This paper describes the automatic leaf recognition based on feature level fusion and score level fusion of vein orientation angles, GLCM, SIFT, SURF as a features. However, to obtain the sophisticated leaf recognition, the system must be undergo through numerous difficulties such as intra and inter-class variations in plants and defining the proper local and global image descriptors which can deal with the color, shape and textual, information for the classification. Selection of the meticulous features plays key role in designing the best classification system. In this paper we proposed multi-modal plant classification where several components are fused together for a more precise classification. The results shows that the proposed system for feature level fusion achieved 93.72% GAR with 6.27% of EER and for score level fusion system achieves 97.13% GAR and 2.86% EER. It is found that the performance of the classification has been increased by 3.79% of EER when score level fusion applied to the system.
CITATION STYLE
Salve, P., Yannawar, P., & Sardesai, M. (2019). Evaluation and Analysis of Plant Classification System Based on Feature Level Fusion and Score Level Fusion. In Communications in Computer and Information Science (Vol. 1037, pp. 453–470). Springer Verlag. https://doi.org/10.1007/978-981-13-9187-3_41
Mendeley helps you to discover research relevant for your work.