Classes

A class specifically designed to deal with images of business cards. It auto-rotates the images and extracts their texts. Can be subclassed and modified to your own needs. Designed to work in conjunction with SBSDKMultipleObjectScannerViewController.

This class provides information about the current cameras exposure settings.

An easy-to-use wrapper around the AVFoundations video capture session. Sets up and configures the camera device and the preview layer. Provides functionality to capture still images.

Helper class, that gathers the image, camera and detection metadata of a document scan.

A realtime detector and recognizer for U.S. Cheques. It detects cheques on CMSampleBufferRef object from an AVFoundation video stream. After successful detection it runs a recognition operation to extract the data fields of the detected credit card.

Contains the result of a cheque recognition.

A specific subclass of SBSDKMachineReadableCode, that represents a QR code with personal contact information (vCard, MeCard). Upon creation it tries to geocode all contained postal adresses. Before accessing contact property, make sure processing equals 0.

UIViewController subclass showing an image and a polygon The user can move edge and corner handles to redefine the polygon manually. This class cannot be instanced from a storyboard. Instead it is installing itself as a child view controller onto a given parent view controller.

Deprecated! Use SBSDKImageEditingViewControllerinstead!

A UILabel subclass with rounded corners and content insets to display the status of the current document detection result.

Delivers information about the iOS device.

Wrapper class for disability certificates recognition. Recognition is performed on still UIImage or SampleBufferRef, result is incapsulated in SBSDKDisabilityCertificatesRecognizerResult instance.

Contains information about recognized Disability certificate checkbox.

Class contains date information retrieved by disability certificates recognizer.

Class contains disability certificates recognizer retrieved information.

The metrics and additional properties returned by ‘SBSDKDocumentDetector’. Can be used to make decisions on polygon acceptance, automatic capturing and user guidance.



• This class provides methods for detecting a documents outline polygon in an UIImage or CMSampleBuffer.

This class represents the result of a document detection on an image.

A specific subclass of SBSDKMachineReadableCode that represents a QR codes with iCal compatible events (VCALENDER: VEVENT:).

A layer class to display a view finder with colored background and transparent cutout with the given aspect ratio. To change the color just set the layers backgroundColor property.

A simple frame limiter class. An object can use this class to limit the number of events fired per second. E.g. A camera device captures video frames with 60 frames per second and passes them to a detector. Use this class to limit the number of detector runs to a smaller value, like 20 frames per second, skipping the remaining 40 frames per second. As the detector needs lots of CPU power this limit can drastically reduce CPU usage and therefore saves battery.

A generic bar code class. An instance of this class is returned from SBSDKMachineReadableCodeManager when no matching parser for this bar code has been found.

A generic QR code class. An instance of this class is returned from SBSDKMachineReadableCodeManager when no matching parser for this QR code has been found.

UIViewController subclass showing an image and a polygon The user can move edge and corner handles to redefine the polygon manually and rotate the image in 90 degree steps (counter)clockwise. This class cannot be instanced from a storyboard. Instead it is installing itself as a child view controller onto a given parent view controller.

Replaces the deprecated SBSDKCropViewController.

Encapsulates image metadata, e.g. EXIF, TIFF, JFIF in a convenient way. Use SBSDKImageMetadataProcessor to extract metadata from and inject changed metadata into image data.

Helper class to extract image metadata, e.g. EXIF, TIFF, JFIF from any images data. Be cautious: changing an images metadata can break it, e.g. by setting wrong width or height.

Example usage:

1. Use UIImageJPEGRepresentation(UIImage *image, CGFloat compressionFactor) to get the NSData object from an UIImage.

2. Call SBSDKImageMetadataProcessor method extractMetadataFromImageData: to retrieve a SBSDKImageMetadata object from the image data.

3. Modify the metadata object as needed.

4. Call SBSDKImageMetadataProcessor method imageDataByInjectingMetadata to inject the modified metadata into the image data.

5. Create a new UIImage object from the returned image data using UIImage(data:) constructor with the modified metadata.

Class containing the metrics for a single image channel.

• Utility class to extract the metrics for each individual channel of an UIImage.
• Metrics are min, max and mean values of each channel, the variance and standard deviation.
• Along with the metrics this class also extracts histograms and cumulative histograms for each channel.

A static class that processes images asynchronously: the heart of the Scanbot SDK image processing module. As image processing is most often consuming lots of memory this class processes images on a serial queue ensuring a minimum of large images in memory. Each operation reads the image from the imageURL, processes it and, if outputImageURL is a valid file URL, writes the result to outputImageURL. Upon completion completionHandler is called on main thread in a synchronous matter. After your completionHandler finished executing the next operation is enqueued. Its resultInfo dictionary contains the processed image, the source image URL and the destination image URL if specified.

A thread-safe mutable ordered collection of disk-backed UIImage objects. When it comes to image processing with large images it is impossible to hold all the of a collection images in memory. The images have to be backed on disk and reloaded from the file when demanded. This class is a simple wrapper around the described technique. You can add/remove images from the storage, access the storage via an index and retrieve images or image URLs. There is also support for (automatically) creating and caching thumbnails from the storages images. Conforms to NSCopying. When copying the disk-backing-storage is also copied to a different physical location. Accessing the storage and changing properties is implemented with thread-safety using a single-writer/multiple-reader pattern.

The deallocation of this object removes all images from disk.

Deprecated! Use SBSDKIndexedImageStorage instead!

A simple thread-safe multiple-reader-single-writer index based disk image cache class. Manages images in an array-like fashion.

A simple thread-safe multiple-reader-single-writer key-value fashioned disk image cache class. Manages images in a dictionary-like fashion.

Encapsulates lens and camera properties. Can be used with polygonal image cropping to achieve a very accurate aspect ratio of the resulting image. If created from image image metadata, uses a dictionary of some known devices to lookup the values.

A specific subclass of SBSDKMachineReadableCode that represents a QR code with a geographic locations (geo:). Performs reverse geocoding to get address from longitude and latitude.

The base class for a machine readable code. Contains string value and metadata. Provides some basic functions to support sharing, opening and displaying QR- or barcode data.

Custom machine readable code parsers return subclasses.

Helper class to parse and transform SBSDKMachineReadableCodeMetadata into specific SBSDKMachineReadableCode subclass instances. To transform a metadata object into a machine readable code object a matching parser is needed.

Parsers must conform to the SBSDKMachineReadableCodeParsing protocol and must be registered using the registerParser method.

Typically a parsers parseMetadata method takes an SBSDKMachineReadableCodeMetadata object, parses its stringValue and eventually returns an instance of a SBSDKMachineReadableCode subclass. If the parsing fails or the stringValue does not match the parsers pattern, the parser should return nil.

ScanbotSDK comes with some basic default parsers and can be extended with custom machine readable codes and their matching parsers.

A class that holds transformed metadata for a detected AVMetadataMachineReadableCodeObject. Pass these objects to SBSDKMachineReadableCodeManager to parse and transform them into more user-friendly representations.

Wrapper class for machine-readable zones recognition. Recognition is performed on still UIImage or SampleBufferRef, result is incapsulated in SBSDKMRZRecognizerResult instance. NOTE: In order to operate, this class requires tesseract languages and trained data to be present in application bundle

A class representing a single data field in a machine-readable-zone.

A class encapsulating a machine-readable-zone recognition result.

A class encapsulating machine-readable-zones text line.

A specific subclass of SBSDKMachineReadableCode that represents a QR code containing an email link (mailto: and smtp:).

A view controller for scanning multiple objects on a single captured still image. *

Detects multiple non-overlapping rectangular objects within an image. Like a bunch of business cards.

This class summarizes the results of an optical character recognition (OCR) operation for a given page. It contains the recognized text, and arrays for each type of SBSDKOCRResultBlock, as well as the result from the page layout analysis. For convenience this immutable class conforms to NSCopying protocol.

This class summarizes the results of an optical character recognition (OCR) operation for number of pages. It contains the array of SBSDKOCRPages. For convenience this immutable class conforms to NSCopying protocol.

Part of the optical character recognitions result. This result contains multiple SBSDKOCRResultBlock objects. Each block represents a text paragraph, a text line or a word and contains information about it, like the found text, the bounding box, where in the image the paragraph/line/word can be found, the confidence value and the type. For convenience this immutable class conforms to NSCopying protocol.

A factory class to enqueue and execute optical character recognition operations: the heart of the Scanbot SDK OCR module. It performs OCR operations on either a single image or a collection of images, see SBSDKImageStorage. Additionally it can perform a standalone page analysis on a single image. All of the operations are dispatched to a single serial queue to ensure low memory footprint and optimum CPU load.

Any text recognition operation will fail with an error if all of the specified languages are missing as language files. If the specified languages are partially available a warning will be logged, but the operation executes without using the missing languages. Please see the section Optical Character Recognition of the main SDK documentation for more details about the OCR language files.

The number of languages passed to OCR operations have a significant impact on the performance: the more languages, the slower the recognition process.

A static class that renders an ordered collection of images into a basic, none-searchable PDF. Very fast. It does not perform OCR.

Describes the result of a page layout analyzing operation. It contains page orientation, writing direction and text line order as well as a deskew angle. This result can be used to process the input image before running an OCR operation on it, e.g.

• Rotate the image to let the page orientation flip to SBSDKPageOrientationUp,
• Apply a rotation with the deskewAngle to remove distortion and get perfectly aligned text lines.
• Limit recognition languages to languages that use the analyzer results writingDirection and textlineOrder to achieve much better OCR performance. For convenience this immutable class conforms to NSCopying protocol.

A class that describes an aspect ratio for a documents page. *

Describes the position and the type of a text box within the image being detected or recognized.

Resulting element of the SBSDKPayFormScanners recognizer. Holds a token and its recognized text value after recognition.

Describes an ordered collection of geometric points in the images coordinate system. Used to express boxes and rectangles in non-perspective-corrected images.

Part of the detector result. Gathers a polygon and a token type.

Common super class for the result classes of the detector and the recognizer.

Result class of SBSDKPayFormScanners detector part. Describes wether the image contains a bank transfer form, and if so it also holds the bounding polygon of the form as well as the detected text boxes.

Result class of SBSDKPayFormScanners recognizer part. Encapsulates the recognized text fields.

The main scanner class that performs detection and recognition on either UIImage or CMSampleBufferRef (from camera). Usually start with detection on a smaller image (video frame) and if detection was successful (isValidForm == YES) run recognition on either the same or a larger image (camera still shot). The larger the image the better the recognition results. Images are rotated accordingly to either the UIImage imageOrientation property or the given video orientation. Detection and recognition are successful only if the rotated image has natural orientation.

Note: This class needs the german (deu.traineddata) language file installed. See SBSDKOpticalTextRecognizer on how to install language files in your app.

A specific subclass of SBSDKMachineReadableCode that represents a QR code with call-a-number links (tel:).

This class represents a 4-gon, a polygon with 4 edges. The polygons points are stored in a normalized unit coordinate system from {0, 0} to {1, 1}. To convert to absolute coordinates use the methods that take a size and return absolute values. An absolute space can be any 2-dimensional space with a size, e.g. an image.

Helper class that encapsulates an edge of a polygon and provides geometric helper methods.

A CAShapeLayer subclass that is used to draw bezier paths with fluid animation. The paths stroke color can be customized.

Factory class providing different NSOperationQueues to work on. There are 3 main queues:

• A serial queue for image processing.
• A serial queue for text processing like optical character recognition.
• A concurrent queue for networking operations allowing 4 concurrent operations.

A NSProgress subclass with a simpler update mechanism and linked operation cancelling. To cancel the underlying operation simply call -cancel on the progress object.

A class that manages installation and removal of resources, required by some features. It also manages installation of custom resources that extend some features functionality.

UIViewController subclass showing a camera screen and running a user guiding Document detector. Detection result is visualized using a polygonal bezier path. This class cannot be instanced from a storyboard. Instead it is installing itself as a child view controller onto a given parent view controller.

A specific subclass of SBSDKMachineReadableCode that represents a QR code contianing an SMS links (sms: and SMSTO:).

@class SBSDKShutterButton The button class used in the camera view controller to take a snapshot. Animates itself depending on its status.

Descriptor and helper class for describing and creating a directory on the file system.

Class used to convert and write images in TIFF format

Class used to define parameters for TIFF format writer.

Class used to describe user-defined TIFF field.

A class, that determines orientation of text on the image. Initialization of this class can be resourse-intensive, so in case you want to analyze multiple images, it is better to create instance of SBSDKTextOrientationRecognizer once and use it multiple times.

A configurable view controller for camera-based detection of QR and bar codes.

A SBSDKUIViewController subclass which adds support for enabled and disabled camera permission.

This class describes the look and feel, as well as the textual contents of the page cropping screen. Use the ‘defaultConfiguration’ class method to retrieve an instance and modify it.

Subconfiguration for the textual content of the page cropping screen.

Subconfiguration for user interface of the page cropping screen.

A configurable view controller to edit a pages polygon or the orientation of the page.

A thread-safe, mutable container for scanned pages of type ‘SBSDKUIPage’.

Subconfiguration for the behavior of the document scanning screen.

This class describes the look and feel, the behavior, as well as the textual contents of the detection screen for documents. Use the ‘defaultConfiguration’ class method to retrieve an instance and modify it.

Subconfiguration for the textual contents of the document scanning screen.

Subconfiguration for user interface of the document scanning screen.

A configurable view controller for camera based document scanning. An edge-detection based algorithm finds documents in the camera stream. A manual or automatic high resolution snapshot can be made. Based on the detected area of the document in the photo, the photo is cropped and perspective-corrected, so that a nice flattened image of the document only is created.

Subconfiguration for the textual contents of camera based view controllers.

Subconfiguration for the user interface of camera based view controllers.

A view controller for camera-based detection of machine readable zones.

Subconfiguration for the behavior of bar codes, QR codes and machine readable zones scanners.

This class describes the look and feel, as well as the textual contents of the detection screens for bar codes, QR codes and machine readable zones. Use the ‘defaultConfiguration’ class method to retrieve an instance and modify it.

Subconfiguration for the textual contents of the detector screens for bar codes, QR codes and machine readable zones.

Subconfiguration for user interface of the detector screens for bar codes, QR codes and machine readable zones.

Factory class for machine readable code types.

Renders a ‘SBSDKUIDocument’ into a PDF.

This class represents a scanned documents page.

A class for storing and restoring images in a memory-efficient way. Stores added images on disk and restores them if requested. Can generate preview images of optimal resolution for the devices screen. Manages two types of images: original images (as from camera or photo library) and document images (cropped and perspective-corrected). Helps you efficiently managing and persisting larger images without keeping them in memory.

A helper class to generate thumbnails from existing UIImage instances.

A ‘SBSDKUIWorkflowStep’ subclass specialized on recognizing barcodes/QR codes on low-res video frames (live detection).

A ‘SBSDKUIWorkflowStep’ subclass specialized on recognizing disability certificates (DC) on either high-res still images or low-res video frames (live detection).

A ‘SBSDKUIWorkflowStep’ subclass specialized on capturing documents from high-res still images.

A ‘SBSDKUIWorkflowStep’ subclass specialized on recognizing machine readable zones (MRZ) on either high-res still images or low-res video frames (live detection).

A ‘SBSDKUIWorkflowStep’ subclass specialized on recognizing payforms on either high-res still images or low-res video frames (live detection).

Abstract base class for SBSDKUI view controllers. Do not use/subclass directly.

Represents a scanning workflow, consisting of multiple steps, each of it performing an image scan. You can combine document scanning with QR code detection or machine readable zone recognition. Many steps can be run on a captured still image or a video frame, so a step can either be a live-detection or a still-image capturing step. In the end, when all steps have finished you can validate all step results in a single block, and restart the whole workflow if the validation fails. By subclassing ‘SBSDKUIWorkflowStep’ the creation of custom steps is possible.

Behaviour configuration object for a ‘SBSDKUIWorkflowScannerViewController’.

Configuration object that holds the subconfigurations for a ‘SBSDKUIWorkflowScannerViewController’.

Text configuration object for a ‘SBSDKUIWorkflowScannerViewController’.

User interface configuration object for a ‘SBSDKUIWorkflowScannerViewController’.

View controller that processes a workflow.

A class describing one single step of a workflow. There are some pre-defined workflow steps but you also may may subclass and create your own custom steps.

This class represents the result of one workflow step

A specific subclass of SBSDKMachineReadableCode that represents a QR code containing http or https links.

A specific subclass of SBSDKMachineReadableCode that represents a QR code containing information about a WiFi hotspot (WIFI:).

Main class of Scanbot SDK. Lets you install the license and allows basic configuration.