Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2012 · 32 citations · 0 references
AgricultureComputer ScienceCrop SciencesMachine LearningTopologyVegetable ScienceClassificationCrop ClassificationFruit QualityFruit Quality EvaluationNeural NetworksPattern RecognitionFruit Image AnalysisIdentificationNeural Network ClassificationNonparametric ClassificationSelf-organizing MapsSelf-organizing NetworksTopological ClassificationTopology Of Neural Networks
By using the classification properties of Kohonen-type networks (Tipping 1996), a neural model was built for the qualitybased identification of tomatoes. The resulting empirical data in the form of digital images of tomatoes at various stages of storage were subsequently used to draw up a topological SOFM (Self-Organizing Feature Map) which features cluster centers of "comparable" cases (Tadeusiewicz 1997, Boniecki 2008). Radial neurons from the Kohonen topological map were labeled appropriately to allow for the practical quality-based classification of tomatoes (De Grano 2007).