ENSIKLOPEDIA TANAMAN MENGGUNAKAN PENCOCOKAN GAMBAR PADA APLIKASI BERBASIS WEB

Oleh :  Briant Stevanus, Agung Wijayakusuma, Christian, Angry Ronald Adam, dan Angry Ronald Adam

The intention in developing Plant Encyclopedia with Image Matching on the Web Based Application is clear. This new application is intended to search data by matching various types of image components, in example like shape based and color based specifically thresholding. This application combined these two image component and blend it with shape and edge detection method. This application is using Speeded Up Robust Features (SURF) algorithm with some considerations and well considerate plan. SURF algorithm are divided into four main stages, the first is retrieve image information and turn it into matrix using integral implementation, this step called image retrieval. The next step is defining objects found in image that have been interpreted into matrix and detect every object using high edge detection method or Fast Hessian method. The third, it came with positioning the objects that already detected in previous stage, in this stage the SURF algorithm searching the most compatible objects, and then the last is search the image orientation and matching it one to another using K-means algorithm. With SURF algorithm, this application could recognize the same object in different angle and in different color.

 Keyword: SURF, Image Matching, Encyclopedia