No abstract

Food waste is one of the key challenges of the agri-food sector: one third of the global food production is wasted yearly, while paradoxically 815 million people do not have access to sufficient and nutritious food. Food waste represents an economic loss for the agri-food supply chain and the whole society and significantly contributes to the GHG emissions. In Italy up to 5.1 million tons of food is wasted: nearly half of it is generated by agri-food supply chain actors. Retailers contribute to the 14% of the overall food waste produced and the main cause relies on products reaching the expiration date. Over the last years retailers have increasingly taken action in order to recover surplus food, encouraged by positive changes in the regulatory environment and the increasing relevance of Corporate Social Responsibility policies adopted by companies. Food donations have been increasing, but in many cases the surplus food redistribution process to food-aid organizations is still occasional and not formalized, leaving space for efficiency improvement. Surplus food close to expiration date, if not properly and timely handled, inevitably turns into waste. In this paper we introduce SIVEQ: a systematic solution which relies on novel technologies such as IoT and big data analytics to tackle this issue. Our system represents an added value to all actors involved, not only for NPOs who collect and redistribute surplus food.

No abstract

In this article we evaluate the impact of using two image pre-processing approaches with the objective of aiding an Optical Character Recognition (OCR) software in correctly retrieving an expiry date from an image of a product containing it. In particular, we analyze the impact of finding the rotation angle of an image using the Hough transform and the impact of image binarization using adaptive Gaussian threshold. We attempt to further increase OCR accuracy through a sliding window approach. Our results show that applying the Hough transform noticeably improves OCR performance with minimal impact on the execution time.

No abstract