Along with technological developments, many innovations have been developed
by humans, including in the field of digital image processing, digital image
processing techniques have now begun to spread to various fields, such as in the
fields of medicine, forensics, law, trade, education and in everyday life, especially
imagery. that can convey information. Image segmentation is an important part in
image analysis, because in this process the desired image or image will be
analyzed for further processing to make it easier to analyze. Elephant ear
ornamental plant or anthurium is one of the favorite plants that is loved by many
people. This unique shape, color, and size resembles an elephant's ear, no wonder
this uniqueness is the hallmark of this plant. This research will try to use one of
the local thresholding methods, namely the sauvola method in the case study of
anthurium leaves. The image segmentation process starts from image acquisition,
converts RGB to grayscale, local mean, standard deviation, then looks for the
thresholding value. The sauvola method (local thresholding) and global
thresholding need to be compared, so the error value is searched using Mean
Square Error (MSE). The results of the MSE test using 20 samples of anthurium
leaf imagery got the percentage of success reaching 80%, the lowest MSE value in
the sauvola method was obtained in the 11th sample with a value of 3423.9238,
while the global thresholding MSE value was 3400.3597. The conclusion is that
the sauvola method (local thresholding) is better than the global thresholding
method based on the MSE calculation of 20 samples which has been calculated
with a percentage reaching 80%.