Skip to main content
Version: 2.5.X

Selphi Component

1. Introduction​

The Component addressed in the current document is called the Selphi Component. This is responsible for capturing a selfie of the user and the subsequent extraction of the most important facial characteristics. Its main functionalities are the following:

  • Internal camera management.
  • Permissions management.
  • Assists the processes of capturing the user's face.
  • Generation of templates with the facial characteristics and the image of the user's face for the Liveness detection process

1.1 Minimum requirements​

  • If the Injection Attack Detection (IAD) feature is used, then: API level 24 RAM: 3 GB

1.2 Plugin version​

The current plugin version can be checked as follows:

  • Look for the pubspec.yaml file at the root of the plugin.
  • The KEY/TAG version indicates the version.

2. Component integration​

ℹ️ Before integrating this component, it is recommended to read the documentation related to Core Component and follow the instructions indicated in said document.

This section will explain step by step how to integrate the current component into an existing project.

⚠️ For this section, the following values ​​will be considered:

  • <%APPLICATION_PATH%> - Path to the root of the application (example: /folder/example)
  • <%PLUGIN_CORE_PATH%> - Path to the root of the Core plugin, which is required (example:/folder/sdk-core)
  • <%PLUGIN_SELPHI_FACE_PATH%> - Path to the root of the current plugin (example /folder/sdk-selphi)

2.1. Plugin installation: Common​

The plugin allows execution on Android and iOS platforms. This section explains the common steps to all platforms. To install the plugin, the following steps must be adopted:

  • Make sure Flutter is installed.
  • Access APPLICATION_PATH at a terminal and run:
dart pub token add "https://facephicorp.jfrog.io/artifactory/api/pub/pub-pro-fphi"
  • Access to <%APPLICATION_PATH%> folder, and in the pubspec.yaml file add:
fphi_sdkmobile_selphi:
  hosted:
    name: fphi_sdkmobile_selphi
    url: https://facephicorp.jfrog.io/artifactory/api/pub/pub-pro-fphi/
  version: ^2.6.0

(Only if you use Selphi IAD)
fphi_sdkmobile_selphi_iad:
  hosted:
    name: fphi_sdkmobile_selphi_iad
    url: https://facephicorp.jfrog.io/artifactory/api/pub/pub-pro-fphi/
  version: ^2.6.0

2.2 Plugin installation: iOS​

2.2.1 Project configuration​

For the iOS version, when adding our plugin to the final application, the following points must be previously taken into account:

  • Add camera permissions: To use the widget, you need to enable the camera permission in the application's info.plist file (included within the project in the ios folder). You will need to edit the file with a text editor and add the following key/value pair:
<key>NSCameraUsageDescription</key>
<string>$(PRODUCT_NAME) uses the camera</string>

2.2.2 Update the Podfile​

In the project podfile it will be necessary to add the information from the private repository (see section 2.1). To do this, the following lines must be added at the beginning of the file:

platform :ios, '13.0' //MIN VERSION
plugin 'cocoapods-art', :sources => ['cocoa-pro-fphi']
source 'https://cdn.cocoapods.org/'

⚠️ To know more about the configuration and use of Cocoapods Artifactory, it is necessary to access the following document of Core Component.

2.2.3 Possible issues​

If environment problems occur or the plugin is not updated after making new changes (for example, problems occurred due to the bundle not being generated correctly, or the libraries not being updated to the correct versions), it is recommended to execute the following sequence of instructions after launching the plugin:

Open the application's ios folder at a terminal.

Run the following command:

pod deintegrate
  • Remove the Podfile.lock
  • Run the following command (or open the project with Xcode and run it):
pod install --repo-update

y

pod repo-art update cocoa-pro-fphi

2.3 Plugin installation: Android​

2.3.0 Set Android SDK version​

For Android, the minimum SDK version required by our native libraries is 24, so if your app has a Minimum SDK defined less than this, it will need to be modified to avoid a compile error. To do this, access the application's build.gradle file (located in the android folder) and modify the following parameter:

buildscript {
  ext {
    minSdkVersion = 24
  }
}

3. Component configuration​

The actual component contains a number of Dart methods and interfaces contained within the SelphiFaceConfiguration file. In this file you can find the necessary API for the communication between the application and the native functionality of the component. It is then explained what each one of those listed is for and the other properties that affect the operation of the component.

Below is the SelphiFaceConfiguration class, which allows you to configure the Selphi component:

class SelphiFaceConfiguration {
  bool mDebug;
  bool mFullscreen;
  bool mLogImages;
  bool mStabilizationMode;
  bool mTemplateRawOptimized;
  bool mEnableGenerateTemplateRaw;
  bool mQRMode;
  bool mShowResultAfterCapture;
  double mCropPercent;
  double mJPGQuality;
  SelphiCompressFormat mCompressFormat;
  String mTranslationsContent;
  String mViewsContent;
  int? mCameraId;
  dynamic mParams;
  SelphiFaceLivenessMode mLivenessMode;
  bool? mShowDiagnostic;
  bool? mShowTutorial;
  bool? mShowPreviousTip;
  SelphiFaceExtractionDuration? mExtractionDuration;
  SelphiCamera mCameraPreferred;
  String? license;
}

ℹ️ All the configuration can be found in the component's SelphiFaceConfiguration class.

When making the call to the widget there is a series of parameters that must be included. They will be briefly discussed below.

3.1. resourcesPath​

type: string

Sets the name of the resource file that the widget will use for its graphical configuration. This file is customisable and is located in the plugin in the src/main/assets folder for Android and in the ios/Frameworks and Resources folder for iOS. Its installation is transparent to the user, it will simply be added to the respective platform's projects during plugin installation. More details about how this resource pack works and how to modify it are explained in section 6.

resourcesPath: "fphi-selphi-widget-resources-sdk.zip",

3.2. crop​

type: boolean

Indicates whether the images returned (in the images parameter that is activated with logImages = true) in the completion event contain only the area of ​​the detected face, at a magnification given by CropPercent, or whether the entire image is returned.

crop: false

3.3. cropPercent​

type: float

Specifies the magnification percentage of the detected face area to compose the returned image.

cropPercent: 1.0

3.4. debug​

type: boolean

Sets the debugging mode of the widget.

debug: false

3.5. livenessMode​

type: SelphiFaceLivenessMode

Sets the liveness mode of the component. The possible values are:

  • NoneMode: Indicates that the photodetection mode should not be enabled in authentication processes.
  • PassiveMode: Indicates that the passive life test is carried out on the server, sending the corresponding "BestImage" for this purpose
  • MoveMode: Indicates that the liveness test is active by displaying instructions during the capture, and returning the corresponding result of the process.
livenessMode: SdkSelphiEnums.SdkLivenessMode.PassiveMode

3.6. stabilizationMode​

type: boolean

Property that allows activating or deactivating the stabilised mode before the face detection process. In the case of being activated, it will give some guidelines to know if it is correctly located or not.

stabilizationMode: true

3.8. fullScreen​

type: string

Sets whether you want the sdk to start in full screen mode, hiding the status bar.

fullscreen: true

3.9. logImages​

type: boolean

Indicates whether the sdk returns to the application the images used during extraction or not. It should be noted that returning images can result in a significant increase in device resource usage:

logImages: false

3.10. templateRawOptimized​

type: boolean

Indicates whether if the selfie token generated must be optimized or not.

templateRawOptimized: false

3.11. showDiagnostic​

type: boolean

Shows a pop-up with the component diagnostic if the process fails.

showDiagnostic: true

3.12 enableGenerateTemplateRaw​

type: boolean

Optional parameter. Only visible if the parameter enableGenerateTemplateRaw is set to true. The widget will return the bestImage encrypted and in stringBase64 format.

enableGenerateTemplateRaw: true

3.13 showResultAfterCapture​

type: boolean

Indicates whether or not to display a screen with the captured image of the selfie after the process. This screen gives the user the option of repeating the capture process if the image obtained is not correct.

showResultAfterCapture: false

3.14 showTutorial​

type: boolean

Indicates whether or not to display the tutorial before the process. After the tutorial finishes, the component process will continue as usual.

showTutorial: true

3.15 compressFormat​

type: boolean

Indicates the image format compression. The possible values are:

  • SelphiCompressFormat.T_PNG
  • SelphiCompressFormat.T_JPEG
compressFormat: SelphiCompressFormat.T_JPEG;

3.16 videoFilename​

type: string

⚠️ This is an advanced property, and in most use cases you don't need to modify it. Incorrect use may cause component malfunction.

Sets the absolute path of the file name in which a video of the capture process will be recorded. The application is responsible for requesting the necessary permissions to the phone in case that route requires additional permissions. The component, by default, will not perform any write processing unless a file path is specified using this method.

videoFilename: “\<videofile-path\>“;

3.17 translationsContent​

type: string

⚠️ This is an advanced property, and in most use cases you don't need to modify it. Incorrect use may cause component malfunction.

This property allows, through a string in xml format, to configure the current location of the widget. The definition of this model can be found, by default, in an internal translations folder within the resources .zip. This property allows an application to update and replace the current location of the component at run time.

translationsContent: “\<translation-content-string\>“;

3.18 viewsContent​

type: string

⚠️ This is an advanced property, and in most use cases you don't need to modify it. Incorrect use may cause component malfunction.

This property allows, through a string in xml format, to configure the views of the current component. The definition of this modelling can be found, by default, in a file called widget.xml inside the resources .zip. This property allows an application to update and override the layout of the component's internal screens whilst running.

viewsContent: “\<views-content-string\>“;

3.19. showPreviousTip​

type: boolean

Displays a pre-launch screen with information about the process to be performed and a launch button.

3.20. extractionDuration​

type: SelphiFaceExtractionDuration

Indicates the Duration of the facial extraction process. The possible values are:

  • SelphiFaceExtractionDuration.SHORT
  • SelphiFaceExtractionDuration.MEDIUM
  • SelphiFaceExtractionDuration.LONG

3.21. cameraPreferred​

type: SelphiCamera

Indicates which camera will carry out the capture process. The possible values are:

  • SelphiCamera.BACK
  • SelphiCamera.FRONT

3.22. license (string)​

Sets the widgets license. This license is provided by Facephi. This parameter is only necesary for selphi IA.

license: JSON.stringify(getPlatform() === 'android' ? lIC_ANDROID : lIC_IOS);

4. Component Usage​

The following will show how to execute the functionality of the current component.

⚠️ Remember that in order to launch a certain component previously, you must initialise the SDK with its respective licence, and then start a new operation. For further information, consult the documentation of the Core Component.

Once the component has been configured, to launch it, the following code must be executed:

Future<Either<Exception, SelphiFaceResult>> launchSelphiAuthenticate(String resourcesPath) async
  {
    return launchSelphiAuthenticateWithConfiguration(resourcesPath, createStandardConfiguration());
  }

  Future<Either<Exception, SelphiFaceResult>>
      launchSelphiAuthenticateWithConfiguration(
          String resourcesPath, SelphiFaceConfiguration configuration) async {
    try
    {
      FphiSdkmobileSelphi selphi = FphiSdkmobileSelphi();
      final Map resultJson = await selphi.startSelphiFaceWidget(
          resourcesPath: resourcesPath,
          widgetConfigurationJSON: configuration);

      if (resultJson != null)
        return Right(SelphiFaceResult.fromMap(resultJson));
      else
        throw Exception('Plugin internal error');
    }
    on Exception catch (e) {
      return (Left(e));
    }
  }

  /// Sample of standard plugin configuration
  SelphiFaceConfiguration createStandardConfiguration()
  {
    SelphiFaceConfiguration configurationWidget;
    configurationWidget = SelphiFaceConfiguration();
    configurationWidget.livenessMode = SelphiFaceLivenessMode.LM_PASSIVE; // Liveness mode
    configurationWidget.fullscreen = true;
    configurationWidget.enableImages = false;
    configurationWidget.jpgQuality = 0.95;
    configurationWidget.enableGenerateTemplateRaw = true;

    return configurationWidget;
  }

5. Return of result​

As shown in the example above, the result is returned in the form of a JSON object via Promises, whether it is a successful operation or an error:

  try
  {
      FphiSdkmobileSelphi selphi = FphiSdkmobileSelphi();
      final Map resultJson = await selphi.startSelphiFaceWidget(resourcesPath: resourcesPath, widgetConfigurationJSON: configuration);

      return Right(SelphiFaceResult.fromMap(resultJson));
  }
  on Exception catch (e) {
      return (Left(e));
  }

Regardless of whether the result is correct/erroneous, the result will have the following format:

class SelphiFaceResult {
  final SdkFinishStatus finishStatus;
  final String finishStatusDescription;
  final String errorDiagnostic;
  final String? errorMessage;
  final String? templateRaw;
  final String? qrData;
  final String? bestImage;
  final String? bestImageCropped;
  final String? bestImageTemplateRaw;
  final String? livenessDiagnostic;
  final String? iad;
}

ℹ️ The result will be returned via a Promise containing an object of class SelphiResult. Information on these fields has been elaborated on below.

5.0 finishStatus​

  • 1: The operation was successful.
  • 2: An error has occurred, which will be indicated in the errorDiagnostic string and, optionally, an extra information message will be displayed in the errorMessage property.

5.1 finishStatusDescription​

  • STATUS_OK: The operation was successful.
  • STATUS_ERROR: An error has occurred, which will be indicated in the errorDiagnostic string and, optionally, an extra information message will be displayed in the errorMessage property.

5.2 errorDiagnostic​

Returns the type of error that occurred (if there was one, which is indicated by the finishStatus parameter with the value Error). The values ​​it can have are the following:

  • NoError: No error has occurred. The process can continue.

  • UnknownError: Unhandled error. Possibly caused by a bug in the resource bundle.

  • CameraPermissionDenied: The exception that is thrown when the sdk does not have permission to access the camera.

  • SettingsPermissionDenied: The exception that is thrown when the widget does not have permission to access system settings (deprecated).

  • HardwareError: Exception that occurs when there is a hardware problem with the device, usually caused by very few available resources.

  • ExtractionLicenceError: Exception that occurs when there has been a licencing problem on the server.

  • UnexpectedCaptureError: Exception that occurs during the capture of frames by the camera.

  • ControlNotInitialisedError: The widget configurator has not been initialised.

  • BadExtractorConfiguration: Problem arose during widget configuration.

  • CancelByUser: The exception that is thrown when the user stops the extraction manually.

  • TimeOut: Exception that is thrown when a maximum time elapses without successfully completing the extraction.

  • InitProccessError: Exception that is thrown when the sdk cannot process the captured images.

  • NfcError: The exception that is thrown when the sdk does not have permission to access the nfc.

  • NetworkConnection: The exception that is thrown when there are issues with the means the device uses to connect to the network.

  • TokenError: The exception that is thrown when an invalid token is passed as a parameter.

  • InitSessionError: The exception that is thrown when session cannot be initialised. The normal thing is that it happens because the SdkCore was not called when calling any other component.

  • ComponentControllerError: The exception that is thrown when the component cannot be instantiated.

5.3 errorMessage​

Indicates an additional error message if necessary. It is an optional value.

5.4 templateRaw​

Returns the raw template that is generated after the extraction process.

5.5 bestImage​

Returns the best image extracted from the registration or authentication process. This image is the original size image extracted from the camera.

5.6 bestImageCropped​

Returns a cropped image centered on the user's face. This image is obtained from the bestImage. This is the image that should be used as the characteristic image of the user who carried out the registration or authentication process as an 'avatar'.

5.7 qrData​

Returns the captured QR code data.

5.8 bestImageTemplateRaw​

Optional parameter. Only visible if the parameter enableGenerateTemplateRaw is set to true. The widget will return the bestImage encrypted and in stringBase64 format.

5.9 livenessDiagnostic​

Optional output. Returns extra information of the scan face. Possible results:

enum class LivenessDiagnostic {
    NotRated,
    PhotoDetected,
    LivenessDetected,
    Unsuccess,
    UnsuccessLowPerformance,
    UnsuccessGlasses,
    UnsuccessLight,
    UnsuccessNoMovement,
    UnsuccessWrongDirection,
    UnsuccessTooFar,
    UnsuccessPlayback,
    UnsuccessDetectionRateFail,
    UnsuccessProximityFail,
    UnsuccessLandmarkDetectionFail,
    UnsuccessDuplicatedFrames,
    UnsuccessContinuityFail,
    UnsuccessOPFail,
    UnsuccessOPDubious,
    UnsuccessOPSpoofing,
    UnsuccessTemplateFail,
    UnsuccessInsufficientDetections,
    UnsuccessNoEyeMovementDetected,
    UnsuccessMovementDerivative,
    UnsuccessMovementBoundaries
}

5.10 iad​

Returns a token/hash that must be used to call a validation service. Adds a layer of defense against sophisticated attacks using digital injection or advanced spoofing. IMPORTANT: Parameter visible only to the selphid IAD component.

6. Component Customisation (Optional)​

This component allows the customisation of texts, images, fonts and colours. Customisation is done using the internally supplied .zip file. This zip is made up of a file called widget.xml that contains the definition of all the widget screens, each one of them with a series of elements which allow customisation. The zip file also contains a folder with graphic resources and another folder with the translations of the texts.

6.1. Basic description​

6.1.1. Text customisation​

Text customisation is carried out by editing the texts of the existing translation files in the resource .zip.

    /strings/strings.es.xml

    /strings/strings.xml

6.1.2. Image customisation​

To customise the images used by the widget, you must add the images to the resource .zip. In the zip there are 3 folders:

    /resources/163dpi

    /resources/326dpi

    /resources/489dpi

These folders correspond to the different screen densities and you can create as many density folders as you want. These folders contain the versions of the images for each of the resolutions.

It is necessary to add the images in all the folders, since once the optimal resolution for the device has been determined, the widget only loads images from the folder with the chosen resolution.

The images are referenced from the widget.xml file.

6.1.3. Colour customisation​

Button colour customisation is carried out from the widget.xml file. In it you can customise any colour of any graphic element that appears in the widget. Simply modify the colour of the desired property.

6.1.4. Font type customisation​

Font files should be placed in the `/resources/163dpi` folder and once there they can be referenced from the `widget.xml` file. To change the font of a text element, it would be enough to modify the 'font' property and put the name of the corresponding file.

In the next section, the information about the content of the resource bundle and the way to modify it will be elaborated on.

6.2. Advanced description​

6.2.1. Widget.xml​

This file contains the definition of all the properties that are configurable in the authentication and registration processes. It is divided by navigation screens and within each screen label are all the properties that can be modified.

6.2.2. Strings folder​

This folder contains a strings.xml file for each translation you want to support. The name must be formed as follows:

    strings.(language).xml

Where (language) is the language code. For example, `strings.es.xml` would be the Spanish translation, strings.en.xml the English translation, strings.es_ES.xml the Spanish of Spain or strings.es_AR.xml the Spanish of Argentina.

You can force the language or let the widget choose it based on device settings. When deciding which language to apply, the following order is followed:

  • Search by location code (for example, “es_AR”).

  • If it doesn't find a match, it would look for one for the generic language (i.e., in this case it would be "es").

  • If there were no results either, then it would use the default language.

At the code level it is possible to select the localisation using the locale property. This parameter accepts a string with the language code to be used (for example, "es" or "es_ES").

6.2.3. Resources folder​

It contains the folders with all the necessary resources to be able to be modified, divided into densities. It is mandatory to generate the images in all densities as the widget expects to find them in the folder corresponding to the density of the device. New folders with the desired density can also be created.

6.2.4. BACKGROUND element​

The background element is made up of 4 segments that can be coloured independently:

  • top: Defines the background colour of the top segment or panel.

  • middle_top: defines the background colour of the segment or panel where the camera image is located.

  • middle_bottom: defines the background colour of the segment or panel below the camera image.

  • bottom: defines the background colour of the bottom segment or panel.

You can also set certain properties that are used only on specific screens. We have listed them below referring to the screens on which they are used:

  • pagination_separator (RegistrationTips, FaceMovementTips): Defines the colour of the separation between the bottom panel and the panel below the camera.

  • mirror_border_colour (RegistrationTips, FaceMovementTips): Defines the colour of the border of the circle around the camera or video image of the registration tips. This element is also called a mirror.

  • mirror_border_width (RegistrationTips, FaceMovementTips): Defines the width of the border of the circle around the camera or video image of the registration tips. If we didn't want to display a border, we would have to assign a value of 0.0 to this property.

  • mirror_mist_colour (StartExtractor): Defines the colour of the centre circle on the pre-extraction screen. This colour should always have a transparency value since we must show the image of the camera so that the user can position himself correctly before starting the extraction. The colour format when a transparency value is included is RGBA (The alpha value will be indicated with the last byte).

  • mirror_colour (Results): defines the background colour of the circle that displays the results of the registration process.

6.2.5. BUTTON element​

  • background: defines the background colour of the button

  • decorator: defines the colour of the button's shadow

  • foreground: defines the colour of the button font in case the content is text

  • content_type: defines the content type of the button. There are 2 different types:

  • resource_id: Content must contain the name of a file in the resource bundle

  • text_id: Content must contain the identifier of a literal from the translations file of the resource bundle

  • content: defines the content of the button. It can be either an image or the identifier of a literal.

  • align: Defines the alignment of the button content, be it an image or a text

  • font: Defines the font used if the button content is text

  • font_size: Defines the font size if the button content is text

6.2.6. TEXT element​

The text elements are used to define the graphic aspect of the texts of each of the widget's screens. These are the properties that can be modified:

  • colour: defines the colour of the text.

  • font: defines the type of font used to display the text.

  • font_size: defines the font size.

It must be borne in mind that on the registration results screen, the two texts that define the quality of the registration have their colour forced to the colour of the bar that indicates the score.

6.2.7. IMAGE element​

  • value: defines the name of the file that contains the image to display.

image elements only have the property that defines the file where the image is physically located in the resource bundle. The images are obtained from the bundle by looking in the appropriate folder according to the density of the device.

6.2.8. VIDEO element​

  • value: defines the name of the file that contains the video to display.

Video elements only have the property that defines the file when the video is physically located in the resource bundle.