Tobacco Leaves Disease Identification and Spot Segmentation Based on the Improved ORB Algorithm

In order to improve the problems of poor accuracy and low efficiency in tobacco leaves disease recognition diagnosis avoid misjudgment recognition, a spot segmentation method based on improved ORB algorithm was proposed. The FAST14-24 used preliminarily extract corners. It overcame deficiency sensitivity traditional corner detection image edges. During experiment, 28 parameters were obtained through extraction color features, morphological texture, other features spots. Through experimental comparisons, it found that fitness 96.68 cross-checking rate 93.21%. validation for samples 96%. identification brown frog eye 92%, 6 categories different periods over results verified effects fully.