This page links to the detailed documentation on several modules
If you are not a java programmer, but you think tango lacks a module (pre-filter, post-filter, segmentation algorithm, thresholding method or measurement) don't hesitate to contact us, we will do our best to include it.

Processing Modules


Pre-filters are classicaly used to reduce noise in your images, most commonly used are gaussian filter and median filter. They should be used first in processing chains. Other filters can on the contrary enhance your signal like TopHat filtering, LoG3D or Hessian filters (in Image Features) to detect spots.
  • Classical filters in 3D (tophat, median, mean, minimum, maximum, variance, sobel, maximum local) : FastFilters3D
  • Other filters in 3D (gaussian, Laplacian of Gaussian, pure denoise, bandpass filter ) : MiscFilters3D
  • Histogram transformations like invert, normalize, scale
  • Translate: from TransformJ
  • Test Threshold Method is just useful to check the returned value of a threshold method and should not be included in a processing chain.
  • Image Features (from FeatureJ) can be useful within several algorithm when filters are asked (for instance in Seeded Watershed 3D where the propagation map can be chosen). In the case of spot detection in very noisy images, a good strategy is to apply thresholding to the inverted hessian transform.
  • Subtract Gaussian: removes background by subtracting a gaussian-blured version of tje image. The radius of the gaussian should be higher than the size of the objects to be detected. useful to correct inhomogeneous illumination.

Segmentation Algorithms

  • Classical segmentation algorithms are based on thresholding. Two plugins are using this technique Simple Segmenter and Hysteresis Segmenter, they are document on the Thresholding page. They can be used both for segmenting nuclei or structures.

Nucleus Segmentation

  • Nucleus Edge Detector : this algorithm is design to segment nuclei, with a good precision on the edge even in case of nucleus-to-nucleus intensity variation on a given image. It is not designed to split nuclei in close contact.

Structure Segmentation


  • Segmented objects can be corrected using morphological filters
  • Size and Edge Filter : remove objects with innapropriate size, or if they touch edges by a certain number of pixels.
  • Watershed 2D : ImageJ function watershed to separate fused objects (only works with 2D images).
  • Region Adjustement : apply a local threshold on each spot
  • Erase Spots : erase spots based on intensity-based criterion
  • Merge Regions : merge connected regions

Quantitative Analysis Modules

Measurement are grouped in serveral categories depending on their output.
  • Measurements on segmented objects will yield one value per object of the selected structure.
  • Measurement between two segmented objects will return one value per pair of objects. If two different segmented structures are selected, containing respectively n1 and n2 objects, those measurement will produce an array of n1 * n2 values; If the same structure is selected, containing n objects, they will produce an array of n * ( n - 1 ) / 2 values.
  • Measurement on structures return any nupmber of value per nucleus, and thus are not indexed

Measurement on segmented Objects

Measurement between two segmented objects

  • Distances
  • Object Colocalization: computes the percentage of overlap between each of the objects of the first selected structure and each of the objects of the 2nd selected structure

Measurement on Structures