Azure OCR的api的使用

最新推荐文章于 2022-11-21 22:09:10 发布

你说过

最新推荐文章于 2022-11-21 22:09:10 发布

阅读量4.2k

点赞数 1

分类专栏： OCR 文章标签： OCR Azure Cognitive service 文本识别

本文链接：https://blog.csdn.net/u013363895/article/details/82626217

版权

OCR 专栏收录该内容

1 篇文章 0 订阅

订阅专栏

本文是调用微软OCR api的一些经验和代码记录：

微软OCR api的地址：https://docs.azure.cn/zh-cn/cognitive-services/computer-vision/quickstarts/java

其中包含各种语言、各种服务的调用示例：

本文主要是传入图片url或者图片stream来获取识别文本的response，通过response来实现自己的业务逻辑。

核心代码：

@Service(value = "azureOcrService")
public class DefaultAzureOcrService implements AzureOcrService {

    private static final Logger logger = Logger.getLogger("DefaultAzureOcrService");

    // Replace the subscriptionKey string value with your valid subscription key.
    public static final String subscriptionKey = "13hc77781f7e4b19b5fcdd72a8df7156";

    // if you want to use the celebrities model, change "landmarks" to "celebrities" here and in
    // uriBuilder.setParameter to use the Celebrities model.
    public static final String uriBase = "https://api.cognitive.azure.cn/vision/v1.0/ocr";

    @Autowired
    private HttpUtils httpUtils;


    @Override
    public String getSerialNumber(String url,InputStream inputStream) {

        HttpClient httpClient = new DefaultHttpClient();
        String serialNumber = null;
        try
        {
            URIBuilder uriBuilder = new URIBuilder(uriBase);

            uriBuilder.setParameter("language", "unk");
            uriBuilder.setParameter("detectOrientation ", "true");

            // Request parameters.
            URI uri = uriBuilder.build();
            HttpPost request = new HttpPost(uri);

            // Request headers.

            request.setHeader("Ocp-Apim-Subscription-Key", subscriptionKey);

            // Request body.
            if (url!=null){

                request.setHeader("Content-Type", "application/json");
                StringEntity requestEntity = new StringEntity("{\"url\":\""+url+"\"}");
                request.setEntity(requestEntity);
            }else {
                request.setHeader("Content-Type", "application/octet-stream");
                InputStreamEntity inputStreamEntity = new InputStreamEntity(inputStream);
                request.setEntity(inputStreamEntity);
            }

            // Execute the REST API call and get the response entity.
            HttpResponse response = httpClient.execute(request);
            HttpEntity entity = response.getEntity();

            if (entity != null)
            {
                // Format and display the JSON response.
                String jsonString = EntityUtils.toString(entity);
                JSONObject json = new JSONObject(jsonString);
                System.out.println("REST Response:\n");
                System.out.println(json.toString(2));
            }
        }
        catch (Exception e)
        {
            // Display error message.
            logger.info(e.getMessage());
        }
        return serialNumber;
    }
}

最好业务实现获取serialNumber可以通过response json来读取和判断。

代码主要是通过写controller 来获取image url、stream来实现图片的识别。（AzureOcrController是OCR的入口）

代码链接：https://github.com/Jacob029049/JWT

如果文中有什么错误，欢迎指出，以免更多的人被误导。谢谢。