Afrocentric palm oil adulteration learning models: An end-to-end deep learning approach for detection of palm oil adulteration

Food safety remains a critical challenge across Africa, where foodborne illness rates are among the highest globally. One pressing issue is the widespread adulteration of red palm oil in West African markets using harmful synthetic dyes. Traditional detection methods rely on costly and time-intensive laboratory testing, limiting their use in local contexts. This research introduces AfroPALM, a deep learning-based solution that uses high-resolution image analysis to detect adulterated palm oil directly from raw image data. The study developed and tested multiple AI models, including AfroPALM-GhostNet, which achieved over 96% accuracy. By offering a fast, scalable, and low-resource alternative to laboratory testing, AfroPALM demonstrates how AI can strengthen food safety systems and protect public health across African markets.