Texture feature extraction using the Sequency-ordered Complex Hadamard Transform

• المؤلف الرئيسي: Lee, Sin Yi.
• مؤلفون آخرون: Ng Boon Poh
• التنسيق: Final Year Project
• اللغة: English
• منشور في: 2011
• الموضوعات: DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
• الوصول للمادة أونلاين: http://hdl.handle.net/10356/45701

In the field of texture classification, signal processing is one of the most common methods used for texture feature extraction. In this paper, we compare the texture classification performance of the sequency-ordered complex Hadamard transform (SCHT) and its real and conjugate symmetric version (Real-CSSCHT) with other existing transforms such as discrete cosine transform (DCT), Walsh Hadamard transform (WHT) and the parametric Slant Hadamard transform (parametric SHT). In our experiments, feature vectors of different texture images were fed into the K-Nearest Neighbor (KNN) classifier to be trained and classified. Classification performance of each transform was analyzed based on factors such as classification accuracy and computational cost.