Batik Images Retrieval Using Pre-trained model and K-Nearest Neighbor

Batik is an Indonesian cultural heritage that should be preserved. Over time, many batik motifs have sprung up, which can lead to mutual claims between craftsmen. Therefore, it necessary create a system measure the similarity of motif. This research focused on making Content-Based Image Retrieval (CBIR) images. The dataset used in this big data authors transfer learning several pre-trained models and Convolutional Neural Network (CNN) Autoencoder from previous studies extract features all images database. extracted calculate Euclidean distance query database retrieve image closest will retrieved according number r, namely 3, 5, 10, or 15. Before retrieved, retrieval re-ranked with K-Nearest Neighbor (KNN), classifies image. results study prove MobileNetV2 + KNN best model terms Batik, followed by InceptionV3 VGG19 as second third ranks. Moreover, CNN InceptionResNetV2 are ranked fourth fifth. In study, was also found use re-ranking increase precision value 0.00272. For further research, deploying these models, especially for approach seeing major impact craftsmanship decreasing motif plagiarism.