Artificial intelligence-based techniques for adulteration and defect detections in food and agricultural industry: A review

Artificial Intelligence (AI) techniques have evolved into practical, fast and effective tools in combination with detecting devices for quality assessment, particularly in adulteration and deficiency detection in the food and agriculture industry. This review discusses recent advances in AI techniques and their integration with a variety of sensing devices to detect food adulteration and agricultural product defects. The results obtained from the sensors require the application of case-specific AI techniques aimed at improving the acquired high-dimensional dataset's understanding, as well as classification and prediction. It is evident that the coupling of AI technique and sensors have shown promising outcome between 81.2 and 100% range of accuracy in adulteration and defect detection for food and agricultural product. The research trends and guidelines are also proposed with the aim to provide references and guidance to both scientific researchers and industrial players in the field of food and agriculture quality assessment. The challenges and prospects regarding AI techniques were also revealed. Furthermore, future potential developments, new sensors and novel algorithms must be pursued and validated. Future AI detection of food adulteration and agricultural product defects can be anticipated, especially with the integration of various sensors and the application of deep learning algorithms. Real-time monitoring and predictive modeling may also receive great attention, which could assist to avoid quality problems before they arise, lower the risk of fraud, and ensure high-valuable products.