In this paper, we propose the so-called continuous Fresnel-wavelet combinatorial transform which means that the mother wavelet undergoes the Fresnel transformation. This motivation can let the mother-wavelet-state its...In this paper, we propose the so-called continuous Fresnel-wavelet combinatorial transform which means that the mother wavelet undergoes the Fresnel transformation. This motivation can let the mother-wavelet-state itself vary from ｜ψ〉 to Ftr, s ｜ψ〉, except for variation within the family of dilations and translations. The Parseval＇s equality, admissibility condition and inverse transform of this continuous Fresnel-wavelet combinatorial transform are analysed. By taking certain parameters and using the admissibility condition of this continuous Fresnel-wavelet combinatorial transform, we obtain some mother wavelets. A comparison between the newly found mother wavelets is presented.展开更多
针对遥感图像数据产生和传输过程中的去噪需求,作者提出一种基于量子衍生研究方向的遥感图像去噪算法,并通过实验方法对该算法的精度提升幅度进行了验证。该算法的实验验证过程包括:首先,对已知图像进行噪声添加,对添加噪声图像数据进...针对遥感图像数据产生和传输过程中的去噪需求,作者提出一种基于量子衍生研究方向的遥感图像去噪算法,并通过实验方法对该算法的精度提升幅度进行了验证。该算法的实验验证过程包括:首先,对已知图像进行噪声添加,对添加噪声图像数据进行对数变换后再进行双密度双树复小波变换;然后,利用贝叶斯理论对变换后的系数进行去噪,再采用最大后验概率(Maxaposterior,MAP)估计双树复小波的方差;最后,通过双树复小波反变换得到去噪后的图像数据,将这些变换过程中得到数据结果与源图像数据进行对比分析,得出该算法的实验结果。实验结果证明该算法的峰值信噪比(Peak Signal to Noise Ratio,PSNR)比其他文献提高2 dB以上,边缘保持指数(Edge Preservation Index,EPI)也高于普通方法的0.1以上。展开更多
基金supported by the Startup Research Fund for Introducing Talents of Anhui Polytechnic University (Grant No. 2009YQQ006)the Research Foundation of the Education Department of Anhui Province of China (Grant No. KJ2011B031)
文摘In this paper, we propose the so-called continuous Fresnel-wavelet combinatorial transform which means that the mother wavelet undergoes the Fresnel transformation. This motivation can let the mother-wavelet-state itself vary from ｜ψ〉 to Ftr, s ｜ψ〉, except for variation within the family of dilations and translations. The Parseval＇s equality, admissibility condition and inverse transform of this continuous Fresnel-wavelet combinatorial transform are analysed. By taking certain parameters and using the admissibility condition of this continuous Fresnel-wavelet combinatorial transform, we obtain some mother wavelets. A comparison between the newly found mother wavelets is presented.
文摘针对遥感图像数据产生和传输过程中的去噪需求,作者提出一种基于量子衍生研究方向的遥感图像去噪算法,并通过实验方法对该算法的精度提升幅度进行了验证。该算法的实验验证过程包括:首先,对已知图像进行噪声添加,对添加噪声图像数据进行对数变换后再进行双密度双树复小波变换;然后,利用贝叶斯理论对变换后的系数进行去噪,再采用最大后验概率(Maxaposterior,MAP)估计双树复小波的方差;最后,通过双树复小波反变换得到去噪后的图像数据,将这些变换过程中得到数据结果与源图像数据进行对比分析,得出该算法的实验结果。实验结果证明该算法的峰值信噪比(Peak Signal to Noise Ratio,PSNR)比其他文献提高2 dB以上,边缘保持指数(Edge Preservation Index,EPI)也高于普通方法的0.1以上。
