Identifikasi Jumlah Bibit Bandeng Menggunakan Metode K-Means Berbasis HSV Color dan Morfologi
Seed milkfish is one commodity of national food security. Availability of seed milkfish as one of the major production support in the cultivation of milkfish in ponds must be fulfilled. Factors seed availability is essential in improving the commodities which impact on improving the living standard of farmers' welfare milkfish seedling cultivation. Seed fish are difficult to identify because the object is small so that farmers banding should be extra seedlings in calculating the amount of seed milkfish contained in one container. Identification of seed milkfish (milkfish seeds) one way to find out information on the number of seeds in a container milkfish. This research proposes the identification number of seeds banding using the K-Means method based on the HSV Color and morphology preprocessing. This research begins with step preprocessing, do transformasi Color original image RGB to HSV and RGB to Grayscale by the threshold value the image of S and V the Color space (Color space) HSV and morphology, the next process then feature extraction based on the area and the latter process is counting the number of seeds that are recognized as banding objects based on the results of clustering using the K-Means method. Based on the results of testing milkfish Seed identification show reached 92.70% accuracy and error rate 7:30%.