List of Nodes

To view which nodes are only available in the Pro version of Retrobatch, read here.

Read

Clipboard: Copy an image from the clipboard and process it, or optionally write a processed image to the clipboard. Or do both! Take an image from the clipboard, process it, and then copy it back. When writing to the clipboard when more than one image is present in the workflow, Retrobatch will write the last image listed. Retrobatch will notice when a new image shows up on the clipboard, and update the preview in the workflow accordingly.

Photos Library: Read images directly from Photos library, with images selectable by album.

Prompt for Files: A prompt via an open panel for files or a folder of images, and optionally subfolders, in order to pick images to be processed when the workflow is run.

RAW Process: Processes RAW images before converting them to the RGB color space. Check the 'draft mode' box for faster processing.

  • Exposure: Adjust the lightness or darkness of your entire image.

  • Bias: Increases and decreases in the amount of black in an image.

  • Boost: Changes the brightness of the lighter areas of the photo.

  • Shadow Boost: Changes the darkness of the darker areas of the photo.

  • Noise Reduction: Decreases the appearance of random speckled dots in areas where color should be even.

  • Temperature: Changes the photo colors from warm to cool.

  • Tint: Gives image a more green tint when slider is all the way to the left, more blue all the way to the right.

  • Scale: Change the scale of the image using a percentage slider. For more control, add a Scale node in your workflow.

  • Color Profile: Various color profile options are available.

  • Bits per Channel/Depth: Choose from 8, 16, or 32 bits per channel.

  • File type: Choose from HEIC, JPEG, JPEG2000, and TIFF.

Read Folder: Read a folder (and optionally subfolders) of images. Images can be sorted in the preview alphabetically by ascending or descending file name. Multiple Read Folders can be chained together in a workflow. Easily add a Read Folder by dragging and dropping a folder of images onto Retrobatch's canvas.

Read Individual Files: Read a single file, or multiple files from different locations. After adding the node, click the '+' to add images, or drag and drop images onto the node. Alternatively, drag and drop an image into Retrobatch, and a Read Individual Files node will be created. Click the '-' to remove images from this node.

Write

Make Symlink (Pro): Write out images to a folder, optionally converting them to TIFF, JPEG, JPEG2K, PNG, or HEIC.

Open In App (Pro): Open an image in a chosen application by adding this node after the 'write images' node.

Write Images: Write out images to a folder, change the file name if desired, and optionally convert images to HEIC, JPG, JPEG2K, PDF, PNG, PSD, TIFF, or WebP.

  • When a Write Images node is added to the canvas, you will be prompted to select a folder to write your images into.
  • Find and select an existing folder.
  • To create a new folder, navigate to where you want a new folder to be located, then click on 'New Folder' located on the lower lefthand corner of the window.
  • The file name field / token field in the Write node has the ability to create subfolders. For example, if you want your images to be in a folder with the year on it, you would add a Capture Date token to the front, then a "/" (path separator) token, and the file name will be after that.
  • In the file name field, tokens can be separated using characters such as underscores, or inserting additional spaces.
  • 'Overwrite existing images' will overwrite any images in the folder that contain the same filename as the images being written.
  • 'Write back to original images' will over-write the original images. Proceed with caution!

Blur

Blur Box Blur: Smooths or sharpens an image using a box-shaped convolution kernel.

Gaussian Blur: Spreads source pixels by an amount specified by a Gaussian distribution.

Noise Reduction: Reduces noise using a threshold value to define what is considered noise. Small changes in luminance below that value are considered noise and get a noise reduction treatment, which is a local blur. Changes above the threshold value are considered edges, so they are sharpened.

Color Adjustment

Auto Enhance: Automatically applies filters to enhance an image.

Change Color Profile (Pro): Assign a new color profile to an image.

Color Controls: Adjust saturation, brightness, and contrast values.

Exposure Adjust: Adjust the exposure setting for an image similar to controlling exposure for a camera when changing the F-stop.

Gamma Adjust: Adjust midtone brightness. This filter is typically used to compensate for nonlinear effects of displays. Adjusting the gamma effectively changes the slope of the transition between black and white.

Instant Alpha: Removes pixels from your image based on a sampling location and tolerance.

Opacity: Adjust the opacity or transparency of an image.

Vibrance: Adjust the saturation of an image while keeping pleasing skin tones.

Color Effect

Add Border: Add a solid color border to images. The width of the border is measured in pixels and the border extends outwards from the image bounds. Adding a transparent border will add an even amount of transparency to all sides of the image. This can be useful for adding a drop shadow or resizing an image without scaling it. Multiple Add Border nodes can be used in succession to create different border effects.

Color Invert: Inverts the colors in an image, similar to converting a photo negative to a photo or vice versa.

Color Posterize: Remaps red, green, and blue color components, flattening colors to achieve a look similar to that of a silk-screened poster.

Dither: Options for Atkinson, Classic Mac, Floyd Steinberg, Gaussian Blue, and Ordered Reproducible algorithms.

Drop Shadow: Creates a drop shadow for the transparent areas of an image. Adjust the color, offset, opacity, and blur of the shadow. Click on the shadow color well to bring up the Colors Palette.

Grayscale: Turns a colored image into grayscale with optional red, blue, and green biases. Black and white only option available by checking 'Assign Gray Profile'.

Mask to Alpha (Pro): Converts a grayscale image to a white image that is masked by alpha. The white values from the source image produce the inside of the mask and the black values become completely transparent.

Matte: Adds a colored background to transparent areas of images. Click on the color well to bring up the Colors Palette to select a matte color.

Photo Chrome: A preconfigured set of effects that imitate vintage photography film with exaggerated color.

Photo Fade: A preconfigured set of effects that imitate vintage photography film with diminished color.

Photo Instant: A preconfigured set of effects that imitate vintage photography film with distorted colors.

Photo Mono: A preconfigured set of effects that imitate black-and-white photography film with low contrast.

Photo Noir: A preconfigured set of effects that imitate black-and-white photography film with exaggerated contrast.

Photo Process: A preconfigured set of effects that imitate vintage photography film with emphasized cool colors.

Photo Tonal: A preconfigured set of effects that imitate black-and-white photography film without significantly altering contrast.

Photo Transfer: A preconfigured set of effects that imitate vintage photography film with emphasized warm colors.

Sepia Tone: Maps the colors of an image to various shades of brown.

Generator

Animated Image: Creates an animated GIF or PNG, with options to set the frames per second (FPS) and optional looping.

Image Grid: Places images into a grid with options to set the number of images going across, down, and the spacing vertically and horizontally between images.

Layer & Page Splitter (Pro): Split up multilayer or multipage GIFs, PNGs, TIFFs, PDFs, Icon, PSD, and Acorn files into individual images.

Limit (Pro): Limit the number of images that can pass through this node.

PDF Image Extractor (Pro): Pulls out bitmaps from PDF files and writes them out either as JPEG or PNG files. A PDF page can have multiple images in it, each at different resolutions. This node grabs the raw data for it, and pulls it out.

PDF Maker (Pro): Combine any images into a multi-page PDF. Editable field to create a PDF name. The order of the images in the Read folder will dictate the order they are placed in the PDF.

PSD Maker (Pro): Combine any images into a multi-layered Photoshop file.

Screenshots & Windows: Produces an image for every visible window, or one image for every screen.

Style Transfer (Pro): Using machine learning, use style transfer models (also known as feedforward style transfers) to imprint styles from one image to another. Example models can be downloaded from: https://developer.apple.com/machine-learning/models/. Download a CoreML model and place it in your ~/Library/Application Support/Retrobatch/MLModels/ folder, restart Retrobatch, and it will then show up in the Style Transfer node.

Metadata

Delete Metadata: Remove all metatdata from an image, such as GPS, time picture created, lens information, etc. Useful when uploading to social sites such as Facebook and Twitter. Checkbox option available to keep orientation data intact to ensure your image will still orient in the correct direction.

Fix EXIF Rotation: Rotates an image correctly based on its EXIF rotation setting.

Remove GPS: Removes the location metadata from images.

Set DPI: Change the DPI (dots per inch) setting for images.

Set General Metadata: Change limited metadata values in an image including title, author, description, and copyright.

Set Specific Metadata (Pro): Change a wide array of metadata values in an image including IPTC and EXIF data.

Misc

Change Bits Per Channel (Pro): Change the bits per channel component of images to 8, 16, or 32 bits.

Delay (Pro): Pauses the processing of the images for a set amount of time measured in seconds.

Duplicate: Makes copies of images. Combine this with the Image Grid node to repeat the same image multiple times.

Extract Depth Image (Pro): Copies out the depth data to the write folder from images which contain it. It turns the image asset into an asset for the depth data

Extract Disparity Image (Pro): Pulls out a disparity map on images which contain it (iPhone X(s), iPhone 8+, etc). MacOS 10.13+.

Extract Portrait Matte (Pro): Pulls out the portrait matte on images taken with iPhones that support it (iPhone X(s), iPhone 8+, etc). MacOS 10.14+.

Indexed PNG: This node builds on top of pngquant, which is a handy tool for reducing the number of colors in a PNG file resulting in smaller images. Creating indexed PNGs results in smaller file sizes.

Limit (Pro): Limits the number of images or PDFs allowed to pass through the node.

Round Corners: Rounds the corners of images with options to set the corner radius, background color, and opacity.

Set Finder Tags (Pro): Assign one or more tags to images in the Finder. This will work on the image that was most recently written, there is no need to write out a copy of an image to set it. Add a Read node, and then a Set Finder Tags right after it and Retrobatch will then set the tags on the original files.

Single Transparent Pixel: Puts a single transparent pixel in the bottom left corner of an image. Great for social sites which will otherwise compress an image if it has no transparency. Only TIFF, PNG, and JPEG2000 file formats support transparency.

Workflow Notes: A place to add text and notes about the workflow. The information in the workflow notes node does not get written out with the images.

Script

AppleScript (Pro): Run an AppleScript file as part of a workflow. The node references a script file and also works by calling 3 different handlers- on processStart(), on processAsset(anAsset), and on processEnd(). The processAsset handler can also return a Boolean value, which lets the AppleScript node know whether or not to continue processing the asset. The AppleScript code can act as a filter by calling various properties on the passed in asset. Open up Retrobatch's scripting dictionary in either Script Debugger (the preferred way) or Script Editor to see the list of properties on assets.

JavaScript (Pro): Run a JavaScript function when a workflow starts, each time an image is processed by the node, and at workflow end.

Shell Script (Pro): Run a shell script before, after, or for each image run through this node. When each image is run through the script, the first argument to the script is the location of the image.

Sharpen

Sharpen Luminance: Increases image detail by sharpening. It operates on the luminance of the image, the chrominance of the pixels remains unchanged.

Unsharp Mask: Increases the contrast of the edges between pixels of different colors in an image.

Sort

Classify Images (Pro): Analyzes images using machine learning to try and figure out subjects present in the image. Useful when combined with the Rules node to filter out wanted or unwanted subjects. Retrobatch allows you to choose additional CoreML classification models (Retrobatch uses MobileNet) to analyze images. Download a CoreML model and place it in your ~/Library/Application Support/Retrobatch/MLModels/ folder, restart Retrobatch, and it will then show up in the Classification node.

Additional CoreML models can be downloaded from Apple: https://developer.apple.com/machine-learning/

Rules (Pro): Filter out images based on set criteria. Click the '+' button in the inspector palette to add additional rules criteria within the same node.

  • File Name: Option for separating images based on whether the file name contains, matches, is like, begins with, ends with, is, is not....compared to the manually entered text.

  • File Path: Option for separating images based on whether the file path contains, matches, is like, begins with, ends with, is, is not....compared to the manually entered text. The file path is what specifies a file's unique location in a file system.

  • Top Classification: Option for sorting images based on whether the top classification term for that image contains, matches, is like, begins with, ends with, is, is not....compared to the manually entered text.

  • Any Classification: Option for sorting images taking into account the first 20 classifications for an image with a classification confidence greater than 1%. Choose whether the image classification term contains, matches, is like, begins with, ends with, is, is not....compared to the manually entered text.

  • Alpha Channel: Option for sorting images based on whether an alpha channel is or is not present.

  • Transparent Pixels: Option for sorting images based on whether transparent pixels are present or absent.

  • Color Model: Option for separating images based on whether it is / is not, and choose from RGB, CMYK, Gray, and Lab.

  • Image type: Option for separating images based on whether it is / is not, and choose a file type from the list.

  • Bits per Channel: Options for separating images based on manually entered bits per channel value and whether image is greater than, equal to, or less than this value.

  • DPI: Options for separating images based on manually entered DPI (dots per inch or pixels per inch) value, and whether image is greater than, equal to, or less than this value.

  • Image Pixel Width: Options for separating images based on manually entered image pixel width value, and whether image is greater than, equal to, or less than this value.

  • Image Pixel Height: Options for separating images based on manually entered image pixel height value, and whether image is greater than, equal to, or less than this value.

  • Megapixels: Options for separating images based on manually entered megapixel value, and whether image is greater than, equal to, or less than this value.

  • Images Tested: Options for limiting which images pass through the node based on the number of images being greater than, equal to, or less than a set value.

  • Page Number: Options for limiting which pages of a multipage image such as a GIF or PDF pass through the node. Add multiple page number rules to obtain specific page ranges. Be sure to add the Layer & Page Splitter node prior to this node in order for all pages in the image to be available to the node.

  • Classification Confidence: Options for separating images based on manually entered classification confidence, and whether image is greater than, equal to, or less than this percentage value.

Transform

Crop: Crops images by specific pixel dimensions, preset aspect ratios, or percent. Adjust the placement of the crop with the location widget in the inspector palette.

Flip: Flips an image vertical, horizontal, or both.

Multi Scale (Pro): Resizes an image multiple times, making copies for each resize to be processed by the next node. Useful for instances where an image needs to be written out in multiple sizes. Click on the '+' button in the inspector palette to add additional scaling fields.

Rotate: Rotates images by a given amount in degrees via manual entry or widget.

Scale: Resizes an image with various options for scaling including by percentage, fixed dimensions, fixed width, fixed height, with various quality settings available. Checkbox option available to match even edges to create images that can more easily be handled when sizing for retina display. Some scaling types include option to 'only scale smaller' which prevents images smaller than the chosen dimensions from being scaled up.

Trim To Edges: Trims the image down in size by removing any transparent pixels from the edges of an image. Useful for trimming up screenshot windows, or after adding a drop shadow.

Watermark

Image Watermark: Add an image watermark image to images. Choose the file to use as a watermark, and adjust its scale, opacity, blending mode, and check the 'tile' box if desired.

Text Watermark: Add a text watermark to images, with various settings. Token options available from the cogwheel drop down menu include:

  • Capture Date: Original capture date for the image if one is present in the image metadata. The date format is modifiable. The capture date can be modified by using the Delete Metadata node, then adding the Set Specific Metadata node and editing TIFF ▸ Date Time.

  • Current Date: The current date as obtained from the macOS. The date format is modifiable.

  • Author: Options for editing the author include the node 'Set General Metadata', and 'Set Specific Metadata' under PNG ▸ Author and TIFF ▸ Artist.

  • ©: Basic copyright symbol.

  • Copyright: Options for editing the copyright include the node 'Set General Metadata', and 'Set Specific Metadata' under PNG ▸ Copyright, IPTC ▸ Copyright Notice, and TIFF ▸ Copyright.

  • Description: Options for editing the description include the node 'Set General Metadata', and 'Set Specific Metadata' under PNG ▸ Description.

  • File Name: The file name of the image. The filename of the currently selected image can be seen by clicking on the 'i' in the lower righthand corner of the canvas.

  • Color Profile: The color profile of the image, such as sRBGB or generic RGB. The color profile of the currently selected image can be seen by clicking on the 'i' in the lower righthand corner of the canvas.

  • DPI: Dots per inch or pixels per inch value. The DPI of the currently selected image can be seen by clicking on the 'i' in the lower righthand corner of the canvas.

  • Image Height: The height of the image in pixels.

  • Image Width: The width of the image in pixels.

  • Keywords: The keywords of the currently selected image can be seen by clicking on the 'i' in the lower righthand corner of the canvas. Keywords can be set in the 'Set Specific Metadata' node via IPTC ▸ Keywords.

  • Pixel Depth: The pixel depth of the image, frequently 24 or 32.

  • File Number: The image number as it falls in the sequence of being processed.

  • Source Folder: The source folder from where the image was originally located.

Blending Mode Definitions

Compositing is the process of taking one or more images and putting them together into a single image. This is what Retrobatch does when it puts a watermark on top of an image, forming a single image. Retrobatch gives you the ability to decide how the watermark is drawn, which is known as its "blend mode". You may want to change a watermark's blend mode depending on how you want it to appear. The default blending mode is Normal. The best way to become familiar with blending modes is to play around and experiment.
Note: Descriptions taken from Apple's Quartz 2D Programming Guide

  • Normal: The default blend mode. Retrobatch draws the watermark on top of the image below, without any specific blending tricks.

  • Darken: Works by keeping whichever pixels are darker on the watermark or the main image.

  • Color Burn: Darkens the main image over the watermark. White produces no change.

  • Color Dodge: Brighten the main image over the watermark. Black produces no change.

  • Linear Burn: Sums the value in the main image and watermark and subtracts 1. This is the same as inverting each image, adding them together (as in Linear Dodge), and then inverting the result. Blending with white leaves the image unchanged.

  • Linear Dodge: Sums the values in the two images. Blending with white gives white. Blending with black does not change the image.

  • Color: Uses the luminance values of the main image with the hue and saturation values of the watermark. This mode preserves the gray levels in the image.

  • Screen: Multiplies the inverse of the watermark with the inverse of the main image to obtain colors that are at least as light as either of the two contributing colors. Screening any color with white will produce white while screening with black will leave the original color unchanged.

  • Lighten: The opposite of Darken, the Lighten blend mode draws whichever parts are lighter on the watermark or the main image.

  • Hue: Uses the luminance and saturation values of the main image with the hue of the watermark.

  • Saturation: Uses the luminance and hue values of the main image with the saturation of the watermark. Pure gray areas do not produce a change.

  • Difference: Subtracts either the watermark color from the main image color, or the reverse, depending on which has the greater brightness value. Watermark values that are black produce no change; white inverts the main image color values.

  • Addition: Looks at the color information in each channel, and adds the watermark color to the main image color.

  • Subtract: Looks at the color information in each channel and subtracts the watermark color from the main image color.

  • Divide: Looks at the color information in each channel and divides the watermark color from the main image color.

  • Multiply: Multiplies the watermark with the main image. The resulting color is at least as dark as either the source or background color. Multiplying any color with black will give black, and multiplying any color with white preserves the original color.

  • Exclusion: Produces a lower-contrast version of the difference blend mode. Watermark values that are black do not produce a change; white inverts the main image color values.

  • Luminosity: Uses the hue and saturation of the main image with the luminance of the watermark to create an effect that is inverse to the effect created by the color blend mode.

  • Pin Light: Replaces the colors, depending on the blend color. If the blend color (light source) is lighter than 50% gray, pixels darker than the blend color are replaced, and pixels lighter than the blend color do not change. If the blend color is darker than 50% gray, pixels lighter than the blend color are replaced, and pixels darker than the blend color do not change.

  • Soft Light: Either darkens or lightens colors, depending on the watermark color. If the watermark color is lighter than 50% gray, the main image lightens, similar to dodging. If the watermark color is darker than 50% gray, the main image darkens, similar to burning.

  • Hard Light: Either multiplies or screens colors, depending on the watermark color. If the watermark color is lighter than 50% gray, the main image is lightened, similar to screening. If the watermark color is darker than 50% gray, the main image is darkened, similar to multiplying.

  • Mask: Works under the same principles as a layer mask. White in the watermark will expose pixels in the main image below, where as black hides them.

  • Overlay: Either multiplies or screens the watermark with the main image, depending on the color of the main image. The result is to overlay the watermark while preserving the highlights and shadows of the main image.

  • Source Atop: The area of the watermark which overlaps the main image replaces that area of the main image, with the remaining area retained.