【作者】 王丽荣； 申铉国； 王延杰；

【Author】 Wang Lirong~(1,2) Shen Xuanguo~1 Wang Yanjie~3 ~1(Communication Engineer College, Jilin University, Changchun 130025, China) ~2(Electronics and Information Engineer College, Changchun University, Changchun 130022, China) ~3(Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Science, Changchun 130033, China)

【机构】 吉林大学通信工程学院； 中国科学院长春光机与物理研究所；

【摘要】 提出一种利用整数型提升小波变换提取目标子图像的方法,选择整数Harr小波做提升模式的整数到整数的小波变换。在变换中被训练的任意参数同被查询图像的低频分量的消失标准相适应。训练参数包含被访问图像特征。把具有这些参数的提升小波变换作用到参考图像上,用所提出的消失标准提取目标子图像。对一幅速拍图像做仿真,结果表明此种算法能准确提取目标图像。更多还原

【Abstract】 A novel approach to extract target image from a large image using integer lifting wavelet transform is presented. Integer to integer lifting wavelet transform based on integer-type of Harr wavelet transform is selected. Free parameters are trained according to a vanishing criterion for low frequency components of query images. The trained parameters have characteristics of the query images. It applies a lifting wavelet transform with such parameters to a reference image and vanishing criterion to extraction. Result of simulation has illustrated that the algorithm is efficient to extract target subimage.更多还原