Segmentation of Bruises in Fruit Thermograms During Post Harvest Stage Using K-Means Clustering Technique

Thermography is a beneficial technique for evaluating fruit quality non-destructively in the food processing industry. It allows for the identification of internal flaws like bruising. In this paper, the K-means clustering technique is applied to three different fruit types: guava, pomegranate, and apple, with the goal of segmenting bruises in fruit thermograms. The main goal is to improve automation in the sectors by quickly assessing quality in post-harvest procedures by easily identifying bruises. Preprocessing the fruit thermograms by scaling and conversion into appropriate colour space is part of the methodology. Because this study only includes three sample fruits, the bruised areas are identified using the simple K means clustering method to determine the nature of the fruit. The study identifies higher similarity regions for guava fruit and demonstrates the potential of K-means clustering as a dependable and adaptable tool for accurate bruise segmentation in guava fruits. In the future, this technique could be used to improve non-destructive inspection techniques in the agricultural sector.