You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
[!INCLUDE [applies to v3.0 and v2.1](includes/applies-to-v3-0-and-v2-1.md)]
18
+
::: moniker range="form-recog-3.0.0"
19
+
[!INCLUDE [applies to v3.0](includes/applies-to-v3-0.md)]
20
+
::: moniker-end
20
21
21
-
## How business card data extraction works
22
+
::: moniker range="form-recog-2.1.0"
23
+
[!INCLUDE [applies to v2.1](includes/applies-to-v2-1.md)]
24
+
::: moniker-end
22
25
23
-
Business cards are a great way of representing a business or a professional. The company logo, fonts and background images found in business cards help the company branding and differentiate it from others. Applying OCR and machine-learning based techniques to automate scanning of business cards is a common image processing scenario. Enterprise systems used by sales and marketing teams typically have business card data extraction capability integrated into them for the benefit of their users.
26
+
The Form Recognizer business card model combines powerful Optical Character Recognition (OCR) capabilities with deep learning models to analyze and extract key information from business card images. The API analyzes printed business cards; extracts key information such as first name, last name, company name, email address, and phone number; and returns a structured JSON data representation.
24
27
25
-
## Form Recognizer Business Card model
28
+
## Business card data extraction
26
29
27
-
The business card model combines powerful Optical Character Recognition (OCR) capabilities with deep learning models to analyze and extract key information from business card images. The API analyzes printed business cards; extracts key information such as first name, last name, company name, email address, and phone number; and returns a structured JSON data representation.
30
+
Business cards are a great way to represent a business or a professional. The company logo, fonts and background images found in business cards help promote the company branding and differentiate it from others. Applying OCR and machine-learning based techniques to automate scanning of business cards is a common image processing scenario. Enterprise systems used by sales and marketing teams typically have business card data extraction capability integration into for the benefit of their users.
28
31
32
+
::: moniker range="form-recog-3.0.0"
29
33
***Sample business card processed with [Form Recognizer Studio](https://formrecognizer.appliedai.azure.com/studio/prebuilt?formType=businessCard)***
30
34
31
-
:::image type="content" source="./media/studio/overview-business-card-studio.png" alt-text="sample business card" lightbox="./media/overview-business-card.jpg":::
35
+
:::image type="content" source="media/studio/overview-business-card-studio.png" alt-text="Screenshot of a sample business card analyzed in the Form Recognizer Studio." lightbox="./media/overview-business-card.jpg":::
36
+
37
+
::: moniker-end
38
+
39
+
::: moniker range="form-recog-2.1.0"
40
+
41
+
***Sample business processed with [Form Recognizer Sample Labeling tool](https://fott-2-1.azurewebsites.net/)***
42
+
43
+
:::image type="content" source="media/business-card-example.jpg" alt-text="Screenshot of a sample business card analyzed with the Form Recognizer Sample Labeling tool.":::
44
+
45
+
::: moniker-end
32
46
33
47
## Development options
34
48
49
+
::: moniker range="form-recog-3.0.0"
50
+
35
51
The following tools are supported by Form Recognizer v3.0:
See how data, including name, job title, address, email, and company name, is extracted from business cards using the Form Recognizer Studio or our Sample Labeling tool. You'll need the following resources:
71
+
See how data, including name, job title, address, email, and company name, is extracted from business cards. You'll need the following resources:
50
72
51
73
* An Azure subscription—you can [create one for free](https://azure.microsoft.com/free/cognitive-services/)
52
74
53
75
* A [Form Recognizer instance](https://portal.azure.com/#create/Microsoft.CognitiveServicesFormRecognizer) in the Azure portal. You can use the free pricing tier (`F0`) to try the service. After your resource deploys, select **Go to resource** to get your key and endpoint.
54
76
55
77
:::image type="content" source="media/containers/keys-and-endpoint.png" alt-text="Screenshot: keys and endpoint location in the Azure portal.":::
56
78
79
+
::: moniker range="form-recog-3.0.0"
80
+
57
81
#### Form Recognizer Studio
58
82
59
83
> [!NOTE]
@@ -70,20 +94,66 @@ See how data, including name, job title, address, email, and company name, is ex
70
94
> [!div class="nextstepaction"]
71
95
> [Try Form Recognizer Studio](https://formrecognizer.appliedai.azure.com/studio/prebuilt?formType=businessCard)
72
96
73
-
## Input requirements
97
+
::: moniker-end
98
+
99
+
::: moniker range="form-recog-2.1.0"
100
+
101
+
## Form Recognizer Sample Labeling tool
102
+
103
+
1. Navigate to the [Form Recognizer Sample Tool](https://fott-2-1.azurewebsites.net/).
104
+
105
+
1. On the sample tool home page, select the **Use prebuilt model to get data** tile.
106
+
107
+
:::image type="content" source="media/label-tool/prebuilt-1.jpg" alt-text="Screenshot of the layout model analyze results operation.":::
108
+
109
+
1. Select the **Form Type** to analyze from the dropdown menu.
110
+
111
+
1. Choose a URL for the file you would like to analyze from the below options:
*[**Sample ID document**](https://raw.githubusercontent.com/Azure-Samples/cognitive-services-REST-api-samples/master/curl/form-recognizer/DriverLicense.png).
*[**Sample business card image**](https://raw.githubusercontent.com/Azure/azure-sdk-for-python/master/sdk/formrecognizer/azure-ai-formrecognizer/samples/sample_forms/business_cards/business-card-english.jpg).
117
+
118
+
1. In the **Source** field, select **URL** from the dropdown menu, paste the selected URL, and select the **Fetch** button.
119
+
120
+
:::image type="content" source="media/label-tool/fott-select-url.png" alt-text="Screenshot of source location dropdown menu.":::
121
+
122
+
1. In the **Form recognizer service endpoint** field, paste the endpoint that you obtained with your Form Recognizer subscription.
123
+
124
+
1. In the **key** field, paste the key you obtained from your Form Recognizer resource.
125
+
126
+
:::image type="content" source="media/fott-select-form-type.png" alt-text="Screenshot of the select-form-type dropdown menu.":::
74
127
75
-
* For best results, provide one clear photo or high-quality scan per document.
76
-
* Supported file formats: JPEG/JPG, PNG, BMP, TIFF, and PDF (text-embedded or scanned). Text-embedded PDFs are best to eliminate the possibility of error in character extraction and location.
77
-
* For PDF and TIFF, up to 2000 pages can be processed (with a free tier subscription, only the first two pages are processed).
78
-
* The file size must be less than 500 MB for paid (S0) tier and 4 MB for free (F0) tier.
79
-
* Image dimensions must be between 50 x 50 pixels and 10,000 x 10,000 pixels.
80
-
* PDF dimensions are up to 17 x 17 inches, corresponding to Legal or A3 paper size, or smaller.
81
-
* The total size of the training data is 500 pages or less.
82
-
* If your PDFs are password-locked, you must remove the lock before submission.
128
+
1. Select **Run analysis**. The Form Recognizer Sample Labeling tool will call the Analyze Prebuilt API and analyze the document.
129
+
130
+
1. View the results - see the key-value pairs extracted, line items, highlighted text extracted and tables detected.
131
+
132
+
:::image type="content" source="media/business-card-results.png" alt-text="Screenshot of the business card model analyze results operation.":::
83
133
84
134
> [!NOTE]
85
135
> The [Sample Labeling tool](https://fott-2-1.azurewebsites.net/) does not support the BMP file format. This is a limitation of the tool not the Form Recognizer Service.
* Supported file formats: JPEG, PNG, PDF, and TIFF
150
+
* For PDF and TIFF, up to 2000 pages are processed. For free tier subscribers, only the first two pages are processed.
151
+
* The file size must be less than 50 MB and dimensions at least 50 x 50 pixels and at most 10,000 x 10,000 pixels.
152
+
153
+
::: moniker-end
154
+
155
+
::: moniker range="form-recog-3.0.0"
156
+
87
157
## Supported languages and locales
88
158
89
159
>[!NOTE]
@@ -112,22 +182,59 @@ See how data, including name, job title, address, email, and company name, is ex
112
182
| WorkPhones | Array of phone numbers | Work phone number(s) from business card | +1 xxx xxx xxxx |
113
183
| OtherPhones | Array of phone numbers | Other phone number(s) from business card | +1 xxx xxx xxxx |
114
184
115
-
## Form Recognizer v3.0
185
+
::: moniker-end
116
186
117
-
Form Recognizer v3.0 introduces several new features and capabilities.
187
+
::: moniker range="form-recog-2.1.0"
188
+
189
+
### Fields extracted
190
+
191
+
|Name| Type | Description | Text |
192
+
|:-----|:----|:----|:----|
193
+
| ContactNames | array of objects | Contact name extracted from business card |[{ "FirstName": "John", "LastName": "Doe" }]|
194
+
| FirstName | string | First (given) name of contact | "John" |
195
+
| LastName | string | Last (family) name of contact | "Doe" |
196
+
| CompanyNames | array of strings | Company name extracted from business card |["Contoso"]|
197
+
| Departments | array of strings | Department or organization of contact |["R&D"]|
198
+
| JobTitles | array of strings | Listed Job title of contact |["Software Engineer"]|
199
+
| Emails | array of strings | Contact email extracted from business card |["[email protected]"]|
200
+
| Websites | array of strings | Website extracted from business card |["https://www.contoso.com"]|
201
+
| Addresses | array of strings | Address extracted from business card |["123 Main Street, Redmond, WA 98052"]|
202
+
| MobilePhones | array of phone numbers | Mobile phone number extracted from business card |["+19876543210"]|
203
+
| Faxes | array of phone numbers | Fax phone number extracted from business card |["+19876543211"]|
204
+
| WorkPhones | array of phone numbers | Work phone number extracted from business card |["+19876543231"]|
205
+
| OtherPhones | array of phone numbers | Other phone number extracted from business card |["+19876543233"]|
206
+
207
+
## Supported locales
208
+
209
+
**Prebuilt business cards v2.1** supports the following locales:
210
+
211
+
***en-us**
212
+
***en-au**
213
+
***en-ca**
214
+
***en-gb**
215
+
***en-in**
216
+
217
+
### Migration guide and REST API v3.0
118
218
119
219
* Follow our [**Form Recognizer v3.0 migration guide**](v3-migration-guide.md) to learn how to use the v3.0 version in your applications and workflows.
120
220
121
-
* Explore our [**REST API**](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-2022-08-31/operations/AnalyzeDocument) to learn more about the v3.0 version and new capabilities.
* Try processing your own forms and documents with the [Form Recognizer Studio](https://formrecognizer.appliedai.azure.com/studio)
129
228
130
-
* Explore our REST API:
229
+
* Complete a [Form Recognizer quickstart](quickstarts/get-started-sdks-rest-api.md?view=form-recog-3.0.0&preserve-view=true) and get started creating a document processing app in the development language of your choice.
230
+
231
+
::: moniker-end
232
+
233
+
::: moniker range="form-recog-2.1.0"
234
+
235
+
* Try processing your own forms and documents with the [Form Recognizer Sample Labeling tool](https://fott-2-1.azurewebsites.net/)
236
+
237
+
* Complete a [Form Recognizer quickstart](quickstarts/get-started-sdks-rest-api.md?view=form-recog-2.1.0&preserve-view=true) and get started creating a document processing app in the development language of your choice.
238
+
239
+
::: moniker-end
131
240
132
-
> [!div class="nextstepaction"]
133
-
> [Form Recognizer API v3.0](https://westus.dev.cognitive.microsoft.com/docs/services/form-recognizer-api-2022-08-31/operations/AnalyzeDocument)
* To compose a model trained with a prior version of the API (v2.1 or earlier), train a model with the v3.0 API using the same labeled dataset. That addition will ensure that the v2.1 model can be composed with other models.
Copy file name to clipboardExpand all lines: articles/applied-ai-services/form-recognizer/concept-custom.md
+4-1Lines changed: 4 additions & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -16,7 +16,7 @@ recommendations: false
16
16
17
17
[!INCLUDE [applies to v3.0 and v2.1](includes/applies-to-v3-0-and-v2-1.md)]
18
18
19
-
Form Recognizer uses advanced machine learning technology to detect and extract information from forms and documents and returns the extracted data in a structured JSON output. With Form Recognizer, you can use pre-built or pre-trained models or you can train standalone custom models. Custom models extract and analyze distinct data and use cases from forms and documents specific to your business. Standalone custom models can be combined to create [composed models](concept-composed-models.md).
19
+
Form Recognizer uses advanced machine learning technology to detect and extract information from forms and documents and returns the extracted data in a structured JSON output. With Form Recognizer, you can use prebuilt or pre-trained models or you can train standalone custom models. Custom models extract and analyze distinct data and use cases from forms and documents specific to your business. Standalone custom models can be combined to create [composed models](concept-composed-models.md).
20
20
21
21
To create a custom model, you label a dataset of documents with the values you want extracted and train the model on the labeled dataset. You only need five examples of the same form or document type to get started.
22
22
@@ -162,6 +162,9 @@ The [Sample Labeling tool](https://fott-2-1.azurewebsites.net/) doesn't support
162
162
163
163
## Supported languages and locales
164
164
165
+
>[!NOTE]
166
+
> It's not necessary to specify a locale. This is an optional parameter. The Form Recognizer deep-learning technology will auto-detect the language of the text in your image.
167
+
165
168
The Form Recognizer v3.0 version introduces more language support for custom models. For a list of supported handwritten and printed text, see [Language support](language-support.md).
Copy file name to clipboardExpand all lines: articles/applied-ai-services/form-recognizer/concept-form-recognizer-studio.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -33,7 +33,7 @@ The following Form Recognizer service features are available in the Studio.
33
33
34
34
***General Documents**: Try out Form Recognizer's General Documents feature to extract key-value pairs and entities. Start with the [Studio General Documents feature](https://formrecognizer.appliedai.azure.com/studio/document). Explore with sample documents and your documents. Use the interactive visualization and JSON output to understand how the feature works. See the [General Documents overview](concept-general-document.md) to learn more and get started with the [Python SDK quickstart for Layout](quickstarts/get-started-sdks-rest-api.md?view=form-recog-3.0.0&preserve-view=true#general-document-model).
35
35
36
-
***Prebuilt models**: Form Recognizer's pre-built models enable you to add intelligent document processing to your apps and flows without having to train and build your own models. As an example, start with the [Studio Invoice feature](https://formrecognizer.appliedai.azure.com/studio/prebuilt?formType=invoice). Explore with sample documents and your documents. Use the interactive visualization, extracted fields list, and JSON output to understand how the feature works. See the [Models overview](concept-model-overview.md) to learn more and get started with the [Python SDK quickstart for Prebuilt Invoice](quickstarts/get-started-sdks-rest-api.md?view=form-recog-3.0.0&preserve-view=true#prebuilt-model).
36
+
***Prebuilt models**: Form Recognizer's prebuilt models enable you to add intelligent document processing to your apps and flows without having to train and build your own models. As an example, start with the [Studio Invoice feature](https://formrecognizer.appliedai.azure.com/studio/prebuilt?formType=invoice). Explore with sample documents and your documents. Use the interactive visualization, extracted fields list, and JSON output to understand how the feature works. See the [Models overview](concept-model-overview.md) to learn more and get started with the [Python SDK quickstart for Prebuilt Invoice](quickstarts/get-started-sdks-rest-api.md?view=form-recog-3.0.0&preserve-view=true#prebuilt-model).
37
37
38
38
***Custom models**: Form Recognizer's custom models enable you to extract fields and values from models trained with your data, tailored to your forms and documents. Create standalone custom models or combine two or more custom models to create a composed model to extract data from multiple form types. Start with the [Studio Custom models feature](https://formrecognizer.appliedai.azure.com/studio/custommodel/projects). Use the online wizard, labeling interface, training step, and visualizations to understand how the feature works. Test the custom model with your sample documents and iterate to improve the model. See the [Custom models overview](concept-custom.md) to learn more and use the [Form Recognizer v3.0 migration guide](v3-migration-guide.md) to start integrating the new models with your applications.
0 commit comments