# Evaluating OCR performance on food packaging labels in South Africa

arXiv: https://arxiv.org/abs/2510.03570 · HTML: https://mlsanity.com/paper/2510.03570/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 25 references that are in the map.
- First posted on arXiv 3 Oct 2025.

## Builds on (most cited first, 9 of 9)

- [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://mlsanity.com/paper/2109.10282/) (2021)
- [What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis](https://mlsanity.com/paper/1904.01906/) (2019)
- [TextOCR: Towards large-scale end-to-end reasoning for arbitrary-shaped scene text](https://mlsanity.com/paper/2105.05486/) (2021)
- [PP-OCR: A Practical Ultra Lightweight OCR System](https://mlsanity.com/paper/2009.09941/) (2020)
- [PaddleOCR 3.0 Technical Report](https://mlsanity.com/paper/2507.05595/) (2025)
- [A Multiplexed Network for End-to-End, Multilingual OCR](https://mlsanity.com/paper/2103.15992/) (2021)
- [Visual Text Meets Low-level Vision: A Comprehensive Survey on Visual Text Processing](https://mlsanity.com/paper/2402.03082/) (2024)
- [Text Detection Forgot About Document OCR](https://mlsanity.com/paper/2210.07903/) (2022)
- [Fast Glare Detection in Document Images](https://mlsanity.com/paper/1911.05189/) (2019)

## Built on by (most cited first, 2 of 2)

- [HalalBench: A Multilingual OCR Benchmark for Food Packaging Ingredient Extraction](https://mlsanity.com/paper/2604.22754/) (2026)
- [A Conservative OCR-Enabled Workflow for R214 Sodium Screening of South African Packaged Foods](https://mlsanity.com/paper/2609.15427/) (2026)

## Abstract (arXiv)

This study evaluates four open-source Optical Character Recognition (OCR) systems which are Tesseract, EasyOCR, PaddleOCR, and TrOCR on real world food packaging images. The aim is to assess their ability to extract ingredient lists and nutrition facts panels. Accurate OCR for packaging is important for compliance and nutrition monitoring but is challenging due to multilingual text, dense layouts, varied fonts, glare, and curved surfaces. A dataset of 231 products (1,628 images) was processed by all four models to assess speed and coverage, and a ground truth subset of 113 images (60 products) was created for accuracy evaluation. Metrics include Character Error Rate (CER), Word Error Rate (WER), BLEU, ROUGE-L, F1, coverage, and execution time. On the ground truth subset, Tesseract achieved the lowest CER (0.912) and the highest BLEU (0.245). EasyOCR provided a good balance between accuracy and multilingual support. PaddleOCR achieved near complete coverage but was slower because it ran on CPU only due to GPU incompatibility, and TrOCR produced the weakest results despite GPU acceleration. These results provide a packaging-specific benchmark, establish a baseline, and highlight directions for layout-aware methods and text localization.