APPLE CLASSIFICATION BASED ON SURFACE BRUISES USING IMAGE PROCESSING AND NEURAL NETWORKS

Abstract

Maintaining prime fruit quality is the key to success in the fresh fruit business. Quality defects such as bruises inapples adversely affect their market value. Linescan xray imaging has shown potential for detecting these quality defects.Quality assessment of apples with computer vision techniques is possible; however, two basic issues must be addressed beforean automatic sorting system can be developed: (1) which image features best correlate with the fruit quality, and (2) whichclassifier should be used for optimal classification. These issues are discussed in this article. Red delicious (RD) and goldendelicious (GD) apples were linescanned for bruise damage. Spatial and transform features were evaluated for theirdiscriminating contributions to fruit classification based on bruise defects. Stepwise discriminant analysis was used forselecting the salient features. Spatial edge features detected using Roberts edge detector, combined with the selected discretecosine transform (DCT) coefficients proved to be good indicators of old (one month) bruises. Separate artificial neuralnetwork (ANN) classifiers were developed for old (one month) and new (24 hour) bruises. When an ANN classifier was usedto sort apples based on old bruises, it achieved an accuracy of 90% for RD apples and 83% (93% after threshold adjustment)for GD apples. For new bruises, the accuracy was approximately 60% for both RD and GD apples. New bruises were notadequately separated using this methodology.