crop2dLayer

2-D crop layer

Description

A 2-D crop layer applies 2-D cropping to the input.

There are two inputs to this layer:

• 'in' — The feature map that will be cropped

• 'ref' — A reference layer used to determine the size, [height width], of the cropped output

Once you create this layer, you can add it to a layerGraph to make serial connections between layers. To connect the crop layer to other layers, call connectLayers and specify the input names. The connectLayers function returns a connected LayerGraph object ready to train a network.

Creation

Syntax

layer = crop2dLayer(Mode)

layer = crop2dLayer(Location)

layer = crop2dLayer(___,'Name',Name)

Description

layer = crop2dLayer(Mode) returns a layer that crops an input feature map, and sets the Mode property.

layer = crop2dLayer(Location) returns a layer that crops an input feature map using a rectangular window, and sets the Location property that indicates the position of the window.

layer = crop2dLayer(___,'Name',Name) creates a layer for cropping and sets the optional Name property.

Properties

Mode — Cropping mode

'centercrop' (default) | 'custom'

Cropping mode, specified as 'centercrop' or 'custom'.

Mode	Description
'centercrop'	The location of the cropping window is the center of the input feature map.
'custom'	The location of the cropping window is based on the Location property. This value is automatically set when the Location property is specified as a 2-element row vector.

Data Types: char

Location — Cropping window location

'auto' (default) | 2-element row vector

Cropping window location, specified as 'auto' or a 2-element row vector.

Location	Description
2-element row vector in the format [x y]	The upper-left corner of the cropping window is at the location [x y] of the input feature map. x indicates the location in the horizontal direction and y is the vertical direction.
'auto'	The cropping window is located at the center of the input feature map. This value is automatically set when the Mode property is specified as 'centercrop'.

Name — Layer name

"" (default) | character vector | string scalar

Layer name, specified as a character vector or a string scalar. For Layer array input, the trainnet, trainNetwork, assembleNetwork, layerGraph, and dlnetwork functions automatically assign names to layers with the name "".

The Crop2DLayer object stores this property as a character vector.

Data Types: char | string

NumInputs — Number of inputs

2 (default)

Number of inputs of the layer. This layer has two inputs.

Data Types: double

InputNames — Input names

{'in' 'ref'} (default)

Input names of the layer. This layer has two inputs, named 'in' and 'ref'.

Data Types: cell

NumOutputs — Number of outputs

1 (default)

This property is read-only.

Number of outputs from the layer, returned as 1. This layer has a single output only.

Data Types: double

OutputNames — Output names

{'out'} (default)

This property is read-only.

Output names, returned as {'out'}. This layer has a single output only.

Data Types: cell

Examples

Create 2-D Crop Layer

Create a 2-D crop layer and connect both of the inputs using a layerGraph object.

Create the layers.

layers = [
     imageInputLayer([32 32 3],'Name','image')
     crop2dLayer('centercrop','Name','crop')]

layers = 
  2x1 Layer array with layers:

     1   'image'   Image Input   32x32x3 images with 'zerocenter' normalization
     2   'crop'    Crop 2D       center crop

Create a layerGraph. The first input of crop2dLayer is automatically connected to the first output of the image input layer.

lgraph = layerGraph(layers)

lgraph = 
  LayerGraph with properties:

     InputNames: {'image'}
    OutputNames: {1x0 cell}
         Layers: [2x1 nnet.cnn.layer.Layer]
    Connections: [1x2 table]

Connect the image input layer to the "ref" input of the 2-D crop layer.

lgraph = connectLayers(lgraph,'image','crop/ref')  

lgraph = 
  LayerGraph with properties:

     InputNames: {'image'}
    OutputNames: {1x0 cell}
         Layers: [2x1 nnet.cnn.layer.Layer]
    Connections: [2x2 table]

Algorithms

Layer Input and Output Formats

Layers in a layer array or layer graph pass data to subsequent layers as formatted dlarray objects. The format of a dlarray object is a string of characters, in which each character describes the corresponding dimension of the data. The formats consists of one or more of these characters:

• "S" — Spatial

• "C" — Channel

• "B" — Batch

• "T" — Time

• "U" — Unspecified

For example, 2-D image data represented as a 4-D array, where the first two dimensions correspond to the spatial dimensions of the images, the third dimension corresponds to the channels of the images, and the fourth dimension corresponds to the batch dimension, can be described as having the format "SSCB" (spatial, spatial, channel, batch).

You can interact with these dlarray objects in automatic differentiation workflows such as developing a custom layer, using a functionLayer object, or using the forward and predict functions with dlnetwork objects.

This table shows the supported input formats of Crop2DLayer objects and the corresponding output format. If the software passes the output of the layer to a custom layer that does not inherit from the nnet.layer.Formattable class, or a FunctionLayer object with the Formattable property set to 0 (false), then the layer receives an unformatted dlarray object with dimensions ordered corresponding to the formats in this table. The formats listed here are only a subset. The layer may support additional formats such as formats with additional "S" (spatial) or "U" (unspecified) dimensions.

Input Format	Output Format
"SSCB" (spatial, spatial, channel, batch)	"SSCB" (spatial, spatial, channel, batch)
"SSC" (spatial, spatial, channel)	"SSC" (spatial, spatial, channel)

In dlnetwork objects, Crop2DLayer objects also support these input and output format combinations.

Input Format	Output Format
"SSCBT" (spatial, spatial, channel, batch, time)	"SSCBT" (spatial, spatial, channel, batch, time)
"SSCT" (spatial, spatial, channel, time)	"SSCT" (spatial, spatial, channel, time)

Extended Capabilities

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

GPU Code Generation
Generate CUDA® code for NVIDIA® GPUs using GPU Coder™.

Version History

Introduced in R2017b