4.21. ImagePreProcess

General preprocessing interface, internally implemented using thread pool.

4.21.1. Constructor ImagePreProcess()

interface:

ImagePreProcess(
    int batch_size,
    sail_resize_type resize_mode,
    int tpu_id=0,
    int queue_in_size=20,
    int queue_out_size=20,
    bool use_mat_flag=false);

Parameters:

batch_size: int

Input parameter. The batch size of the output results.

resize_mode: sail_resize_type

Input parameter. Internal scale transformation method.

tpu_id: int

Input parameter. The tpu id used, the default is 0.

queue_in_size: int

Input parameter. Enter the maximum length of the image queue buffer. The default is 20.

queue_out_size: int

Input parameter. The maximum length of the output Tensor queue cache, the default is 20.

use_mat_output: bool

Input parameter. Whether to use Mat of OPENCV as the output of the image, the default is False, not used.

4.21.2. SetResizeImageAtrr

Set the properties of the image resizing.

Interface:

void SetResizeImageAtrr(
    int output_width,
    int output_height,
    bool bgr2rgb,
    bm_image_data_format_ext  dtype);

parameters:

output_width: int

Input parameter. Image width after scale transformation.

output_height: int

Input parameter. Image height after scale transformation.

bgr2rgb: bool

Input parameter. Whether to convert images with BGR to GRB.

dtype: ImgDtype

Input parameter. The data type after image scale conversion, the current version only supports BM_FLOAT32, BM_INT8, BM_UINT8. Can be set according to the input data type of the model.

4.21.3. SetPaddingAtrr

Setting the properties of Padding only takes effect when resize_mode is BM_PADDING_VPP_NEAREST, BM_PADDING_TPU_NEAREST, BM_PADDING_TPU_LINEAR, or BM_PADDING_TPU_BICUBIC.

interface:

void SetPaddingAtrr(
    int padding_b=114,
    int padding_g=114,
    int padding_r=114,
    int align=0);

Parameters: * padding_b: int

Input parameter. The b channel pixel value to be pdding, default is 114.

padding_g: int

Input parameter. The g channel pixel value to be pdding, default is 114.

padding_r: int

Input parameter. The r channel pixel value to be pdding, default is 114.

align: int

Input parameter. Image padding is position, 0 means filling from the upper left corner, 1 means center filling, default is 0.

4.21.4. SetConvertAtrr

Set the properties of the linear transformation.

Interface:

int SetConvertAtrr(
   const std::tuple<
       std::pair<float, float>,
       std::pair<float, float>,
       std::pair<float, float>> &alpha_beta);

Parameters:

alpha_beta: (a0, b0), (a1, b1), (a2, b2)。输入参数。

a0 describes the linear transformation coefficient of the 0th channel;

b0 describes the linear transformation offset of the 0th channel;

a1 describes the linear transformation coefficient of the first channel;

b1 describes the linear transformation offset of the first channel;

a2 describes the linear transformation coefficient of the second channel;

b2 describes the linear transformation offset of the second channel;

Returns:

If the setting is successful, 0 is returned. If other values are set, the setting fails.

4.21.5. PushImage

Send data.

interface:

int PushImage(
    int channel_idx,
    int image_idx,
    BMImage &image);

Parameters:

channel_idx: int

Input parameter. Enter the channel number of the image.

image_idx: int

Input parameter. Enter the number of the image.

image: BMImage

Input parameter. The input image.

返回值说明:

Returns 0 if set successfully, other values indicate failure.

4.21.6. GetBatchData

Get the result of processing.

Interface:

std::tuple<sail::Tensor,
    std::vector<BMImage>,
    std::vector<int>,
    std::vector<int>,
    std::vector<std::vector<int>>> GetBatchData();

Returns: tuple[data, images, channels, image_idxs, padding_attrs]

data: Tensor

The processed result Tensor.
images: std::vector<BMImage>

Original image sequence.
channels: std::vector<int>

The channel sequence of the original image.
image_idxs: std::vector<int>

Numbered sequence of original images.
padding_attrs: std::vector<std::vector<int> >

Attribute list of the filled image, filled starting point coordinate x, starting point coordinate y, width after scale transformation, height after scale transformation

4.21.7. set_print_flag

Set the flag for printing logs. If this interface is not called, the log will not be printed.

Interface:

void set_print_flag(bool print_flag);

Returns:

flag: bool

The printing flag, False means not printing, True means printing.