摘要
为了提高医院图像检索系统对医学图像的检索精度,对基于内容的图像检索方法进行了全面的研究。深入地分析了小波变换和游程长度矩阵在图像检索中的优点,创造性地提出了一种融合小波分解和游程长度矩阵的检索算法。该算法充分利用小波变换可以对图像进行的多尺度分析的优点,以及游程长度矩阵可以描述图像内灰度与游程长度分布规律的优点,使用高斯分布将小波分解后的各尺度图像游程长度矩阵特征进行合理融合。将仿真算法和其它算法进行比较,比较结果表明该方法能有效地提高图像检索的精确度。
In order to improve the retrieval accuracy in image retrieval system for hospital medical image, the content-based image retrieval method is studied comprehensive. The advantages of wavelet transform and run-length matrix in image retrieval are analyzzed deeply. A algorithm with wavelet decomposition and the run-length matrix is proposed. It make full use of the advantages of multiscale analysis, as well as the advantages of run-length matrix can be described within the gray-scale images. And use Gaussian distribution to the wavelet decomposition scale image after run-length matrix features integration reasonable. And simulation algorithms and other algorithms are compared, the result show that the method effectively improve the accuracy of image retrieval.
出处
《计算机工程与设计》
CSCD
北大核心
2010年第8期1771-1774,共4页
Computer Engineering and Design
关键词
小波变换
多尺度分析
小波分解
游程长度矩阵
图像特征
图像检索
wavelet transform
multi-scale analysis
wavelet decomposition
run length matrix
image feature
image retrieval