| Summary: | 碩士 === 南台科技大學 === 資訊工程系 === 102 === In today's society, people are often not so interested in the study of plants, and since they don't know the name of the plant and how to observe plants, it may cause problems of recognizing them. The purpose of this study is to develop a Plant Feature Search Systems, in the situation of not knowing the names of the plants, it leads users through observing the feature's of plants and filling the results in order. The system will find out the plant from the database that will fit in with the features base on the features that the users filled in. But even if the system has a guiding function, users may still filled in the wrong features since general public are not experts in the field of plants. Therefore, this study will be based on fuzzy functions, by setting up a plant searching system that has the ability to accept filling wrong features.
The plant searching system of this research will be aimed at the simulation by searching data of a single plant from the considerable quantities. To search up, the system uses a query based on 12 kinds of plant characteristics by Leaf-base, Leaf-apex, Leaf-shape, Leaf-phyllotaxy, Leaf-parallel, Leaf-margin, Leaf-size, Tree-dwell, Flower-inflorescence, Flower-shape, Flower-color and Flower-florescence characteristics parts,etc. By entering the above feature parts that the users observed, the system will find out the right plant according to the comparison method. The data analyze of this research is to enter the feature data that will be effected by the accuracy rate and using strict standards to test the search results. Moreover, the tests results show that the system's feature options are quite acceptable towards wrong features. Users selecting the site from the 12 kinds of features are allowed to choose 4-5 kinds of wrong features, yet can still maintain a high accuracy of query results. Hence, proving this system research is effective.
|
|---|
