作者: Luefeng Chen , Min Wu , Witold Pedrycz , Kaoru Hirota
DOI: 10.1007/978-3-030-61577-2_8
关键词:
摘要: The Dempster-Shafer (D-S) theory based on multi-SVM to deal with multimodal gesture images for intention understanding is proposed, in which the Sparse Coding (SC) Speeded-Up Robust Features (SURF) are used feature extraction of depth and RGB image. Aiming at problems small sample, high dimensionality redundancy image data, we use SURF algorithm extract features original image, then perform their Coding, means that subjected two-dimensional reduction. dimensionally reduced by classification.