免费在 GitHub 公开

azure-ai-formrecognizer-java

Azure AI Document Intelligence SDK for Java (com.azure:azure-ai-documentintelligence). Use for extracting text, tables, key-value pairs from documents, receipts, invoices, IDs, or building custom document models. Triggers: "document intelligence java", "form recognizer java", "extract text from PDF java", "OCR document java", "analyze invoice receipt java", "custom document model java", "document classification java".

中文简介正在准备中。当前显示原文说明。

microsoft/skills

集成工具Azure

查看安装方法

包含的文件（2）

SKILL.md13.1 KB
references/examples.md20.4 KB

SKILL.md（原文）

安装前可以查看交给 AI 代理的指令内容。

Azure AI Document Intelligence SDK for Java

Rebranding: Azure AI Form Recognizer is now Azure AI Document Intelligence. New projects should use com.azure:azure-ai-documentintelligence. The legacy azure-ai-formrecognizer package targets API version 2023-07-31 only. See Migration Guide.

Before Implementation

Search microsoft-docs MCP for current API patterns:

Query: "azure-ai-documentintelligence Java SDK"
Verify: Parameters match installed SDK version (latest GA: 1.0.7)

Installation

<dependency>
    <groupId>com.azure</groupId>
    <artifactId>azure-ai-documentintelligence</artifactId>
    <version>1.0.0</version>
</dependency>

<!-- For DefaultAzureCredential -->
<dependency>
    <groupId>com.azure</groupId>
    <artifactId>azure-identity</artifactId>
    <version>1.14.2</version>
</dependency>

Environment Variables

DOCUMENT_INTELLIGENCE_ENDPOINT=https://<resource>.cognitiveservices.azure.com/ # Required for all auth methods
AZURE_TOKEN_CREDENTIALS=prod  # Required only if DefaultAzureCredential is used in production

Authentication

DefaultAzureCredential (Recommended)

import com.azure.ai.documentintelligence.DocumentIntelligenceClient;
import com.azure.ai.documentintelligence.DocumentIntelligenceClientBuilder;
import com.azure.core.credential.TokenCredential;
import com.azure.identity.AzureIdentityEnvVars;
import com.azure.identity.DefaultAzureCredentialBuilder;
import com.azure.identity.ManagedIdentityCredentialBuilder;

TokenCredential credential = new DefaultAzureCredentialBuilder()
    .requireEnvVars(AzureIdentityEnvVars.AZURE_TOKEN_CREDENTIALS)
    .build();
// Or use a specific credential directly in production:
// See https://learn.microsoft.com/java/api/overview/azure/identity-readme?view=azure-java-stable#credential-classes
// TokenCredential credential = new ManagedIdentityCredentialBuilder().build();

DocumentIntelligenceClient client = new DocumentIntelligenceClientBuilder()
    .endpoint(System.getenv("DOCUMENT_INTELLIGENCE_ENDPOINT"))
    .credential(credential)
    .buildClient();

API Key

import com.azure.core.credential.AzureKeyCredential;

DocumentIntelligenceClient client = new DocumentIntelligenceClientBuilder()
    .endpoint(System.getenv("DOCUMENT_INTELLIGENCE_ENDPOINT"))
    .credential(new AzureKeyCredential(System.getenv("DOCUMENT_INTELLIGENCE_KEY")))
    .buildClient();

Administration Client

import com.azure.ai.documentintelligence.DocumentIntelligenceAdministrationClient;
import com.azure.ai.documentintelligence.DocumentIntelligenceAdministrationClientBuilder;
import com.azure.core.credential.TokenCredential;
import com.azure.identity.AzureIdentityEnvVars;
import com.azure.identity.DefaultAzureCredentialBuilder;
import com.azure.identity.ManagedIdentityCredentialBuilder;

TokenCredential credential = new DefaultAzureCredentialBuilder()
    .requireEnvVars(AzureIdentityEnvVars.AZURE_TOKEN_CREDENTIALS)
    .build();
// Or use a specific credential directly in production:
// See https://learn.microsoft.com/java/api/overview/azure/identity-readme?view=azure-java-stable#credential-classes
// TokenCredential credential = new ManagedIdentityCredentialBuilder().build();

DocumentIntelligenceAdministrationClient adminClient = new DocumentIntelligenceAdministrationClientBuilder()
    .endpoint(System.getenv("DOCUMENT_INTELLIGENCE_ENDPOINT"))
    .credential(credential)
    .buildClient();

Async Client

import com.azure.ai.documentintelligence.DocumentIntelligenceAsyncClient;
import com.azure.core.credential.TokenCredential;
import com.azure.identity.AzureIdentityEnvVars;
import com.azure.identity.DefaultAzureCredentialBuilder;
import com.azure.identity.ManagedIdentityCredentialBuilder;

TokenCredential credential = new DefaultAzureCredentialBuilder()
    .requireEnvVars(AzureIdentityEnvVars.AZURE_TOKEN_CREDENTIALS)
    .build();
// Or use a specific credential directly in production:
// See https://learn.microsoft.com/java/api/overview/azure/identity-readme?view=azure-java-stable#credential-classes
// TokenCredential credential = new ManagedIdentityCredentialBuilder().build();

DocumentIntelligenceAsyncClient asyncClient = new DocumentIntelligenceClientBuilder()
    .endpoint(System.getenv("DOCUMENT_INTELLIGENCE_ENDPOINT"))
    .credential(credential)
    .buildAsyncClient();

Prebuilt Models

Model ID	Purpose
`prebuilt-read`	Extract text, lines, words, languages
`prebuilt-layout`	Text, tables, selection marks, structure
`prebuilt-receipt`	Receipt data extraction
`prebuilt-invoice`	Invoice field extraction
`prebuilt-idDocument`	ID documents (passport, license)
`prebuilt-tax.us.w2`	US W2 tax forms
`prebuilt-healthInsuranceCard.us`	US health insurance cards
`prebuilt-contract`	Contract field extraction

Retired models: prebuilt-businessCard and prebuilt-document are retired in API version 2024-11-30. Use the legacy azure-ai-formrecognizer package for these.

Core Patterns

Analyze from File

import com.azure.ai.documentintelligence.models.*;
import com.azure.core.util.BinaryData;
import com.azure.core.util.polling.SyncPoller;
import java.io.File;

File document = new File("document.pdf");
BinaryData documentData = BinaryData.fromFile(document.toPath(), (int) document.length());

SyncPoller<AnalyzeOperationDetails, AnalyzeResult> poller =
    client.beginAnalyzeDocument("prebuilt-layout",
        new AnalyzeDocumentOptions(documentData));

AnalyzeResult result = poller.getFinalResult();

Analyze from URL

String documentUrl = "https://example.com/invoice.pdf";

SyncPoller<AnalyzeOperationDetails, AnalyzeResult> poller =
    client.beginAnalyzeDocument("prebuilt-invoice",
        new AnalyzeDocumentOptions(documentUrl));

AnalyzeResult result = poller.getFinalResult();

Extract Layout

AnalyzeResult result = poller.getFinalResult();

for (DocumentPage page : result.getPages()) {
    System.out.printf("Page has width: %.2f and height: %.2f, measured with unit: %s%n",
        page.getWidth(), page.getHeight(), page.getUnit());

    // Lines
    for (DocumentLine line : page.getLines()) {
        System.out.printf("Line '%s' is within bounding box %s.%n",
            line.getContent(), line.getPolygon());
    }

    // Selection marks
    for (DocumentSelectionMark mark : page.getSelectionMarks()) {
        System.out.printf("Selection mark is '%s' with confidence %.2f.%n",
            mark.getState(), mark.getConfidence());
    }
}

// Tables
for (DocumentTable table : result.getTables()) {
    System.out.printf("Table: %d rows x %d columns%n",
        table.getRowCount(), table.getColumnCount());
    for (DocumentTableCell cell : table.getCells()) {
        System.out.printf("Cell[%d,%d]: %s%n",
            cell.getRowIndex(), cell.getColumnIndex(), cell.getContent());
    }
}

Extract Document Fields

SyncPoller<AnalyzeOperationDetails, AnalyzeResult> poller =
    client.beginAnalyzeDocument("prebuilt-receipt",
        new AnalyzeDocumentOptions(receiptUrl));

AnalyzeResult result = poller.getFinalResult();

for (AnalyzedDocument doc : result.getDocuments()) {
    Map<String, DocumentField> fields = doc.getFields();

    DocumentField merchantName = fields.get("MerchantName");
    if (merchantName != null && merchantName.getType() == DocumentFieldType.STRING) {
        System.out.printf("Merchant: %s (confidence: %.2f)%n",
            merchantName.getValueString(), merchantName.getConfidence());
    }

    DocumentField transactionDate = fields.get("TransactionDate");
    if (transactionDate != null && transactionDate.getType() == DocumentFieldType.DATE) {
        System.out.printf("Date: %s%n", transactionDate.getValueDate());
    }
}

Analyze with Options

SyncPoller<AnalyzeOperationDetails, AnalyzeResult> poller =
    client.beginAnalyzeDocument("my-custom-model",
        new AnalyzeDocumentOptions(documentUrl)
            .setPages(Collections.singletonList("1-3"))
            .setLocale("en-US")
            .setDocumentAnalysisFeatures(Arrays.asList(DocumentAnalysisFeature.LANGUAGES))
            .setOutputContentFormat(DocumentContentFormat.TEXT));

Custom Models

Build Custom Model

String blobContainerUrl = "{SAS_URL_of_training_data}";

SyncPoller<DocumentModelBuildOperationDetails, DocumentModelDetails> poller =
    adminClient.beginBuildDocumentModel(
        new BuildDocumentModelOptions("my-custom-model", DocumentBuildMode.TEMPLATE)
            .setAzureBlobSource(new AzureBlobContentSource(blobContainerUrl)));

DocumentModelDetails model = poller.getFinalResult();
System.out.printf("Model ID: %s%n", model.getModelId());
System.out.printf("Created: %s%n", model.getCreatedOn());

model.getDocumentTypes().forEach((docType, details) -> {
    details.getFieldSchema().forEach((field, schema) -> {
        System.out.printf("Field: %s (%s)%n", field, schema.getType());
    });
});

Manage Models

// Resource limits
DocumentIntelligenceResourceDetails resourceDetails = adminClient.getResourceDetails();
System.out.printf("Models: %d / %d%n",
    resourceDetails.getCustomDocumentModels().getCount(),
    resourceDetails.getCustomDocumentModels().getLimit());

// List models
PagedIterable<DocumentModelDetails> models = adminClient.listModels();
for (DocumentModelDetails model : models) {
    System.out.printf("Model: %s, Created: %s%n",
        model.getModelId(), model.getCreatedOn());
}

// Get model
DocumentModelDetails model = adminClient.getModel("model-id");

// Delete model
adminClient.deleteModel("model-id");

Document Classification

Build Classifier

Map<String, ClassifierDocumentTypeDetails> docTypes = new HashMap<>();
docTypes.put("invoice", new ClassifierDocumentTypeDetails()
    .setAzureBlobSource(new AzureBlobContentSource(containerUrl).setPrefix("invoices/")));
docTypes.put("receipt", new ClassifierDocumentTypeDetails()
    .setAzureBlobSource(new AzureBlobContentSource(containerUrl).setPrefix("receipts/")));

SyncPoller<DocumentClassifierBuildOperationDetails, DocumentClassifierDetails> poller =
    adminClient.beginBuildClassifier(
        new BuildDocumentClassifierOptions("my-classifier", docTypes));

DocumentClassifierDetails classifier = poller.getFinalResult();

Classify Document

SyncPoller<AnalyzeOperationDetails, AnalyzeResult> poller =
    client.beginClassifyDocument("my-classifier",
        new ClassifyDocumentOptions(documentUrl));

AnalyzeResult result = poller.getFinalResult();
for (AnalyzedDocument doc : result.getDocuments()) {
    System.out.printf("Classified as: %s (confidence: %.2f)%n",
        doc.getDocumentType(), doc.getConfidence());
}

Error Handling

import com.azure.core.exception.HttpResponseException;

try {
    client.beginAnalyzeDocument("prebuilt-receipt",
        new AnalyzeDocumentOptions("invalid-url"));
} catch (HttpResponseException e) {
    System.out.printf("Status: %d, Error: %s%n",
        e.getResponse().getStatusCode(), e.getMessage());
}

Migration from azure-ai-formrecognizer

Old (formrecognizer v4.x)	New (documentintelligence v1.x)
`DocumentAnalysisClient`	`DocumentIntelligenceClient`
`DocumentAnalysisClientBuilder`	`DocumentIntelligenceClientBuilder`
`DocumentModelAdministrationClient`	`DocumentIntelligenceAdministrationClient`
`beginAnalyzeDocumentFromUrl(modelId, url)`	`beginAnalyzeDocument(modelId, new AnalyzeDocumentOptions(url))`
`beginAnalyzeDocument(modelId, data)`	`beginAnalyzeDocument(modelId, new AnalyzeDocumentOptions(data))`
`SyncPoller<OperationResult, AnalyzeResult>`	`SyncPoller<AnalyzeOperationDetails, AnalyzeResult>`
`field.getValueAsString()`	`field.getValueString()`
`field.getValueAsDate()`	`field.getValueDate()`
`field.getValueAsDouble()`	`field.getValueNumber()`
`field.getValueAsList()`	`field.getValueList()`
`field.getValueAsMap()`	`field.getValueObject()`
`mark.getSelectionMarkState()`	`mark.getState()`
`adminClient.beginBuildDocumentModel(url, mode, prefix, options, ctx)`	`adminClient.beginBuildDocumentModel(new BuildDocumentModelOptions(id, mode).setAzureBlobSource(...))`
`adminClient.getResourceDetails()` → `.getCustomDocumentModelCount()`	`adminClient.getResourceDetails()` → `.getCustomDocumentModels().getCount()`
`FORM_RECOGNIZER_ENDPOINT`	`DOCUMENT_INTELLIGENCE_ENDPOINT`

Reference Files

File	Contents
references/examples.md	Complete code examples for all scenarios

评价

还没有评价。欢迎分享你的使用感受。

无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

同一仓库的技能

agent-framework-azure-ai-py

免费

Build Azure AI Foundry agents using the Microsoft Agent Framework Python SDK (agent-framework-azure-ai). Use when creating persistent agents with AzureAIAgentsProvider, using hosted tools (code interpreter, file search, web search), integrating MCP servers, managing conversation threads, or implementing streaming responses. Covers function tools, structured outputs, and multi-tool agents.

中文简介正在准备中。当前显示原文说明。