Applied Engineering in Agriculture · 2004 · 19 citations · 0 references
ComputingComputer ScienceFood ScienceComputer VisionSensory EvaluationVisual PerceptionAppearance EvaluationImage AnalysisImage ClassificationImage TextureTexture AnalysisTexture EvaluationCo-occurrence AnalysisColor Coordinate SystemsColor SystemsColor Texture AnalysisEvaluationStatistical Texture AnalysisTexture ClassificationTexture Features
A computer vision technique was used to evaluate internal hollowness, an important textural attribute of Frenchfries. Three computer vision algorithms (co-occurrence matrices, texture features corresponding to visual perception, andrun-length methods) were modified, implemented, and evaluated on color French fry cross section images. Evaluation andclassification of internal hollowness were based on textural features in two color coordinate systems (RGB and I1, I2, I 3).Statistical procedures were used to determine the classification capabilities of the selected significant texture features of thethree algorithms. A maximum of 100% classification accuracy was obtained using any of the three algorithms.