1 - 13 of 13 results

CIDRE / corrected intensity distributions using regularized energy minimization

A retrospective illumination correction method for optical microscopy. CIDRE is designed to correct collections of images by building a model of the illumination distortion directly from the image data. Larger image collections provide more robust corrections. CIDRE achieves correction quality similar to that of prospective methods. Unlike existing retrospective methods, CIDRE estimates both linear intensity gain function and an additive term. The key insight to our approach is that the distribution of intensities from a single location across many images is related to an underlying distribution common to all locations by a linear transform. This assumes that objects may appear anywhere in the image with equal probability.

Anisotropic Diffusion 2D

Serves for vector-valued image regularization. Anisotropic Diffusion 2D is a program based on variational methods and partial differential equations (PDEs). This tool consists in a single generic anisotropic diffusion equation that provides a simple interpretation of the regularization process in terms of local filtering with spatially adaptive Gaussian kernels. It also offers the possibility to specialize a generic expression into different regularization PDEs depending on different applications: flow visualization, image restoration, magnification, or inpainting.


Automates image analysis of estrogen receptor (ER), progesterone receptor (PR), and Ki-67 tissue sections. ImmunoRatio uses monoclonal antibodis 6F11, PgR636 and MIB-1 to detect respectively ER, PR and Ki-67. It is able to make blank field correction thank to the utilization of the Calculator Plus plugin. The tool segments the DAB- and hematoxylin-stained nuclei areas from a microscope image, calculates the labeling index, and generates a pseudo-colored result image matching of the segmentation.