Otsu thresholding on monitoring system PDAM water meter.

Image segmentation is a process of separating the foreground or object and the background or background. This separation process has problems, including those that often occur in pattern recognition, namely from the characteristics, thickness of numbers and also the results of image capture. Pattern recognition also has many identifications, one of which is using Otsu Thresholding. This thresholding method is used to automatically divide the histogram of the gray image into two different regions. The numbers detected are numbers on the PDAM water meter. For this PDAM meter, misunderstandings often occur in recording and many payments are not in accordance with the amount used. Therefore, a system was made to detect the PDAM meter number so that homeowners can see the amount of usage and the amount of fees to be paid. The system has 2 process flows, namely the process for training data is carried out manually and also the process for value identification is carried out through comparisons and differences in values from the threshold. The last step is to calculate the accuracy of image recognition. After testing the accuracy for image recognition obtained by 91% with testing as many as 100 images. [ABSTRACT FROM AUTHOR]