Detection of sugar syrup adulteration in unifloral honey using deep learning framework: An effective quality analysis technique

The adulteration of honey is a common practice in the production and processing, industries. Various analytical methods such as chromatography, DNA-based, and NMR technology have been used to detect honey adulteration. However, these methods still face challenges in addressing honey authentication issues due to their sophistication, cost, and feasibility. For instance, these methods are expensive, time-consuming, require trained professionals for their operation, and are inaccessible to most of the population. This paper proposes a deep learning framework based on a two-dimensional convolutional neural network to overcome these challenges. The framework uses high-resolution video sequences (one minute) of each honey sample with varying levels of sugar syrup adulteration. Before using the videos in the model, they undergo pre-processing. The performance of the proposed method is validated using several performance indices, including accuracy (0.94), precision, sensitivity (0.99), specificity (1.00), F1-score (0.98), ROC-AUC (0.98), etc. The experimental results demonstrate the potential application of the proposed model in honey quality evaluation. However, further research is required to improve and expand the application of the model in this field, which could positively impact the global honey market.