Image based workflow screens sodium levels in south african packaged foods
A Conservative OCR-Enabled Workflow for R214 Sodium Screening of South African Packaged Foods
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Monitoring sodium content on food packages in South Africa is difficult because it requires recognizing the product, reading nutrition facts, and comparing them to specific legal limits. The authors designed a careful image-based process that uses computer vision and text recognition to check sodium levels on packages and decide if they meet rules or need review. They tested their method on hundreds of real food packages and showed it can accurately find cases that clearly pass or fail, and flag uncertain ones for further checking. This helps reduce mistakes by not forcing uncertain products into pass or fail categories.
What this means in practice
- •For food regulatory agencies: Automatically screen packaged foods for sodium compliance using image analysis to organize evidence and flag uncertain products for review.
- •For food safety auditors: Use an image-based tool to efficiently identify sodium labeling issues in packaged foods during on-site inspections.
Authors
Mayimunah Nagayi, Alice Scaria Khan, Tamryn Frank, Rina Swart, Clement Nyirenda
Abstract
Using food package images to monitor sodium and salt content against South Africa's R214 sodium limits is challenging when screening decisions require product identity, nutrition facts panel evidence, reporting basis, and category-specific thresholds. This study presents a conservative image-based workflow that combines region detection, optical character recognition (OCR), product identity and sodium evidence extraction, R214 category assignment, deterministic threshold comparison, and independent vision language model comparison. The evaluation used 442 packaged food products and 3 929 full package images from a real-world South African food packaging dataset. A YOLO26s small detector generated 4 195 region crops, and strict post-processing produced one sodium evidence row per product. The integrated workflow produced 290 OUTSIDE R214 SCOPE, 139 REVIEW, seven SCREEN-PASS, and six SCREEN-FAIL outcomes. The independent Qwen2.5-VL 7B vision language model workflow produced 387 OUTSIDE R214 SCOPE, 31 REVIEW, twenty SCREEN-PASS, and four SCREEN-FAIL outcomes. The workflows agreed on exact R214 category assignment for 415 of 442 products (93.9%) and on whether the assigned category was within R214 scope for 416 of 442 products (94.1%). Final screening outcome agreement was 307 out of 442 products, or 69.5%. Manual verification on 60 products showed lower strict outcome agreement than regulated status agreement, while all manual INSUFFICIENT DATA cases were kept out of SCREEN-PASS and SCREEN-FAIL by both automated workflows. The findings show that conservative image-based screening can organise package evidence, identify clear cases, and assign uncertain cases to REVIEW rather than forcing SCREEN-PASS or SCREEN-FAIL decisions.