hi_transform_config

Description

Defines the affine transformation, perspective transformation, or remapping configuration.

Prototype

hi_transform_config = {"interpolation": interpolation,
                       "border_type": hi_vpc_bord_type,
                       "scalar_value": hi_vpc_scalar}

Members

Member	Description
interpolation	Int. The following resizing algorithms are supported: For the Atlas 200I/500 A2 inference products, the following resizing algorithms are supported: 0: the conventional bilinear algorithm 1: the conventional nearest neighbor algorithm. For the Atlas A2 training products/Atlas A2 inference products, the following resizing algorithms are supported: 0: the conventional bilinear algorithm 1: the conventional nearest neighbor algorithm. For the Atlas A3 training products/Atlas A3 inference products, the following resizing algorithms are supported: 0: the conventional bilinear algorithm 1: the conventional nearest neighbor algorithm.
border_type	Int, border making type. For the Atlas 200I/500 A2 inference products, the following options are supported: 0: fixed value of the register. 1: border replication mode. For the Atlas A2 training products/Atlas A2 inference products, the following options are supported: 0: fixed value of the register. 1: border replication mode. For the Atlas A3 training products/Atlas A3 inference products, the following options are supported: 0: fixed value of the register. 1: border replication mode.
scalar_value	Dictionary, pixel value for image padding. Only three channels are supported. For the Atlas 200I/500 A2 inference products, if the input image format is YUV400, val[0] stores the Y component. val[1], val[2], and val[3] are reserved and need to be initialized to 0. For the Atlas A2 training products/Atlas A2 inference products, if the input image format is YUV400, val[0] stores the Y component. val[1], val[2], and val[3] are reserved and need to be initialized to 0. For the Atlas A3 training products/Atlas A3 inference products, if the input image format is YUV400, val[0] stores the Y component. val[1], val[2], and val[3] are reserved and need to be initialized to 0.

Member

Description

interpolation

Int. The following resizing algorithms are supported:

For the Atlas 200I/500 A2 inference products, the following resizing algorithms are supported:

0: the conventional bilinear algorithm
1: the conventional nearest neighbor algorithm.

For the Atlas A2 training products/Atlas A2 inference products, the following resizing algorithms are supported:

0: the conventional bilinear algorithm
1: the conventional nearest neighbor algorithm.

For the Atlas A3 training products/Atlas A3 inference products, the following resizing algorithms are supported:

0: the conventional bilinear algorithm
1: the conventional nearest neighbor algorithm.

border_type

Int, border making type.

For the Atlas 200I/500 A2 inference products, the following options are supported:

0: fixed value of the register.
1: border replication mode.

For the Atlas A2 training products/Atlas A2 inference products, the following options are supported:

0: fixed value of the register.
1: border replication mode.

For the Atlas A3 training products/Atlas A3 inference products, the following options are supported:

0: fixed value of the register.
1: border replication mode.

scalar_value

Dictionary, pixel value for image padding. Only three channels are supported.

For the Atlas 200I/500 A2 inference products, if the input image format is YUV400, val[0] stores the Y component. val[1], val[2], and val[3] are reserved and need to be initialized to 0.

For the Atlas A2 training products/Atlas A2 inference products, if the input image format is YUV400, val[0] stores the Y component. val[1], val[2], and val[3] are reserved and need to be initialized to 0.

For the Atlas A3 training products/Atlas A3 inference products, if the input image format is YUV400, val[0] stores the Y component. val[1], val[2], and val[3] are reserved and need to be initialized to 0.

Parent topic: Image Processing