Identification of citrus disease using color texture features and discriminant analysis

摘要: The citrus industry is an important constituent of Florida's overall agricultural economy. Proper disease control measures must be undertaken in groves to minimize losses. Technological strategies using machine vision and artificial intelligence are being investigated achieve intelligent farming, including early detection diseases groves, selective fungicide application, etc. This research used the color co-occurrence method (CCM) determine whether texture based hue, saturation, intensity (HSI) features conjunction with statistical classification algorithms could identify diseased normal leaves under laboratory conditions. Normal leaf samples greasy spot, melanose, scab were evaluated. sample discriminant analysis CCM textural achieved accuracies over 95% for all classes when hue saturation features. Data models that relied on suffered a reduction accuracy categorizing fronts, due darker pigmentation fronts. was not experienced backs where lighter clearly revealed discoloration. Although, high unreduced dataset consisting HSI features, best performer determined reduced data model selected computational load elimination which robust presence ambient light variation.