Azure AI Vision Imageanalysis Java

1---
2name: azure-ai-vision-imageanalysis-java
3description: Build image analysis applications with Azure AI Vision SDK for Java. Use when implementing image captioning, OCR text extraction, object detection, tagging, or smart cropping.
4package: com.azure:azure-ai-vision-imageanalysis
5---
6 
7# Azure AI Vision Image Analysis SDK for Java
8 
9Build image analysis applications using the Azure AI Vision Image Analysis SDK for Java.
10 
11## Installation
12 
13```xml
14<dependency>
15    <groupId>com.azure</groupId>
16    <artifactId>azure-ai-vision-imageanalysis</artifactId>
17    <version>1.1.0-beta.1</version>
18</dependency>
19```
20 
21## Client Creation
22 
23### With API Key
24 
25```java
26import com.azure.ai.vision.imageanalysis.ImageAnalysisClient;
27import com.azure.ai.vision.imageanalysis.ImageAnalysisClientBuilder;
28import com.azure.core.credential.KeyCredential;
29 
30String endpoint = System.getenv("VISION_ENDPOINT");
31String key = System.getenv("VISION_KEY");
32 
33ImageAnalysisClient client = new ImageAnalysisClientBuilder()
34    .endpoint(endpoint)
35    .credential(new KeyCredential(key))
36    .buildClient();
37```
38 
39### Async Client
40 
41```java
42import com.azure.ai.vision.imageanalysis.ImageAnalysisAsyncClient;
43 
44ImageAnalysisAsyncClient asyncClient = new ImageAnalysisClientBuilder()
45    .endpoint(endpoint)
46    .credential(new KeyCredential(key))
47    .buildAsyncClient();
48```
49 
50### With DefaultAzureCredential
51 
52```java
53import com.azure.identity.DefaultAzureCredentialBuilder;
54 
55ImageAnalysisClient client = new ImageAnalysisClientBuilder()
56    .endpoint(endpoint)
57    .credential(new DefaultAzureCredentialBuilder().build())
58    .buildClient();
59```
60 
61## Visual Features
62 
63| Feature | Description |
64|---------|-------------|
65| `CAPTION` | Generate human-readable image description |
66| `DENSE_CAPTIONS` | Captions for up to 10 regions |
67| `READ` | OCR - Extract text from images |
68| `TAGS` | Content tags for objects, scenes, actions |
69| `OBJECTS` | Detect objects with bounding boxes |
70| `SMART_CROPS` | Smart thumbnail regions |
71| `PEOPLE` | Detect people with locations |
72 
73## Core Patterns
74 
75### Generate Caption
76 
77```java
78import com.azure.ai.vision.imageanalysis.models.*;
79import com.azure.core.util.BinaryData;
80import java.io.File;
81import java.util.Arrays;
82 
83// From file
84BinaryData imageData = BinaryData.fromFile(new File("image.jpg").toPath());
85 
86ImageAnalysisResult result = client.analyze(
87    imageData,
88    Arrays.asList(VisualFeatures.CAPTION),
89    new ImageAnalysisOptions().setGenderNeutralCaption(true));
90 
91System.out.printf("Caption: \"%s\" (confidence: %.4f)%n",
92    result.getCaption().getText(),
93    result.getCaption().getConfidence());
94```
95 
96### Generate Caption from URL
97 
98```java
99ImageAnalysisResult result = client.analyzeFromUrl(
100    "https://example.com/image.jpg",
101    Arrays.asList(VisualFeatures.CAPTION),
102    new ImageAnalysisOptions().setGenderNeutralCaption(true));
103 
104System.out.printf("Caption: \"%s\"%n", result.getCaption().getText());
105```
106 
107### Extract Text (OCR)
108 
109```java
110ImageAnalysisResult result = client.analyze(
111    BinaryData.fromFile(new File("document.jpg").toPath()),
112    Arrays.asList(VisualFeatures.READ),
113    null);
114 
115for (DetectedTextBlock block : result.getRead().getBlocks()) {
116    for (DetectedTextLine line : block.getLines()) {
117        System.out.printf("Line: '%s'%n", line.getText());
118        System.out.printf("  Bounding polygon: %s%n", line.getBoundingPolygon());
119        
120        for (DetectedTextWord word : line.getWords()) {
121            System.out.printf("  Word: '%s' (confidence: %.4f)%n",
122                word.getText(),
123                word.getConfidence());
124        }
125    }
126}
127```
128 
129### Detect Objects
130 
131```java
132ImageAnalysisResult result = client.analyzeFromUrl(
133    imageUrl,
134    Arrays.asList(VisualFeatures.OBJECTS),
135    null);
136 
137for (DetectedObject obj : result.getObjects()) {
138    System.out.printf("Object: %s (confidence: %.4f)%n",
139        obj.getTags().get(0).getName(),
140        obj.getTags().get(0).getConfidence());
141    
142    ImageBoundingBox box = obj.getBoundingBox();
143    System.out.printf("  Location: x=%d, y=%d, w=%d, h=%d%n",
144        box.getX(), box.getY(), box.getWidth(), box.getHeight());
145}
146```
147 
148### Get Tags
149 
150```java
151ImageAnalysisResult result = client.analyzeFromUrl(
152    imageUrl,
153    Arrays.asList(VisualFeatures.TAGS),
154    null);
155 
156for (DetectedTag tag : result.getTags()) {
157    System.out.printf("Tag: %s (confidence: %.4f)%n",
158        tag.getName(),
159        tag.getConfidence());
160}
161```
162 
163### Detect People
164 
165```java
166ImageAnalysisResult result = client.analyzeFromUrl(
167    imageUrl,
168    Arrays.asList(VisualFeatures.PEOPLE),
169    null);
170 
171for (DetectedPerson person : result.getPeople()) {
172    ImageBoundingBox box = person.getBoundingBox();
173    System.out.printf("Person at x=%d, y=%d (confidence: %.4f)%n",
174        box.getX(), box.getY(), person.getConfidence());
175}
176```
177 
178### Smart Cropping
179 
180```java
181ImageAnalysisResult result = client.analyzeFromUrl(
182    imageUrl,
183    Arrays.asList(VisualFeatures.SMART_CROPS),
184    new ImageAnalysisOptions().setSmartCropsAspectRatios(Arrays.asList(1.0, 1.5)));
185 
186for (CropRegion crop : result.getSmartCrops()) {
187    System.out.printf("Crop region: aspect=%.2f, x=%d, y=%d, w=%d, h=%d%n",
188        crop.getAspectRatio(),
189        crop.getBoundingBox().getX(),
190        crop.getBoundingBox().getY(),
191        crop.getBoundingBox().getWidth(),
192        crop.getBoundingBox().getHeight());
193}
194```
195 
196### Dense Captions
197 
198```java
199ImageAnalysisResult result = client.analyzeFromUrl(
200    imageUrl,
201    Arrays.asList(VisualFeatures.DENSE_CAPTIONS),
202    new ImageAnalysisOptions().setGenderNeutralCaption(true));
203 
204for (DenseCaption caption : result.getDenseCaptions()) {
205    System.out.printf("Caption: \"%s\" (confidence: %.4f)%n",
206        caption.getText(),
207        caption.getConfidence());
208    System.out.printf("  Region: x=%d, y=%d, w=%d, h=%d%n",
209        caption.getBoundingBox().getX(),
210        caption.getBoundingBox().getY(),
211        caption.getBoundingBox().getWidth(),
212        caption.getBoundingBox().getHeight());
213}
214```
215 
216### Multiple Features
217 
218```java
219ImageAnalysisResult result = client.analyzeFromUrl(
220    imageUrl,
221    Arrays.asList(
222        VisualFeatures.CAPTION,
223        VisualFeatures.TAGS,
224        VisualFeatures.OBJECTS,
225        VisualFeatures.READ),
226    new ImageAnalysisOptions()
227        .setGenderNeutralCaption(true)
228        .setLanguage("en"));
229 
230// Access all results
231System.out.println("Caption: " + result.getCaption().getText());
232System.out.println("Tags: " + result.getTags().size());
233System.out.println("Objects: " + result.getObjects().size());
234System.out.println("Text blocks: " + result.getRead().getBlocks().size());
235```
236 
237### Async Analysis
238 
239```java
240asyncClient.analyzeFromUrl(
241    imageUrl,
242    Arrays.asList(VisualFeatures.CAPTION),
243    null)
244    .subscribe(
245        result -> System.out.println("Caption: " + result.getCaption().getText()),
246        error -> System.err.println("Error: " + error.getMessage()),
247        () -> System.out.println("Complete")
248    );
249```
250 
251## Error Handling
252 
253```java
254import com.azure.core.exception.HttpResponseException;
255 
256try {
257    client.analyzeFromUrl(imageUrl, Arrays.asList(VisualFeatures.CAPTION), null);
258} catch (HttpResponseException e) {
259    System.out.println("Status: " + e.getResponse().getStatusCode());
260    System.out.println("Error: " + e.getMessage());
261}
262```
263 
264## Environment Variables
265 
266```bash
267VISION_ENDPOINT=https://<resource>.cognitiveservices.azure.com/
268VISION_KEY=<your-api-key>
269```
270 
271## Image Requirements
272 
273- Formats: JPEG, PNG, GIF, BMP, WEBP, ICO, TIFF, MPO
274- Size: < 20 MB
275- Dimensions: 50x50 to 16000x16000 pixels
276 
277## Regional Availability
278 
279Caption and Dense Captions require GPU-supported regions. Check [supported regions](https://learn.microsoft.com/azure/ai-services/computer-vision/concept-describe-images-40) before deployment.
280 
281## Trigger Phrases
282 
283- "image analysis Java"
284- "Azure Vision SDK"
285- "image captioning"
286- "OCR image text extraction"
287- "object detection image"
288- "smart crop thumbnail"
289- "detect people image"
290