摘要
基于内容的图像检索是图像检索中的一项重要方法,可是图像处理中存在着一个难题就是维数过大,处理效率不高。本文目的是探索对于图像在降维之后的特征向量能否可以直接进行图像匹配检索,这样处理既能解决维数问题,也可以提高处理的速度。本文在HSI模型的基础上采用LLE降维,分别对H、S、I进行降维。通过对降维后的特征向量直接进行匹配检索,争取提高图像处理速度,及图像检索精度。实验结果表明该算法在图像检索中是有效的。
CBIR (content-based image retrieval)is one of the important image retrieval method. But there is a problem in the image processing which is too large dimension and low processing efficiency. The purpose of this paper is to explore the possibility of the image retrieval by matching image feature vector directly after dimensionality reduction, which can not only solve the problem of dimension, it can also increase the processing speed. This paper, by using the HSI color model, reduce dimension of hue, saturation, intensity by LLE (Locally Linear Embedding) dimension reduction. We can take the retrieval of feature vector directly after the dimension reduction, in order to improve the image processing speed, and Image retrieval precision. The experimental results show that the algorithm is effective in image retrieval.
出处
《无线互联科技》
2013年第2期150-151,共2页
Wireless Internet Technology
