"myHerbs": A mobile based application for herbal leaf recognition using sift / Nur Nabilah Abu Mangshor ...[et al.]

• Main Authors: Abu Mangshor, Nur Nabilah (Author), Abdul Rahman, Mohamed Al Arabee (Author), Sabri, Nurbaity (Author), Ibrahim, Shafaf (Author), Ibrahim, Zaidah (Author), Shari, Anis Amilah (Author)
• Format: Book
• Published: Universiti Teknologi MARA Cawangan Pahang, 2020-09.
• Subjects: Article; PeerReviewed
• Online Access: Link Metadata
• Notes: https://ir.uitm.edu.my/id/eprint/46086/1/46086.pdf
• Keywords: Mobile computing; Algorithms; Organic chemistry

Herbs are plant with exquisite or sweet-smelling properties that been widely used since ancient times and are still used until today. Herbs generally refers to the leafy green, which are some of them have the same appearance, color and shape. Due to that, most of the ordinary people have trouble in recognizing the herbal species because of the similar features and appearances of the herbs leaf. In addition, the complexity of the structure of the herbs leaf itself contributes to the difficulty in recognizing its species. Other than that, botanist also had spent a lot of time to examine herbal species and classify them into group. Hence, this study proposed an automated mobile-based application for herb leaf recognition, "myHerbs. This study covers two species of the local herbs. The species of the herbs used in this study are Basil (Selasih) and Centella (Pegaga). All images used in this study are self-collected. Scale Invariant Feature Transform (SIFT) algorithm is used for extracting features from the herbs leaf and Fast Library for Approximate Nearest Neighbors (FLANN) algorithm is used for the classification purpose. 55 images have been evaluated for the testing purpose and the accuracy rate of 74.55% is achieved. The outcome of this study is believed to help the botanist and people in recognizing herbs species. In addition, it also contributes to the exploration and implementation of learning algorithm in mobile-based application.