Automatic classification of paddy leaf disease

Shafaf Ibrahim (Universiti Teknologi MARA Cawangan Melaka (Kampus Jasin))
Nurnazihah Wahab (Universiti Teknologi MARA Cawangan Melaka (Kampus Jasin))
Ahmad Firdaus Ahmad Fadzil (Universiti Teknologi MARA Cawangan Melaka (Kampus Jasin))
Nur Nabilah Abu Mangshor (Universiti Teknologi MARA Cawangan Melaka (Kampus Jasin))
Zaaba Ahmad (Universiti Teknologi MARA Cawangan Melaka (Kampus Jasin))

Indonesian Journal of Electrical Engineering and Computer Science
Vol 16, No 2: November 2019

RiceisastaplefoodinmostoftheAsiancountries.Itisanimportantcrop, andoverhalfoftheworldpopulationreliesonitforfood.However,paddy leafdiseasecanaffectboththequalityandquantityofpaddyinagriculture production.Theclassificationofpaddyleafdiseaseisanimportantand urgenttaskasitdestroysabout10%to15%ofproductioninAsia.Thus,a studyonautomaticclassificationofpaddyleafdiseaseusingimage processingispresented.Featureextractiontechniquesofcolor,texture,and shapewereimplementedtoanalyzethecharacteristicsofthepaddyleaf disease.Inanother note,aSupportVector Machine(SVM)isused toclassify thefourtypesofpaddyleafdiseasewhicharethebrownspot,bacterialleaf blight,tungrovirus,andleaf scald.Theperformanceofthe proposedstudyis evaluatedto160testingimageswhichreturned86.25%ofclassification accuracy.Theoutcomeofthisstudyisexpectedtoassisttheagrotechnology industryinearlydetectionofpaddyleafdiseaseinwhichanappropriate actioncould be taken accordingly.