Recent advances in imaging techniques for bruise detection in fruits and vegetables

Background: Bruising is the most common type of mechanical damage, which is inevitable along the postharvest chain, including harvesting, sorting, packaging and so on. It's the major cause of postharvest loss. However, the detection of bruises mainly relies on manual inspection which is time-consuming and mistake-prone, especially for the early bruises. In recent years, with the development of computer vision, more and more imaging techniques for real-time and automated bruise detection are proposed and studied again. Scope and approach: This review provides an overview of recent advances in imaging techniques for bruise detection. The imaging techniques presented include spectral techniques (biospeckle imaging, hyperspectral/multispectral imaging, fluorescence imaging and structured-illumination reflectance imaging), nuclear magnetic techniques (magnetic resonance imaging and X-ray imaging) and thermal imaging. The basic principles of these imaging techniques and their application in bruise detection are concisely introduced. Key findings and conclusion: The imaging techniques mentioned above exhibit potential to detect the bruised tissue and the bruise volume in real time and to overcome the fallibility, tediousness, destructiveness, and time-consumption, which are the disadvantages of traditional methods. In future research, integration of these techniques, developing specific algorithms for bruise detection and attempting to apply them into an assembly line for commercial value will be the research hotpots. In addition, the range of applications of these techniques for fruit and vegetable types also needs to be extended, and applying them in a practically assembly line should be enhanced.