此文件包括三个关于图像显著性的代码,每个模型单独成一个文件夹,并用其模型名称命名,三个模型分别是ITTi模型、GBVS模型和SR模型,这三个模型都是使用MATLAB实现的,下载后可以直接使用MATLAB运行,另文件附有一个链接。点赞(0) 踩踩(0) 反馈 所需:30 积分 电信网络下载 ...
资源大小9.2MB,文件格式.zip,此文件包括三个关于图像显著性的代码,每个模型单独成一个文件夹,并用其模型名称命名,三个模型分别是ITTi模型、GBVS模型和SR模型,这三个模型都是使用MATLAB实现的,下载后可以直
spectral residual (SRphase spectrum of Fourier transform (PFTphase of quaternion Fourier transform (PQFTpulsed discrete cosine transform (PCTfrequency domain division normalization (FDNpatch FDN (PFDNpulsed PCA (principal components analysisamplitude spectrum of quaternion Fourier transform (AQFT...
Existing leading methods for spectral reconstruction (SR) focus on designing deeper or wider convolutional neural networks (CNNs) to learn the end-to-end mapping from the RGB image to its hyperspectral image (HSI). These CNN-based methods achieve impressive restoration performance while showing limit...
Finally, each SpecProcessBlock is connected via residual connections. It takes a 1D feature representation \(X_{l-1}^{1D}\) from the previous layer, processes it to generate \(\text {SpecProcessBlock}\) \((X_{l-1}^{1D})\), and then adds the processed features to the original inpu...
'quat:fft:eigensr': The principle of 'fft:residual' transferred to 'quat:fft:eigenpqft'. 'quat:dct' Converts the image into a quaternion-based representation, uses quaternion DCT/IDCT operations. 'quat:dct:fast' Same as 'quad:dct', but with a fixed image ...
This paper proposes a novel end鈥揺nd deep learning-based network named Separable-Spectral and Inception Network (SSIN) for HSI SR. In SSIN, the feature extraction module independently extracts features of each band image, and then these features are fused together to further exploit residual ...
If compared to the residual distributions associated to a ground motion pre- diction equation previously derived using the same Fourier amplitude spectra, the source parameter and the empirical site amplification effects correlate well with the inter-event and inter-station residuals, respectively. ...
下面是SR算法主要的步骤 x只是输入的序列数据,一般是滑动窗口数据。首先通过傅里叶变换之后计算振幅谱A(f),然后计算相位谱P(f)(复数x+i*y的相位是arctan(y/x)),然后对振幅谱做Log得到L(f),AL(f)是L(f)进行均值滤波之后的结果,R(f)就是Spectral Residual谱,再进行一个傅里叶反变换就可以得到显著区域,...
Residual learning alleviates the degradation prob- lem due to increased depth, and simplifies the learning task, which improves network convergence. 尽管这些进步提高了性能并且现在在 SR 网络中很常见,但这些方法仍然存在严重缺陷(见图 1)。已经证明神经网络表现出对低频信号的偏向。图 2 说明了这一点。在...