## DESCRIPTION

This module counts the number of occurrences of data values from the input lattice, storing counts in an output lattice. The input lattice may be of any dimensionality and type. The output lattice is either a 1-D lattice, or an N-D lattice in Multi-Dimension Mode (see below).The Num Buckets Slider specifies how many buckets to divide the data into. In Multi-Dimension mode, it says how many buckets per dimension.

- One Channel Mode
- In One Channel mode, the Channel slider chooses which channel to display. If the slider value is greater then the last channel, then the last channel is selected. In All Channels mode, the selected channel is drawn in white. It has no effect in Multi-Dimension mode. In All Channels mode, the Histograms are drawn, in order, in red, green, cyan, yellow and magenta. If there are more than 5 channels, the colors are cycled. The selected channel, as mentioned, is the exception in that it is drawn in white.
The two text type-ins specify the minimum and maximum of the data range. If they are not specified, they are taken from the data. If they are specified, then anything lying below the range is put in the minimum bucket, and anything above the range is put in the maximum bucket. If neither is specified, and all the data is the same value, then the minimum becomes that value, and the maximum becomes the value plus one.

In One Channel and All Channels mode, the output is a 1-D lattice. If the input is an N vector, the output is also an N vector, where the array of first elements is the Histogram of the input's first elements, and so on.

Values outside the range are handled in one of two ways, depending on the selection of the Outliers widget. When Outliers is set to Add, values above the maximum are accumulated into the highest bucket and values below the minimum are accumulated into the lowest bucket. When Outliers is set to Omit, values outside the range are simply ignored. The Omit state permits a much closer look at a subset of the data values.

- Multi-Dimension Mode
- In Multi-Dimension Mode nothing is displayed, and the output lattice is in a different format. The output lattice has as its number of dimensions the number of data variables of the input lattice (for instance, an RGB image input will produce a 3-D output).
The output lattice is Num_Buckets on a side, with each side uniformly distributed in the range from the minimum to the maximum data value in that data channel on the input (for instance, an RGB image will produce an array of counts lying in 256 * 256 * 256 space). The count in the output lattice at point (i,j,k) is the number of input data values lying in the (i,j,k) portion of the data range space.

**Note**: the Histogram module is not present in every implementation of IRIS Explorer.

## INPUTS

**Port:** Data

**Type:** Lattice

This is the input lattice.

## WIDGETS

**Port:** Num Buckets

**Type:** Slider

This is the discretization level of data range space.

**Port:** Histogram Window

**Type:** Drawing Area

**Port:** Min Domain

**Type:** Text

This is the minimum data value to histogram. If left blank, it will be taken from the data.

**Port:** Max Domain

**Type:** Text

This is the maximum data value to histogram. If left blank, it will be taken from the data.

**Port:** Scale Window

**Type:** Drawing Area

**Port:** Channel

**Type:** Slider

This selects the channel of the lattice from which to take the histogram.

**Port:** Channel Mode

**Type:** Radio Box

**Menu Item:** One Channel

**Menu Item:** All Channels

**Menu Item:** Multi Dimension

This selects between one and multi-channel modes.

**Port:** Range Window

**Type:** Drawing Area

**Port:** Outliers

**Type:** Option Menu

**Menu Item:** Add

**Menu Item:** Omit

This selects whether to add out-of-range values to the extremal buckets or to omit them from the histogram.

## OUTPUTS

**Port:** Counts

**Type:** Lattice

**Constraints:** long

**Constraints:** uniform

This is the output histogram.

## KNOWN PROBLEMS

The scale ticks and numbers are not displayed if the number of buckets is large relative to the width of the large drawing (in pixels).## SEE ALSO

Graph, HistEqualImg, HistScaleImg, HistNormImg.[Documentation Home]

© The Numerical Algorithms Group Ltd, Oxford UK. 2000