基于局部敏感哈希算法的图像高维数据索引技术的研究

摘要

局部敏感哈希(LSH)算法是有效的高维数据索引方法之一，该算法成功地解决了“维数灾难”问题。分析了 LSH 算法中主要参数对索引性能的影响，在规模不同的图像数据集上应用了 LSH 算法，实验结果表明选择合适的参数时，其性能接近顺序搜索方法。%Locality sensitive hashing (LSH) was quite an efficient high dimensional data indexing method, which solved the problem on“disaster of dimension”. How the key parameters of LSH affect the performance of retrieval were analyzed, and LSH is also applied to different scale image datasets. The experimental results show that the performance of LSH is near to that of linear scan with the suitable parameters selected.