern Recognition and Artificial Intelligence. 08. 10.1142/S0218001494000528.