Get started with AI Builder Text recognition

-

Introduction 

AI Builder Text recognition is a prebuilt AI model that extracts lines of text from documents and images. AI models usually require that you provide samples of data to train before you are able to perform predictions. You can pretrain prebuilt models by using a set of samples that are provided by Microsoft so they are instantly ready for use in predictions.


When you are predicting, AI Builder Text recognition will try to identify printed or handwritten text in images or documents, with help from Azure Computer Vision OCR service technology. Thus far, only English texts are supported, and the text is extracted from JPG, PNG, BMP, TIFF images, and PDF.AI Builder Text recognition provides makers with Optical Character Recognition (OCR) capabilities. Being fully integrated with the rest of the business platform means that you can automate a wide range of scenarios that are currently only achieved with human intervention.


First scenario category: Assisted typing

Front line workers will take a picture of a product or an object that contains an identifier or a price. This information can be used to update and/or retrieve information from Microsoft CRM.


Example:


A car rental company's employee takes pictures of cars during checkout to report damage. The agent takes the picture, which is then processed by AI Builder to extract the car license plate number. This number is used to get information about the car into CRM and to display it to the agent. The agent records damage on the car, records the returned date, and then updates CRM.


Second scenario category: Person or product presence check

Front line workers take a picture of a person or an object that contains an identifier. This information can be used to confirm the presence.


Example:


An airport uses pilots' cards to confirm that they are boarding the correct flight. Airport staff will take a picture of the pilot's card, which is processed by AI Builder to extract the pilot's identifier. This identifier is used to get information about the pilot's incoming flight, check consistency with the pilot's presence, and confirm access.


Third scenario category: Automated data entry

Agents take pictures of paper forms that contain records. This information is automatically entered into CRM.


Example:


A public company is auditing pharmacies. The auditors go on site and use a paper form to record results. Afterward, an agent manually enters the results in CRM. The agent takes a picture of each audit record and saves it in a folder. A flow processes all new pictures in the folders, sends them to AI Builder to extract the audited notes, and then saves them in CRM.

Build a Power Apps application that uses AI Builder Text recognition

Your task for this exercise is to take a picture, or upload an existing one from the file system,and then retrieve text lines that are present in the image. Then, you will build a Power Apps canvas app and will use the Text recognizer component, which can be useful for assisted data entry. For more information, see the What are canvas apps in Power Apps? and the Use the Text recognizer component articles in Power Apps documentation.


Create an application in Power Apps


To create an application in Power Apps, follow these steps:


1. Select Create in the Power Apps menu.


2. Select Canvas app from blank.


3. Name the canvas app Contoso Text Recognizer.


4. Select Phone.


5. Select Create.


Your next task is to give users the ability to automatically retrieve the text lines that are visible in the image.1. From Insert > AI Builder, select Text recognizer (preview). This control allows you to take or upload a picture and automatically recognizes the container identifier.

 2. Resize the control to fit in the screen.


Provide feedback about recognized text

To provide feedback to the user about recognized text, follow these steps:


1. Select Insert a new label.


2. Beneath the Text Recognizer component, change the Text property to Selected text:


3. Beside the Insert a text field, change the Default property to TextRecognizer1.SelectedText. 


This property will display the text that was manually selected by the user from the list of texts that were recognized by AI Builder.


Comments

Post a Comment

Popular posts from this blog

Configure forms, charts, and dashboards in model driven apps

-
Image
    Model-driven apps           Model-driven app design is a component-focused approach to app development. Model-driven app design doesn’t require code and the apps you make can be simple or very complex. Unlike canvas app development where the designer has complete control over app layout, with model-driven apps much of the layout is determined for you and largely designated by the components you add to the app.   Model-driven apps start with your data model – building up from the shape of your core business data and processes in the Dataverse to model forms, views, and other components. Model-driven apps automatically generate great UI that is responsive across devices.           Model-driven app design doesn’t require code and the apps you make can be simple or very complex.  With model-driven apps, much of the layout is determined for you and largely designated by the components you add to the app. Form            Forms provide a structured way to represent the data when it is ren
Read more

Create and Manage tables in Dataverse

-
 Create and Manage tables in Dataverse           Dataverse allows data to be integrated from multiple sources into a single store, which can then be used in Power Apps. If you are using Power Apps, then the license will cover Dataverse so you don’t have to pay extra.  Microsoft Dataverse “Common data service” is a cloud-based storage space that organizations can use to store business application data. Dataverse is a straightforward platform that you can use to begin designing your data structures, it is cloud-based storage space.           The data within Dataverse is stored within a set of tables. A table is a set of records that used to store data, each table consists of some columns. Dataverse uses Azure Active Directory identity and access management mechanisms to help ensure that only authorized users can access the environment, data, and reports. Advantage to use Dataverse 1) Easy to manage 2) Secure store 3) Work with any type of data 4) Work with any type of apps 5) Metadata 6)
Read more

Use the Admin center to manage environments and data policies in Power Automate

-
Image
           Microsoft Power Automate is a no-code/low-code drag-and-drop solution that allows users to create workflows to automate repetitive tasks and business processes.           The admin center is the central location where tenant admins and environment admins manage an organization’s data policies and environments. Any changes you make in the admin center are immediately available to users within the organization. Open the Admin center 1.Power Automate settings 2.Go to Power Automate, and sign in by using your organizational account. 3.Select the Settings button (the gear symbol), and then select Admin Center on the menu. The Admin center is opened. Environments An environment is a space where you can store, manage, and share your organization's business data, apps, and flows. It also serves as a container to separate apps that might have different roles, security requirements, or target audiences. Export a flow 1.Go to Power Automate, and sign in by using your organizational
Read more