Threshold Class Reference

3 min read time

Applies to Linux

• roccv
• Threshold

Public Member Functions | List of all members

roccv::Threshold Class Referencefinal

Class for managing the Threshold operator. More...

#include <op_thresholding.hpp>

Inheritance diagram for roccv::Threshold:

Public Member Functions
void	operator() (hipStream_t stream, const roccv::Tensor &input, const roccv::Tensor &output, const roccv::Tensor &thresh, const roccv::Tensor &maxVal, eDeviceType device=eDeviceType::GPU)
	Construct a new Threshold object. The object can be used to choose a global threshold value that is the same for all pixels across the image. More...
	Threshold (eThresholdType threshType, int32_t maxBatchSize)
	Constructs a Threshold object. More...
	~Threshold ()
	Destroy the Threshold object. More...
Public Member Functions inherited from roccv::IOperator
virtual	~IOperator ()

Detailed Description

Class for managing the Threshold operator.

Constructor & Destructor Documentation

◆ Threshold()

roccv::Threshold::Threshold	(	eThresholdType	threshType,
		int32_t	maxBatchSize
	)

Constructs a Threshold object.

◆ ~Threshold()

roccv::Threshold::~Threshold

(

)

Destroy the Threshold object.

Member Function Documentation

◆ operator()()

void roccv::Threshold::operator()	(	hipStream_t	stream,
		const roccv::Tensor &	input,
		const roccv::Tensor &	output,
		const roccv::Tensor &	thresh,
		const roccv::Tensor &	maxVal,
		eDeviceType	device = eDeviceType::GPU
	)

Construct a new Threshold object. The object can be used to choose a global threshold value that is the same for all pixels across the image.

Limitations:

Input: Supported TensorLayout(s): [HWC, NHWC] Channels: [1, 3, 4] Supported DataType(s): [U8]

Output: Supported TensorLayout(s): [HWC, NHWC] Channels: [1, 3, 4] Supported DataType(s): [U8]

Input/Output dependency

  Property      |  Input == Output
 -------------- | -------------
  Data Layout   | Yes
  Data Type     | Yes
  Number        | Yes
  Channels      | Yes
  Width         | Yes
  Height        | Yes

threshold Tensor

 Must be of layout type 'NW' (dim = 2) with N = maxBatchSize
 and W = 1 Data Type must be an 8bit unsigned integer (DATA_TYPE_U8).

maxval Tensor

 Must be of layout type 'NW' (dim = 2) with N = maxBatchSize and
 W = 1 Data Type must be 8bit unsigned integer (DATA_TYPE_U8).

thresh and maxVal are expected to be contiguous in memory.

Current Supported Threshold Types (threshType)

• THRESH_BINARY
• THRESH_BINARY_INV
• THRESH_TRUNC
• THRESH_TOZERO
• THRESH_TOZERO_INV

Parameters

[in]	stream	The HIP stream to run this operation on.
[in]	input	Input tensor with image batch data
[out]	output	Output tensor for storing modified image batch data
[in]	thresh	thresh an array of size maxBatch that gives the threshold value of each image.
[in]	maxVal	maxval an array of size maxBatch that gives the maxval value of each image, used with the THRESH_BINARY and THRESH_BINARY_INV thresholding types.
[in]	device	The device which this operation should run on. (Default: eDeviceType::GPU)

---

The documentation for this class was generated from the following file:

• /home/docs/checkouts/readthedocs.org/user_builds/advanced-micro-devices-roccv/checkouts/latest/include/op_thresholding.hpp