IDENTIFIKASI TANAMAN HIAS DAUN MENGGUNAKAN CITRA DAUN

• Main Author: KADIR, ABDUL
• Format: Theses and Dissertations NonPeerReviewed
• Published: [Yogyakarta] : Universitas Gadjah Mada 2012
• Subjects: Plant Biology; Neural Evolutionary and Fuzzy Computation; Signal Processing
• Online Access: https://repository.ugm.ac.id/99898/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=56081
• Institution: Universitas Gadjah Mada

Several methods to identify plants by using a leaf of plant have been proposed by several researchers. Commonly, the methods did not capture color, because color was not recognized as an important aspect to the identification. The main reason was caused by a fact that they used green colored leaves as samples. However, for foliage plants-plants with colorful leaves, fancy patterns in their leaves, and interesting plants with unique shape-color and also texture could not be neglected. Therefore, combination of shape, color, texture features, and other attribute contained on the leaf is very useful in leaf identification. In this research, shape and vein, color, and texture features were incorporated to classify a leaf and to retrieve plants which have most similar with leaf of query. The plant retrieval was intended to give suggestion to users five plants that may help them to identify the leaf when classification process failed. Besides, this research also tried to develop an algorithm to find all leaves of plants in a database that have certain dominant color, where the dominant color is determined by human perception, such as green or dark green. In this riset, a neural network called Probabilistic Neural network (PNN) was used as a classifier, the Euclidean distance with weighting coefficients of shape, vein, and color features was used to retrieve leaves and fuzzy approach was used to retrieve leaves which fulfill the dominant color. Moreover, the research also included Principal Component Analysis (PCA) in order to reduce features. The experiments was accomplished by using Flavia and Foliage datasets. Flavia is a dataset came from other researchers that was commonly used in leaf retrievals. It contains 32 kinds of plants with all of them have green color leaves. Meanwhile, Foliage are dataset with various color leaves that were prepared by authors. Foliage contains 60 kinds of foliage plants. The results shows that the method for classification gave average accuracy of 94.6875% when it was tested on Flavia dataset. It means that the method gave better performance compared to the original work. The identification system gave average accuracy of 92.9167% for 60 kinds of foliage plants by using 52 features. By reducing features to 20 using PCA, the performance increased to 94.5%. The performance was still more than 80% when the total features was compressed to 10.