Combining Filter Bank and KSH for Image Retrieval in Bone Scintigraphy.
ICONIP (1)(2020)
摘要
Bone scintigraphy is widely used to diagnose bone tumor and metastasis. Accurate bone scan image retrieval is of great importance for tumor metastasis diagnosis. In this paper, we propose a framework to retrieve images by integrating the techniques of texture feature extraction and supervised hashing with kernels (KSH). We first use a filter bank to extract the texture features. Then KSH is used to train a set of hashing functions with constructed features, which can convert images to hashing codes. We can obtain the most similar retrieval images by comparing Hamming distance of these hashing codes. We evaluate the proposed framework quantitatively on the testing dataset and compare it with other methods.
更多查看译文
关键词
Bone scintigraphy, Image retrieval, Filter bank, KSH
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络