Use Case: Image Processing Workflows
There are two workflows for image processing described in this article:
These workflows require workers that are part of the orkesworkers GitHub repository (and are running in the Orkes Playground for use).
Image Encoding and Resizing
| Run it in Orkes Playground |
|---|
| Sample run on Orkes Playground |
|---|
The image_multiple_convert_resize.json workflow takes the URL of an image and converts it to multiple formats and sizes, hosting the completed images on S3.
The workflow takes in 3 inputs:
- fileLocation: A HTTP location to an image file
- outputFormats: An array of image types (E.g., jpg, png, & webp)
- outputSizes: An array of sizes. Each size object has a width and height property
For example:
{
"maintainAspectRatio": true,
"fileLocation": "https://static01.nyt.com/images/2022/02/24/sports/24soccerRussia-02/merlin_195120864_4c81c841-1349-43dd-8204-c792708e5570-superJumbo.jpg",
"outputFormats": [
"jpg",
"png",
"webp"
],
"outputSizes": [
{
"width": 1200,
"height": 800
},
{
"width": 600,
"height": 400
},
{
"width": 300,
"height": 200
}
]
}
This will result in 9 output images, i.e., three different sizes with three formats each.
Output of the workflow
{
"fileLocations": {
"image_convert_resize_png_1200x800_3": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/1d932e8f-417c-41e6-a001-c2b3c230c43f.png"
},
"image_convert_resize_png_600x400_4": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/a4cde72e-8314-408a-bab7-56a277f32a67.png"
},
"image_convert_resize_webp_1200x800_6": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/839fa6c5-1315-4aad-950a-044356e232d2.webp"
},
"image_convert_resize_png_300x200_5": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/7e2bd2ba-8629-47cc-aa33-69b4192edef1.png"
},
"image_convert_resize_jpg_1200x800_0": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/887490b3-c47a-4b89-abb7-39ec356a2fe2.jpg"
},
"image_convert_resize_webp_300x200_8": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/a34d4f88-e3ea-4252-9486-534861da45ed.webp"
},
"image_convert_resize_jpg_300x200_2": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/5093f7b0-29a2-4ae8-95aa-50908ec1ac32.jpg"
},
"image_convert_resize_webp_600x400_7": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/31dfea0c-b684-450e-9a3a-66bcbde89411.webp"
},
"image_convert_resize_jpg_600x400_1": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/15836483-e3bf-4dad-a852-46afb1748b5f.jpg"
}
}
}
Image Effects
The image_effects.json workflow demonstrates how to add effects to images. The following recipes are currently supported:
- Applying sepia tone
- Making an image more vibrant
- Adding a watermark proportional to the size of an image
Workflow Inputs
The workflow takes in 3 inputs:
- fileLocation: A http location to an image file
- recipe: Supported recipes are sepia, vibrant and watermark
- recipeParameters: A json object of key value pairs that are specific to the recipe. The list of recipe parameters for each recipe
sepia
| Key | Value type | Description |
|---|---|---|
| sepiaIntensityThreshold | Integer (E.g., 80 ) | Applies a special effect to the image, similar to the effect achieved in a photo darkroom by sepia toning. Threshold ranges from 0 to QuantumRange and is a measure of the extent of the sepia toning. A threshold of 80 is a good starting point for a reasonable tone. |
vibrant
| Key | Value type | Description |
|---|---|---|
| vibrance | Integer (E.g., 4 ) | Positive values make the image more vibrant, whereas negative values make the image less vibrant. |
watermark
| Key | Value type | Description |
|---|---|---|
| watermarkFileLocation | String (url) | A logo file that will be applied as a watermark on top of the file that is specified by the fileLocation parameter. |
| gravity | String (E.g., southeast) | Specifies the location to apply the logo/watermark image. Allowed values: NorthWest, North, NorthEast, West, Center, East, SouthWest, South & SouthEast. |
Workflow Diagram
Example Input
{
"fileLocations": [
"https://static01.nyt.com/images/2022/02/24/sports/24soccerRussia-02/merlin_195120864_4c81c841-1349-43dd-8204-c792708e5570-superJumbo.jpg"
],
"recipeInfos": [
{
"recipe": "sepia",
"recipeParameters": {
"sepiaIntensityThreshold": 80
}
},
{
"recipe": "vibrant",
"recipeParameters": {
"vibrance": 4
}
},
{
"recipe": "watermark",
"recipeParameters": {
"watermarkFileLocation": "https://orkes.io/logo/orkes/png/4X/orkes-logo-purple-inverted-4x.png",
"gravity" : "southwest"
}
}
]
}
Example Output
{
"fileLocations": {
"image_effect_merlin_195120864_4c81c841-1349-43dd-8204-c792708e5570-superJumbo.jpg_sepia_0": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/e52e3ef6-956d-4b5c-9a7c-17ffa64645d2-SEPIA.jpg",
"recipe": "SEPIA",
"recipeParameters": {
"sepiaIntensityThreshold": 80
}
},
"image_effect_merlin_195120864_4c81c841-1349-43dd-8204-c792708e5570-superJumbo.jpg_vibrant_1": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/333ca485-6482-4e64-8547-4828289904d9-VIBRANT.jpg",
"recipe": "VIBRANT",
"recipeParameters": {
"vibrance": 4
}
},
"image_effect_merlin_195120864_4c81c841-1349-43dd-8204-c792708e5570-superJumbo.jpg_watermark_2": {
"fileLocation": "https://image-processing-orkes.s3.amazonaws.com/c065e32a-bc8d-448f-9d07-73b7373478a3-WATERMARK.jpg",
"recipe": "WATERMARK",
"recipeParameters": {
"watermarkFileLocation": "https://orkes.io/logo/orkes/png/4X/orkes-logo-purple-inverted-4x.png",
"gravity": "southwest"
}
}
}
}
Images from the workflow
Original image
Image on applying sepia recipe
Image on applying vibrant recipe
Image on applying watermark recipe (Logo gets added as per the gravity key mentioned in the input)
