mxpi_channelimagesstitcher

Function

Combines multiple images into a large image, dynamically outputs the preprocessing information of each image, and provides the information to the coordinate assembly plugin.

Synchronous/Asynchronous (Status)

Synchronous

Constraints

  • The value of channelIds cannot be empty.
  • The width and height of images from all channels must be the same.
  • The value of the width of the output image is an integer within the range of [32, 4096]. The default value is 1920.
  • The value of the height of the output image is an integer within the range of [32, 4096]. The default value is 1080.

Plugin Base Class (Factory)

mxpi_channelimagesstitcher

Input/Output

  • Input: buffer (data type: MxpiBuffer), metadata (data type: MxpiVisionList)
  • Output: buffer (data type: MxpiBuffer), metadata (data type: MxpiVisionList)

Port Format (Caps)

  • Dynamic input: {"image/yuv"}
  • Static output: {"image/yuv"}; dynamic output: {"metadata/stitch-info"}

Property

For details, see Table Properties of the mxpi_channelimagesstitcher plugin.

Table 1 Properties of the mxpi_channelimagesstitcher plugin

Property Name

Description

Mandatory or Not

Modifiable or Not

dataSource

Index of the input data. Multiple indexes can be configured, but the number of indexes must be the same as that of input ports. The default value is the key value of the output port of the upstream plugin.

No

Yes

channelIds

Input channel IDs, which are separated by commas (,). Each channel ID must be unique.

Example: "channelIds":"0,1"

Yes

Yes

outputWidth

The value of the width of the output image is an integer within the range of [32, 4096]. The default value is 1920.

Yes

Yes

outputHeight

The value of the height of the output image is an integer within the range of [32, 4096]. The default value is 1080.

Yes

Yes

RGBValue

Background color value. Enter the R, G, and B values in sequence, for example, 255,255,255. By default, the value is empty and color padding is not performed. The default background color of DVPP is used.

Yes

Yes

Example

Input buffer 1 to N from multiple channels, specify a channel, and filter out buffers of other channels. Then, the input images are combined into a large image and the preprocessing information of each image is dynamically output to the coordinate assembly plugin.

Pipeline sample:
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
"mxpi_channelimagesstitcher0":{
    "props":{
         "channelIds":"0,1",
         "outputWidth": "1920",
         "outputHeight": "1080",
         "RGBValue": "255,255,255"
    },
    "factory":"mxpi_channelimagesstitcher",
    "next":["queue8","queue9"]
},