Label

Label connected components in binary image

• 

Libraries:
Computer Vision Toolbox / Morphological Operations

Description

The Label block finds and labels the connected components or objects in a 2-D binary image. The block ignores the background pixels in the input image and searches for connected components within remaining pixels. The block then labels all the pixels of each connected component with a unique number and computes the number of labeled objects.

Examples

Label Objects in Binary Image

Label objects in a binary image.

Open Model

Ports

Input

BW — Input binary image

vector | matrix

The input image must be a vector or matrix of binary values, each element representing a pixel of the image. The block categorizes pixels with a value of 0 as background pixels, searching for connected components within pixels with a value of 1.

Data Types: Boolean

Output

Label — Labeled objects

vector | matrix

Labeled objects in the input image, returned as a vector or matrix of nonnegative integers. The elements of the output vector or matrix labeled 0 represent the background, the elements labeled 1 represent the first object, the elements labeled 2 represent the second object, and so on. The size of this output matches the size of the BW input.

Dependencies

To enable this port, set the Output parameter to either Label matrix or Label matrix and number of labels.

Data Types: uint8 | uint16 | uint32

Count — Number of labeled objects

scalar

Number of labeled objects, returned as a scalar.

Dependencies

To enable this port, set the Output parameter to either Number of labels or Label matrix and number of labels.

Data Types: uint8 | uint16 | uint32

Parameters

Connectivity — Type of pixel connectivity

8 (default) | 4

Specify how the block defines which pixels are connected to each other. If you want to connect adjacent pixels in only the horizontal and vertical directions, select 4. If you want to connect adjacent pixels in the horizontal, vertical, and diagonal directions, select 8. These figures show how the block defines a connected component for 4-connected and 8-connected pixels. The block considers the white pixels marked by black circles to be connected to each other.

Type of Connectivity	Connected Pixels
4-connected pixels
8-connected pixels

Output — Block output

Label matrix and number of labels (default) | Label matrix | Number of labels

Specify the block output as one of these options:

• Label matrix — The block outputs the labeled objects at the Label port.

• Number of labels — The block outputs the number of labeled objects at the Count port.

• Label matrix and number of labels — The block outputs both the labeled objects and number of labeled objects to the Label and Count ports, respectively.

Output data type — Data type of output

Automatic (default) | uint8 | uint16 | uint32

Specify the data type of the outputs at the Label and Count ports. When you specify this parameter as Automatic, the block calculates the maximum number of objects that can fit inside the image based on the image size and the specified connectivity. Using this calculation, it chooses the output data type with the minimum value that guarantees unique region labels.

If label exceeds data type size, mark remaining regions using — Overflow mode of output labels

maximum value of the output data type (default) | zero

Use this parameter to specify the behavior of the block when the block finds more objects than the maximum possible value the output data type can represent. The block labels the remaining objects with the value determined by one of these options:

• maximum value of the output data type — The block labels the remaining objects with the maximum value of the output data type.

• zero — The block labels the remaining objects with 0.

Dependencies

To enable this parameter, set the Output data type parameter to either uint8 or uint16.

Block Characteristics

Data Types	Boolean | integer
Multidimensional Signals	no
Variable-Size Signals	yes

Extended Capabilities

C/C++ Code Generation
Generate C and C++ code using Simulink® Coder™.

Version History

Introduced before R2006a