Leaf Disease Recognition in Vine Plants Based on Local Binary Patterns and One Class Support Vector Machines

The current application concerns a new approach for disease recognition of vine leaves based on Local Binary Patterns (LBPs). The LBP approach was applied on color digital pictures with a natural complex background that contained infected leaves. The pictures were captured with a smartphone camera from vine plants. A 32-bin histogram was calculated by the LBP characteristic features that resulted from a Hue plane. Moreover, four One Class Support Vector Machines (OCSVMs) were trained with a training set of 8 pictures from each disease including healthy, Powdery Mildew and Black Rot and Downy Mildew. The trained OCSVMs were tested with 100 infected vine leaf pictures corresponding to each disease which were capable of generalizing correctly, when presented with vine leave which was infected by the same disease. The recognition percentage reached 97 %, 95 % and 93 % for each disease respectively while healthy plants were recognized with an accuracy rate of 100 %.