Use of Image Analysis to Investigate Human Quality Classification of Apples

Abstract The quality grade of apple fruit has to be distinguished before marketing, but such classification is highly sensitive to human error. Image analysis can be used to extract external quality properties from digitized videoimages. Algorithms were developed to characterize shape, size and colour of apples. Classification of apple quality was simulated by “tree-based modelling” using objective measurements of the external properties. This methodology gave insight into the way in which external product features affect the human perception of quality. The characteristics influencing the classification were found to differ according to the variety. It was also found that humans have poor ability to repeat their quality estimation; this is defined as inconsistency. A methodology was developed to study the product features influencing this inconsistency. As the classification involved more product properties and became more complex, the error of human classification increased.