FeatureJ: Structure

General Description

This plugin enables one to compute, for all elements (pixels/voxels) in an image, the eigenvalues of the so-called structure tensor, which can be used for example in texture analysis [1] and image enhancement filtering [2].

Dialog Description

Largest/Middle/Smallest eigenvalue of structure tensor. After computation of the structure tensor the resulting eigenvalues are ordered for each image element (pixel/voxel). All largest eigenvalues are put in a separate image, as are all smallest (and middle, in 3D). These options determine which of these so-called eigenimages will be displayed by the plugin. Due to the positive-semidefiniteness of the structure tensor, the eigenvalues are always larger than or equal to zero. If the size of the image is one in the z-dimension (a single slice), the plugin computes 2D-tensor eigenvalues, otherwise it computes 3D-tensor eigenvalues (for every time frame and channel in a 5D image).

Smoothing scale. The smoothing scale is equal to the standard deviation of the Gaussian derivative kernels used for computing the elements of the structure tensor and must be larger than zero. See the algorithmic details in the description of the Derivatives dialog for boundary conditions in setting this parameter. If physically isotropic Gaussian image smoothing is to be applied (which can be specified in the Options dialog), then in each dimension the scale is divided by the sampling interval in that dimension (the pixel width/height/depth as specified in ImageJ > Image > Properties).

Integration scale. The integration scale is equal to the standard deviation of the Gaussian kernel used for smoothing the elements of the structure tensor and must be larger than zero. See the algorithmic details in the description of the Derivatives dialog for boundary conditions in setting this parameter. If physically isotropic Gaussian image smoothing is to be applied (which can be specified in the Options dialog), then in each dimension the scale is divided by the sampling interval in that dimension (the pixel width/height/depth as specified in ImageJ > Image > Properties).

References