diff --git a/docs/ocr/umn/ALL_META.TXT.json b/docs/ocr/umn/ALL_META.TXT.json
new file mode 100644
index 00000000..0640a43c
--- /dev/null
+++ b/docs/ocr/umn/ALL_META.TXT.json
@@ -0,0 +1,635 @@
+[
+ {
+ "dockw":"User Guide"
+ },
+ {
+ "uri":"ocr_01_0002.html",
+ "node_id":"ocr_01_0002.xml",
+ "product_code":"ocr",
+ "code":"1",
+ "des":"Optical Character Recognition (OCR) detects and extracts text from images and converts the text into an editable JSON format.OCR provides open APIs, so you can use progra",
+ "doc_type":"usermanual",
+ "kw":"What Is OCR?,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"What Is OCR?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0028.html",
+ "node_id":"ocr_01_0028.xml",
+ "product_code":"ocr",
+ "code":"2",
+ "des":"General TableDetects and extracts text and their row and column locations from images of tables in various formats, as well as the text areas outside tables. It is used t",
+ "doc_type":"usermanual",
+ "kw":"Functions,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Functions",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0006.html",
+ "node_id":"ocr_01_0006.xml",
+ "product_code":"ocr",
+ "code":"3",
+ "des":"There are various factors, such as technology and cost, that limit the performance of OCR services. The system-level constraints are the most significant limitations that",
+ "doc_type":"usermanual",
+ "kw":"Constraints and Limitations,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Constraints and Limitations",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0024.html",
+ "node_id":"ocr_01_0024.xml",
+ "product_code":"ocr",
+ "code":"4",
+ "des":"Identity and Access Management (IAM) lets you control user authentication and access to OCR.Object Storage Service (OBS) is a stable, secure, efficient, and easy-to-use c",
+ "doc_type":"usermanual",
+ "kw":"Related Services,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Related Services",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0050.html",
+ "node_id":"ocr_01_0050.xml",
+ "product_code":"ocr",
+ "code":"5",
+ "des":"Log in to the OCR console and choose Service Monitoring. On the Service Monitoring page displayed, view the total number of API calls, number of successful API calls, and",
+ "doc_type":"usermanual",
+ "kw":"Monitoring,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Monitoring",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0153.html",
+ "node_id":"ocr_01_0153.xml",
+ "product_code":"ocr",
+ "code":"6",
+ "des":"Use Postman to call the General Text OCR API.To call an OCR API, perform the following steps:Step 1: Subscribing to a ServiceStep 2: Configuring the EnvironmentStep 3: Us",
+ "doc_type":"usermanual",
+ "kw":"Getting Started,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Getting Started",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0015.html",
+ "node_id":"ocr_01_0015.xml",
+ "product_code":"ocr",
+ "code":"7",
+ "des":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "doc_type":"usermanual",
+ "kw":"FAQ",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"FAQ",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0109.html",
+ "node_id":"ocr_01_0109.xml",
+ "product_code":"ocr",
+ "code":"8",
+ "des":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "doc_type":"usermanual",
+ "kw":"Token-based Authentication",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Token-based Authentication",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0062.html",
+ "node_id":"ocr_01_0062.xml",
+ "product_code":"ocr",
+ "code":"9",
+ "des":"username indicates the name of the user, and domainname indicates the name of the account to which the user belongs.If the token is obtained by an account, user name and ",
+ "doc_type":"usermanual",
+ "kw":"What Are the Username, Domain Name, and Project Name in the Token Message Body?,Token-based Authenti",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"What Are the Username, Domain Name, and Project Name in the Token Message Body?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0044.html",
+ "node_id":"ocr_01_0044.xml",
+ "product_code":"ocr",
+ "code":"10",
+ "des":"If you receive a 401 status code when calling the OCR service using a token, it means that the token has expired. The token is valid for 24 hours, so it is recommended to",
+ "doc_type":"usermanual",
+ "kw":"Why Do I Receive a 401 Status Code Even After Successfully Obtaining a Token?,Token-based Authentica",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Why Do I Receive a 401 Status Code Even After Successfully Obtaining a Token?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0040.html",
+ "node_id":"ocr_01_0040.xml",
+ "product_code":"ocr",
+ "code":"11",
+ "des":"When obtaining a token, refer to the error message and select the appropriate solution.Check whether the service region in the body and the corresponding key value are co",
+ "doc_type":"usermanual",
+ "kw":"Why Am I Experiencing Token Retrieval Failure When Calling an OCR API Using Postman?,Token-based Aut",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Why Am I Experiencing Token Retrieval Failure When Calling an OCR API Using Postman?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0120.html",
+ "node_id":"ocr_01_0120.xml",
+ "product_code":"ocr",
+ "code":"12",
+ "des":"When calling an API, if you receive the error message \"code\":403,\"message\":\"This user only supports console access, not programmatic access.\", it means that you do not ha",
+ "doc_type":"usermanual",
+ "kw":"Why Don't I Have Programming Access Permission?,Token-based Authentication,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Why Don't I Have Programming Access Permission?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0133.html",
+ "node_id":"ocr_01_0133.xml",
+ "product_code":"ocr",
+ "code":"13",
+ "des":"If the error message \"Error: Header overflow\" is displayed when calling an API using Postman to obtain an authentication token, it means that the header has exceeded its ",
+ "doc_type":"usermanual",
+ "kw":"Why Is the Header Overflowing When I Call an OCR API Using Postman?,Token-based Authentication,User ",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Why Is the Header Overflowing When I Call an OCR API Using Postman?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0046.html",
+ "node_id":"ocr_01_0046.xml",
+ "product_code":"ocr",
+ "code":"14",
+ "des":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "doc_type":"usermanual",
+ "kw":"APIs",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"APIs",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0056.html",
+ "node_id":"ocr_01_0056.xml",
+ "product_code":"ocr",
+ "code":"15",
+ "des":"Log in to the OCR console and choose Service Monitoring. On the Service Monitoring page displayed, view the total number of API calls, number of successful API calls, and",
+ "doc_type":"usermanual",
+ "kw":"How Do I View the API Usage?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I View the API Usage?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0102.html",
+ "node_id":"ocr_01_0102.xml",
+ "product_code":"ocr",
+ "code":"16",
+ "des":"No. It is not recommended to use one OCR API for a different function as each API has its own unique algorithms. Doing so may result in poor recognition results. For inst",
+ "doc_type":"usermanual",
+ "kw":"Can I Use an OCR API for a Different Function?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Can I Use an OCR API for a Different Function?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0093.html",
+ "node_id":"ocr_01_0093.xml",
+ "product_code":"ocr",
+ "code":"17",
+ "des":"At present, the OCR API can only recognize one image per call. To process multiple images in batches, you would need to write additional code that allows for repeated cal",
+ "doc_type":"usermanual",
+ "kw":"Does OCR Support Batch Processing?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Does OCR Support Batch Processing?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0033.html",
+ "node_id":"ocr_01_0033.xml",
+ "product_code":"ocr",
+ "code":"18",
+ "des":"OCR results are initially in JSON format, but they can be easily saved as Word, TXT, or PDF files.",
+ "doc_type":"usermanual",
+ "kw":"Is It Possible to Convert OCR Results into Word, TXT, or PDF Files?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Is It Possible to Convert OCR Results into Word, TXT, or PDF Files?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0032.html",
+ "node_id":"ocr_01_0032.xml",
+ "product_code":"ocr",
+ "code":"19",
+ "des":"To recognize an image, it must first be converted to its Base64 coding. This section provides an example using Python to explain how to convert a local image to Base64 co",
+ "doc_type":"usermanual",
+ "kw":"How Can I Get the Base64 Encoding of an Image?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Can I Get the Base64 Encoding of an Image?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0078.html",
+ "node_id":"ocr_01_0078.xml",
+ "product_code":"ocr",
+ "code":"20",
+ "des":"Cross-region OBS is not supported, and the OBS region must match the region of the service being called.For OBS resources with public read authorization, they can be acce",
+ "doc_type":"usermanual",
+ "kw":"Is It Possible to Call the OCR Service From a Different Region Than OBS Resources?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Is It Possible to Call the OCR Service From a Different Region Than OBS Resources?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0122.html",
+ "node_id":"ocr_01_0122.xml",
+ "product_code":"ocr",
+ "code":"21",
+ "des":"Use high-quality images with clear text and no glare as much as possible. When collecting images, try to increase the proportion of the area with text to be recognized, r",
+ "doc_type":"usermanual",
+ "kw":"How Can I Improve Recognition Accuracy?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Can I Improve Recognition Accuracy?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0123.html",
+ "node_id":"ocr_01_0123.xml",
+ "product_code":"ocr",
+ "code":"22",
+ "des":"The recognition speed is related to the image size, which affects the time required for network transmission and image Base64 decoding. To speed up image processing by th",
+ "doc_type":"usermanual",
+ "kw":"How Can I Improve Recognition Speed?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Can I Improve Recognition Speed?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0125.html",
+ "node_id":"ocr_01_0125.xml",
+ "product_code":"ocr",
+ "code":"23",
+ "des":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "doc_type":"usermanual",
+ "kw":"How Do I Convert a PDF to an Image?,APIs,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Convert a PDF to an Image?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0076.html",
+ "node_id":"ocr_01_0076.xml",
+ "product_code":"ocr",
+ "code":"24",
+ "des":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "doc_type":"usermanual",
+ "kw":"Error Codes",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Error Codes",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0017.html",
+ "node_id":"ocr_01_0017.xml",
+ "product_code":"ocr",
+ "code":"25",
+ "des":"Locate the cause based on the API calling result or error code.Check whether you have subscribed to the service.Check whether the token is correct or has expired.Check wh",
+ "doc_type":"usermanual",
+ "kw":"Why Am I Unable to Call an OCR API?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"Why Am I Unable to Call an OCR API?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0063.html",
+ "node_id":"ocr_01_0063.xml",
+ "product_code":"ocr",
+ "code":"26",
+ "des":"If an error message and error code are returned when an API is called:If error message \"Incorrect IAM authentication information: decrypt token fail\" and error code \"APIG",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error APIG.0301?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error APIG.0301?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0065.html",
+ "node_id":"ocr_01_0065.xml",
+ "product_code":"ocr",
+ "code":"27",
+ "des":"If error message \"Backend timeout.\" and error code \"APIG.0201\" are displayed when you call an OCR API, the request timed out.Perform the following steps to rectify the fa",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error APIG.0201?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error APIG.0201?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0121.html",
+ "node_id":"ocr_01_0121.xml",
+ "product_code":"ocr",
+ "code":"28",
+ "des":"The error message \"The API does not exist or has not been published in the environment.\" is displayed.Check whether the API URL is correct. For example, check whether the",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error APIG.0101?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error APIG.0101?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0137.html",
+ "node_id":"ocr_01_0137.xml",
+ "product_code":"ocr",
+ "code":"29",
+ "des":"If error message and error code \"error_msg\":\"Orchestration error.\",\"error_code\":\"APIG.0106\" are returned when an API is called, check whether the frontend and backend par",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error APIG.0106?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error APIG.0106?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0041.html",
+ "node_id":"ocr_01_0041.xml",
+ "product_code":"ocr",
+ "code":"30",
+ "des":"The error message \"The throttling threshold has been reached: policy user over ratelimit,limit:XX,time:1 minute\" and error code \"APIG.0308\" are displayed when you call an",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error APIG.0308?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error APIG.0308?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0089.html",
+ "node_id":"ocr_01_0089.xml",
+ "product_code":"ocr",
+ "code":"31",
+ "des":"If error message \"The token must be updated.\" and error code \"APIG.0307\" are displayed when you call an OCR API, the token has expired and needs to be updated.Perform the",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error APIG.0307?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error APIG.0307?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0103.html",
+ "node_id":"ocr_01_0103.xml",
+ "product_code":"ocr",
+ "code":"32",
+ "des":"If error message \"Invalid Token header. The Token not contain project item.\" and error code \"ModelArts.0212\" are displayed when you call an OCR API, the token is invalid ",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error ModelArts.0212?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error ModelArts.0212?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0064.html",
+ "node_id":"ocr_01_0064.xml",
+ "product_code":"ocr",
+ "code":"33",
+ "des":"If \"error_code\":\"ModelArts.4603\",\"error_msg\":\"Obtaining the file from the URL failed.\" is displayed, it indicates that the image file fails to be obtained from the URL. T",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error ModelArts.4603 or ModelArts.4704?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error ModelArts.4603 or ModelArts.4704?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0031.html",
+ "node_id":"ocr_01_0031.xml",
+ "product_code":"ocr",
+ "code":"34",
+ "des":"If \"\"error_code\":\"ModelArts.4204\",\"error_msg\":\"Request api error! The API XXX is not subscribed.\" is displayed when you call an API, it indicates that you have not subscr",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle the Error ModelArts.4204?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle the Error ModelArts.4204?",
+ "githuburl":""
+ },
+ {
+ "uri":"ocr_01_0096.html",
+ "node_id":"ocr_01_0096.xml",
+ "product_code":"ocr",
+ "code":"35",
+ "des":"When calling an OCR API, the following error codes may occur related to image quality:Error code AIS.0102: Unsupported image format.Error code AIS.0103: Image size does n",
+ "doc_type":"usermanual",
+ "kw":"How Do I Handle Image Quality Errors?,Error Codes,User Guide",
+ "search_title":"",
+ "metedata":[
+ {
+ "prodname":"ocr",
+ "documenttype":"usermanual"
+ }
+ ],
+ "title":"How Do I Handle Image Quality Errors?",
+ "githuburl":""
+ }
+]
\ No newline at end of file
diff --git a/docs/ocr/umn/CLASS.TXT.json b/docs/ocr/umn/CLASS.TXT.json
new file mode 100644
index 00000000..1757c93c
--- /dev/null
+++ b/docs/ocr/umn/CLASS.TXT.json
@@ -0,0 +1,317 @@
+[
+ {
+ "desc":"Optical Character Recognition (OCR) detects and extracts text from images and converts the text into an editable JSON format.OCR provides open APIs, so you can use progra",
+ "product_code":"ocr",
+ "title":"What Is OCR?",
+ "uri":"ocr_01_0002.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"1"
+ },
+ {
+ "desc":"General TableDetects and extracts text and their row and column locations from images of tables in various formats, as well as the text areas outside tables. It is used t",
+ "product_code":"ocr",
+ "title":"Functions",
+ "uri":"ocr_01_0028.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"2"
+ },
+ {
+ "desc":"There are various factors, such as technology and cost, that limit the performance of OCR services. The system-level constraints are the most significant limitations that",
+ "product_code":"ocr",
+ "title":"Constraints and Limitations",
+ "uri":"ocr_01_0006.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"3"
+ },
+ {
+ "desc":"Identity and Access Management (IAM) lets you control user authentication and access to OCR.Object Storage Service (OBS) is a stable, secure, efficient, and easy-to-use c",
+ "product_code":"ocr",
+ "title":"Related Services",
+ "uri":"ocr_01_0024.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"4"
+ },
+ {
+ "desc":"Log in to the OCR console and choose Service Monitoring. On the Service Monitoring page displayed, view the total number of API calls, number of successful API calls, and",
+ "product_code":"ocr",
+ "title":"Monitoring",
+ "uri":"ocr_01_0050.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"5"
+ },
+ {
+ "desc":"Use Postman to call the General Text OCR API.To call an OCR API, perform the following steps:Step 1: Subscribing to a ServiceStep 2: Configuring the EnvironmentStep 3: Us",
+ "product_code":"ocr",
+ "title":"Getting Started",
+ "uri":"ocr_01_0153.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"6"
+ },
+ {
+ "desc":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "product_code":"ocr",
+ "title":"FAQ",
+ "uri":"ocr_01_0015.html",
+ "doc_type":"usermanual",
+ "p_code":"",
+ "code":"7"
+ },
+ {
+ "desc":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "product_code":"ocr",
+ "title":"Token-based Authentication",
+ "uri":"ocr_01_0109.html",
+ "doc_type":"usermanual",
+ "p_code":"7",
+ "code":"8"
+ },
+ {
+ "desc":"username indicates the name of the user, and domainname indicates the name of the account to which the user belongs.If the token is obtained by an account, user name and ",
+ "product_code":"ocr",
+ "title":"What Are the Username, Domain Name, and Project Name in the Token Message Body?",
+ "uri":"ocr_01_0062.html",
+ "doc_type":"usermanual",
+ "p_code":"8",
+ "code":"9"
+ },
+ {
+ "desc":"If you receive a 401 status code when calling the OCR service using a token, it means that the token has expired. The token is valid for 24 hours, so it is recommended to",
+ "product_code":"ocr",
+ "title":"Why Do I Receive a 401 Status Code Even After Successfully Obtaining a Token?",
+ "uri":"ocr_01_0044.html",
+ "doc_type":"usermanual",
+ "p_code":"8",
+ "code":"10"
+ },
+ {
+ "desc":"When obtaining a token, refer to the error message and select the appropriate solution.Check whether the service region in the body and the corresponding key value are co",
+ "product_code":"ocr",
+ "title":"Why Am I Experiencing Token Retrieval Failure When Calling an OCR API Using Postman?",
+ "uri":"ocr_01_0040.html",
+ "doc_type":"usermanual",
+ "p_code":"8",
+ "code":"11"
+ },
+ {
+ "desc":"When calling an API, if you receive the error message \"code\":403,\"message\":\"This user only supports console access, not programmatic access.\", it means that you do not ha",
+ "product_code":"ocr",
+ "title":"Why Don't I Have Programming Access Permission?",
+ "uri":"ocr_01_0120.html",
+ "doc_type":"usermanual",
+ "p_code":"8",
+ "code":"12"
+ },
+ {
+ "desc":"If the error message \"Error: Header overflow\" is displayed when calling an API using Postman to obtain an authentication token, it means that the header has exceeded its ",
+ "product_code":"ocr",
+ "title":"Why Is the Header Overflowing When I Call an OCR API Using Postman?",
+ "uri":"ocr_01_0133.html",
+ "doc_type":"usermanual",
+ "p_code":"8",
+ "code":"13"
+ },
+ {
+ "desc":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "product_code":"ocr",
+ "title":"APIs",
+ "uri":"ocr_01_0046.html",
+ "doc_type":"usermanual",
+ "p_code":"7",
+ "code":"14"
+ },
+ {
+ "desc":"Log in to the OCR console and choose Service Monitoring. On the Service Monitoring page displayed, view the total number of API calls, number of successful API calls, and",
+ "product_code":"ocr",
+ "title":"How Do I View the API Usage?",
+ "uri":"ocr_01_0056.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"15"
+ },
+ {
+ "desc":"No. It is not recommended to use one OCR API for a different function as each API has its own unique algorithms. Doing so may result in poor recognition results. For inst",
+ "product_code":"ocr",
+ "title":"Can I Use an OCR API for a Different Function?",
+ "uri":"ocr_01_0102.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"16"
+ },
+ {
+ "desc":"At present, the OCR API can only recognize one image per call. To process multiple images in batches, you would need to write additional code that allows for repeated cal",
+ "product_code":"ocr",
+ "title":"Does OCR Support Batch Processing?",
+ "uri":"ocr_01_0093.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"17"
+ },
+ {
+ "desc":"OCR results are initially in JSON format, but they can be easily saved as Word, TXT, or PDF files.",
+ "product_code":"ocr",
+ "title":"Is It Possible to Convert OCR Results into Word, TXT, or PDF Files?",
+ "uri":"ocr_01_0033.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"18"
+ },
+ {
+ "desc":"To recognize an image, it must first be converted to its Base64 coding. This section provides an example using Python to explain how to convert a local image to Base64 co",
+ "product_code":"ocr",
+ "title":"How Can I Get the Base64 Encoding of an Image?",
+ "uri":"ocr_01_0032.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"19"
+ },
+ {
+ "desc":"Cross-region OBS is not supported, and the OBS region must match the region of the service being called.For OBS resources with public read authorization, they can be acce",
+ "product_code":"ocr",
+ "title":"Is It Possible to Call the OCR Service From a Different Region Than OBS Resources?",
+ "uri":"ocr_01_0078.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"20"
+ },
+ {
+ "desc":"Use high-quality images with clear text and no glare as much as possible. When collecting images, try to increase the proportion of the area with text to be recognized, r",
+ "product_code":"ocr",
+ "title":"How Can I Improve Recognition Accuracy?",
+ "uri":"ocr_01_0122.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"21"
+ },
+ {
+ "desc":"The recognition speed is related to the image size, which affects the time required for network transmission and image Base64 decoding. To speed up image processing by th",
+ "product_code":"ocr",
+ "title":"How Can I Improve Recognition Speed?",
+ "uri":"ocr_01_0123.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"22"
+ },
+ {
+ "desc":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "product_code":"ocr",
+ "title":"How Do I Convert a PDF to an Image?",
+ "uri":"ocr_01_0125.html",
+ "doc_type":"usermanual",
+ "p_code":"14",
+ "code":"23"
+ },
+ {
+ "desc":"HUAWEI CLOUD Help Center presents technical documents to help you quickly get started with HUAWEI CLOUD services. The technical documents include Service Overview, Price Details, Purchase Guide, User Guide, API Reference, Best Practices, FAQs, and Videos.",
+ "product_code":"ocr",
+ "title":"Error Codes",
+ "uri":"ocr_01_0076.html",
+ "doc_type":"usermanual",
+ "p_code":"7",
+ "code":"24"
+ },
+ {
+ "desc":"Locate the cause based on the API calling result or error code.Check whether you have subscribed to the service.Check whether the token is correct or has expired.Check wh",
+ "product_code":"ocr",
+ "title":"Why Am I Unable to Call an OCR API?",
+ "uri":"ocr_01_0017.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"25"
+ },
+ {
+ "desc":"If an error message and error code are returned when an API is called:If error message \"Incorrect IAM authentication information: decrypt token fail\" and error code \"APIG",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error APIG.0301?",
+ "uri":"ocr_01_0063.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"26"
+ },
+ {
+ "desc":"If error message \"Backend timeout.\" and error code \"APIG.0201\" are displayed when you call an OCR API, the request timed out.Perform the following steps to rectify the fa",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error APIG.0201?",
+ "uri":"ocr_01_0065.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"27"
+ },
+ {
+ "desc":"The error message \"The API does not exist or has not been published in the environment.\" is displayed.Check whether the API URL is correct. For example, check whether the",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error APIG.0101?",
+ "uri":"ocr_01_0121.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"28"
+ },
+ {
+ "desc":"If error message and error code \"error_msg\":\"Orchestration error.\",\"error_code\":\"APIG.0106\" are returned when an API is called, check whether the frontend and backend par",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error APIG.0106?",
+ "uri":"ocr_01_0137.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"29"
+ },
+ {
+ "desc":"The error message \"The throttling threshold has been reached: policy user over ratelimit,limit:XX,time:1 minute\" and error code \"APIG.0308\" are displayed when you call an",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error APIG.0308?",
+ "uri":"ocr_01_0041.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"30"
+ },
+ {
+ "desc":"If error message \"The token must be updated.\" and error code \"APIG.0307\" are displayed when you call an OCR API, the token has expired and needs to be updated.Perform the",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error APIG.0307?",
+ "uri":"ocr_01_0089.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"31"
+ },
+ {
+ "desc":"If error message \"Invalid Token header. The Token not contain project item.\" and error code \"ModelArts.0212\" are displayed when you call an OCR API, the token is invalid ",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error ModelArts.0212?",
+ "uri":"ocr_01_0103.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"32"
+ },
+ {
+ "desc":"If \"error_code\":\"ModelArts.4603\",\"error_msg\":\"Obtaining the file from the URL failed.\" is displayed, it indicates that the image file fails to be obtained from the URL. T",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error ModelArts.4603 or ModelArts.4704?",
+ "uri":"ocr_01_0064.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"33"
+ },
+ {
+ "desc":"If \"\"error_code\":\"ModelArts.4204\",\"error_msg\":\"Request api error! The API XXX is not subscribed.\" is displayed when you call an API, it indicates that you have not subscr",
+ "product_code":"ocr",
+ "title":"How Do I Handle the Error ModelArts.4204?",
+ "uri":"ocr_01_0031.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"34"
+ },
+ {
+ "desc":"When calling an OCR API, the following error codes may occur related to image quality:Error code AIS.0102: Unsupported image format.Error code AIS.0103: Image size does n",
+ "product_code":"ocr",
+ "title":"How Do I Handle Image Quality Errors?",
+ "uri":"ocr_01_0096.html",
+ "doc_type":"usermanual",
+ "p_code":"24",
+ "code":"35"
+ }
+]
\ No newline at end of file
diff --git a/docs/ocr/umn/PARAMETERS.txt b/docs/ocr/umn/PARAMETERS.txt
new file mode 100644
index 00000000..6da8d5f0
--- /dev/null
+++ b/docs/ocr/umn/PARAMETERS.txt
@@ -0,0 +1,3 @@
+version=""
+language="en-us"
+type=""
\ No newline at end of file
diff --git a/docs/ocr/umn/en-us_image_0000001707122028.png b/docs/ocr/umn/en-us_image_0000001707122028.png
new file mode 100644
index 00000000..dcf684db
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001707122028.png differ
diff --git a/docs/ocr/umn/en-us_image_0000001707228246.png b/docs/ocr/umn/en-us_image_0000001707228246.png
new file mode 100644
index 00000000..3d6f6338
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001707228246.png differ
diff --git a/docs/ocr/umn/en-us_image_0000001707396210.png b/docs/ocr/umn/en-us_image_0000001707396210.png
new file mode 100644
index 00000000..d6f55cbd
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001707396210.png differ
diff --git a/docs/ocr/umn/en-us_image_0000001707404882.png b/docs/ocr/umn/en-us_image_0000001707404882.png
new file mode 100644
index 00000000..58cfdb76
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001707404882.png differ
diff --git a/docs/ocr/umn/en-us_image_0000001707409654.png b/docs/ocr/umn/en-us_image_0000001707409654.png
new file mode 100644
index 00000000..1375e92b
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001707409654.png differ
diff --git a/docs/ocr/umn/en-us_image_0000001755213865.png b/docs/ocr/umn/en-us_image_0000001755213865.png
new file mode 100644
index 00000000..f7b42f24
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001755213865.png differ
diff --git a/docs/ocr/umn/en-us_image_0000001755220149.png b/docs/ocr/umn/en-us_image_0000001755220149.png
new file mode 100644
index 00000000..f37637c4
Binary files /dev/null and b/docs/ocr/umn/en-us_image_0000001755220149.png differ
diff --git a/docs/ocr/umn/ocr_01_0002.html b/docs/ocr/umn/ocr_01_0002.html
new file mode 100644
index 00000000..1433ae4f
--- /dev/null
+++ b/docs/ocr/umn/ocr_01_0002.html
@@ -0,0 +1,18 @@
+
+
+What Is OCR?
+Optical Character Recognition (OCR) detects and extracts text from images and converts the text into an editable JSON format.
+
OCR provides open APIs, so you can use programming languages such as Python and Java to call OCR APIs to extract text from images. OCR allows you to automate the collection of key data. It helps you build an intelligent service system to improve efficiency. For details about how to obtain APIs, see Optical Character Recognition API Reference.
+
Before You Start
You must have programming capabilities and be familiar with Java, Python, iOS, Android, and Node.js.
+
To use OCR, call APIs to detect and extract text from images or scanned documents, convert the text into an editable JSON format, and enter the results into business systems by coding or save them in formats such as TXT or Excel.
+
+
+
Using OCR for the First Time
If you are a first-time user, the following sections are a good place to start:
+
- Functions
Going through this Functions section, you will be able to learn about a collection of OCR functions.
+ - Getting Started
OCR provides open APIs for you. For details about how to use OCR, see Getting Started and Optical Character Recognition API Reference.
+
+
+
Constraints and Limitations
+There are various factors, such as technology and cost, that limit the performance of OCR services. The system-level constraints are the most significant limitations that affect all sub-services. In addition to these system-level constraints, each sub-service also has its own independent limitations.
+
General Table OCR
- Only images in PNG, JPG, JPEG, BMP, or TIFF format can be recognized.
- No side of the image can be smaller than 15 or larger than 8,192 pixels.
- The area to be recognized must occupy more than 80% of the image. When scanning a table, ensure that the entire table and its surrounding area are included in the image.
- An image can be rotated to any angle.
- Text in images with complex backgrounds (such as outdoor scenery or anti-counterfeit watermarks) or distorted table lines cannot be recognized.
- English and Chinese are supported but support for traditional Chinese characters is limited.
+
+
General Text OCR
- Only images in PNG, JPG, JPEG, BMP, GIF, TIFF, WebP, PCX, ICO, or PSD format can be recognized.
- No side of the image can be smaller than 15 or larger than 8,192 pixels.
- The area to be recognized must occupy more than 80% of the image. When scanning a table, ensure that all text and its surrounding area are included in the image.
- An image can be rotated to any angle.
- Text in images with complex backgrounds (such as outdoor scenery or anti-counterfeit watermarks) or distorted text cannot be recognized.
- Supported languages: Chinese, English, some traditional Chinese, Malay, Ukrainian, Hindi, Russian, Vietnamese, Indonesian, Thai, Arabic, German, Latin, French, Italian, Spanish, Portuguese, Romanian, Polish Amharic, Japanese, Korean, Turkish, Norwegian, Danish, and Swedish.
+
+
FAQ
+
+
+
diff --git a/docs/ocr/umn/ocr_01_0017.html b/docs/ocr/umn/ocr_01_0017.html
new file mode 100644
index 00000000..c9c84161
--- /dev/null
+++ b/docs/ocr/umn/ocr_01_0017.html
@@ -0,0 +1,12 @@
+
+
+Why Am I Unable to Call an OCR API?
+Fault Locating
- Locate the cause based on the API calling result or error code.
- Check whether you have subscribed to the service.
- Check whether the token is correct or has expired.
- Check whether the API is correctly called.
+
+
Related Services
+IAM
Identity and Access Management (IAM) lets you control user authentication and access to OCR.
+
+
OBS
Object Storage Service (OBS) is a stable, secure, efficient, and easy-to-use cloud storage service. OCR APIs involve processing user data, which can be efficiently handled in batches using OBS.
+
OCR APIs allow for data retrieval and processing from OBS through temporary or anonymous public authorization.
+
+
Functions
+Function Description
- General Table
Detects and extracts text and their row and column locations from images of tables in various formats, as well as the text areas outside tables. It is used to store information on documents and reports as structured data.
+ - General Text
Detects and extracts text and their locations from images and converts them into structured data.
+
+
+
+
How Do I Handle the Error ModelArts.4204?
+If ""error_code":"ModelArts.4204","error_msg":"Request api error! The API XXX is not subscribed." is displayed when you call an API, it indicates that you have not subscribed to the API. For details about how to subscribe to an API, see Subscribing to a Service.
+
If the service has been subscribed to, check whether the region (or account) where the service is subscribed to is the same as the region (or account) where the service is called. If they are the same, check whether the URL of the API is spelled correctly.
+
How Can I Get the Base64 Encoding of an Image?
+To recognize an image, it must first be converted to its Base64 coding. This section provides an example using Python to explain how to convert a local image to Base64 code. You can also use an online conversion tool.
+
Replace d:\demo.jpg in the code with the actual image path.
+
import base64
+with open("d:\demo.jpg", "rb") as image_file:
+ encoded_string = base64.b64encode(image_file.read()).decode()
+print(encoded_string)
+
Is It Possible to Convert OCR Results into Word, TXT, or PDF Files?
+OCR results are initially in JSON format, but they can be easily saved as Word, TXT, or PDF files.
+
Why Am I Experiencing Token Retrieval Failure When Calling an OCR API Using Postman?
+When obtaining a token, refer to the error message and select the appropriate solution.
+
- Check whether the service region in the body and the corresponding key value are correct.
- If error message "The userInfo is wrong" is displayed, do the following:
Set username and domainname correctly. Typically, the value of username is the same as that of domainname. If you are not sure about the value, log in to the My Credentials page. If you use an IAM account to obtain the token, set the parameters as follows:
+username: IAM username (subaccount name)
+domainname: account name
+ - If the error code "APIGW.0101" is displayed, perform the following operation:
Check whether the URL used to obtain the token is correct.
+
+
How Do I Handle the Error APIG.0308?
+The error message "The throttling threshold has been reached: policy user over ratelimit,limit:XX,time:1 minute" and error code "APIG.0308" are displayed when you call an OCR API.
+
Rectify the fault using either of the following methods:
+
- Use the retry mechanism to rectify the fault by checking the return value in the code and retrying the requests after a short period of time (for example, 2 to 5 seconds).
- You can also check whether the result of the previous request is returned at the backend. After the result of the previous request is returned, send the next request.
+
Why Do I Receive a 401 Status Code Even After Successfully Obtaining a Token?
+If you receive a 401 status code when calling the OCR service using a token, it means that the token has expired. The token is valid for 24 hours, so it is recommended to obtain a new token to call the OCR service.
+
APIs
+
+
+
diff --git a/docs/ocr/umn/ocr_01_0050.html b/docs/ocr/umn/ocr_01_0050.html
new file mode 100644
index 00000000..ac7d1779
--- /dev/null
+++ b/docs/ocr/umn/ocr_01_0050.html
@@ -0,0 +1,6 @@
+
+
+Monitoring
+Log in to the OCR console and choose Service Monitoring. On the Service Monitoring page displayed, view the total number of API calls, number of successful API calls, and number of failed API calls.
+
How Do I View the API Usage?
+Log in to the OCR console and choose Service Monitoring. On the Service Monitoring page displayed, view the total number of API calls, number of successful API calls, and number of failed API calls.
+
What Are the Username, Domain Name, and Project Name in the Token Message Body?
+username indicates the name of the user, and domainname indicates the name of the account to which the user belongs.
+
- If the token is obtained by an account, user name and domain name are the same.
- If the token is obtained by an IAM user (multiple IAM users can be created under an account), user name is a real-world username and is different from domain name.
+
project name indicates the project name, for example, eu-de. For details about how to obtain a project ID, see "Obtaining a Project ID" in Optical Character Recognition API Reference.
+
How Do I Handle the Error APIG.0301?
+If an error message and error code are returned when an API is called:
+
- If error message "Incorrect IAM authentication information: decrypt token fail" and error code "APIG.0301" are displayed, the token fails to decrypt.
Solution:
+(1) Check whether the token has expired.
+(2) Check whether the request body is correct and whether the token is correct and complete.
+(3) Check whether the environment where the token is obtained is the same as that where the token is invoked.
+
+
How Do I Handle the Error ModelArts.4603 or ModelArts.4704?
+"error_code":"ModelArts.4603","error_msg":"Obtaining the file from the URL failed."
If "error_code":"ModelArts.4603","error_msg":"Obtaining the file from the URL failed." is displayed, it indicates that the image file fails to be obtained from the URL. To locate the fault, follow these steps:
+
(1) Make sure that the provided URL supports the HTTP/HTTPS request protocol, which should be in the format of http/https URL.
+
(2) Check if the server where the images are stored is stable and reliable, if the network connection is normal, and if it is publicly accessible.
+
(3) Check if the content-type of the downloaded images is a standard type, such as image/gif, image/jpeg, image/png, image/tiff. It is recommended to use OBS URL for the request.
+
+
"error_code":"ModelArts.4704","error_msg":"Obtaining the file from the OBS failed."
If ""error_code":"ModelArts.4704","error_msg":"Obtaining the file from the OBS failed."" is displayed, it means that the image data fails to be obtained from OBS. Make sure that the OBS path where the images are stored exists and is accessible. If the path exists, make sure that the OBS bucket policy is set to public.
+
It is recommended not to use OBS paths across regions. If OBS and the deployment region of the service to be called are not in the same region, it is suggested to download the images locally and call the service using the image method.
+
+
How Do I Handle the Error APIG.0201?
+If error message "Backend timeout." and error code "APIG.0201" are displayed when you call an OCR API, the request timed out.
+
Perform the following steps to rectify the fault:
+
Use a tool, such as Postman, to call the service and check whether the call is successful. If the call is successful, the service API is normal. Perform the following steps to proceed:
+
- Check whether the original API call requests are excessively frequent. If so, check the return value in the code and resend the requests later (for example, 2 to 5 seconds later). You can also check whether the result of the previous request is returned at the backend. After the result of the previous request is returned, send the next request.
- Check whether the image is too large or the network delay is too long. If the image is too large, compress the image in proportion while ensuring the image definition. If the network delay is long, increase the network transmission speed.
+
If the fault persists, contact technical support.
+
Error Codes
+
+
+
diff --git a/docs/ocr/umn/ocr_01_0078.html b/docs/ocr/umn/ocr_01_0078.html
new file mode 100644
index 00000000..3722ee9c
--- /dev/null
+++ b/docs/ocr/umn/ocr_01_0078.html
@@ -0,0 +1,12 @@
+
+
+Is It Possible to Call the OCR Service From a Different Region Than OBS Resources?
+Cross-region OBS is not supported, and the OBS region must match the region of the service being called.
+
For OBS resources with public read authorization, they can be accessed over the Internet and can support cross-region calls. Although this is convenient, there is a risk of sensitive information leakage, such as personal private data. It is recommended that you use OCR and OBS services in the same region to avoid this risk.
+
How Do I Handle the Error APIG.0307?
+If error message "The token must be updated." and error code "APIG.0307" are displayed when you call an OCR API, the token has expired and needs to be updated.
+
Perform the following steps to rectify the fault:
+
- The validity period of a token is 24 hours. Obtain the token again to call the API.
- Check whether the endpoint in the API URL is correct. Services deployed in different regions cannot be called across regions. If APIs in different regions are called, the token is invalid and error code APIG.0307 is displayed.
+
Does OCR Support Batch Processing?
+At present, the OCR API can only recognize one image per call. To process multiple images in batches, you would need to write additional code that allows for repeated calls to the OCR API.
+
How Do I Handle Image Quality Errors?
+Symptom
When calling an OCR API, the following error codes may occur related to image quality:
+
- Error code AIS.0102: Unsupported image format.
- Error code AIS.0103: Image size does not meet requirements.
- Error code AIS.0104: Unsupported image type or poor image quality.
+
+
Solution
- Check whether the image format and pixel meet the requirements by referring to Constraints and Limitations in Optical Character Recognition Service Overview.
- Verify that the Base64 encoding of the image is complete.
- Check the image quality to ensure that the text in the image is clear and legible to the naked eye.
- Verify that the API function matches the input image.
+
+
Can I Use an OCR API for a Different Function?
+No. It is not recommended to use one OCR API for a different function as each API has its own unique algorithms. Doing so may result in poor recognition results. For instance, using the General Text OCR API to recognize table images is not advised.
+
How Do I Handle the Error ModelArts.0212?
+If error message "Invalid Token header. The Token not contain project item." and error code "ModelArts.0212" are displayed when you call an OCR API, the token is invalid because the project information is missing.
+
OCR is a project-level service. To obtain the token for calling an OCR API, you need to set scope to project.
+
Token-based Authentication
+
+
+
diff --git a/docs/ocr/umn/ocr_01_0120.html b/docs/ocr/umn/ocr_01_0120.html
new file mode 100644
index 00000000..af198c6a
--- /dev/null
+++ b/docs/ocr/umn/ocr_01_0120.html
@@ -0,0 +1,13 @@
+
+
+Why Don't I Have Programming Access Permission?
+When calling an API, if you receive the error message "code":403,"message":"This user only supports console access, not programmatic access.", it means that you do not have the programming access permission.
+
IAM users do not have programming access permissions. You need to contact the main account holder to adjust the permissions.
+
The main account holder can log in to the IAM console, click Security Settings in the Operation column for the corresponding IAM user, and enter the settings page.
+
How Do I Handle the Error APIG.0101?
+The error message "The API does not exist or has not been published in the environment." is displayed.
+
- Check whether the API URL is correct. For example, check whether the project ID is included in the URL.
- Check whether the region information in the URI is consistent with the endpoint for subscribing to the service.
+
- Check whether the HTTP request method (such as POST and GET) is correct.
- For details about the URI, see the corresponding API page.
+
How Can I Improve Recognition Accuracy?
+- Use high-quality images with clear text and no glare as much as possible. When collecting images, try to increase the proportion of the area with text to be recognized, reduce the proportion of irrelevant background, and keep the text in the image clear and recognizable to the human eye.
- If the image has a rotation angle, the algorithm supports automatic correction, and it is recommended not to excessively tilt the image.
- Regarding image size, it is recommended that the longest side should not exceed 8192 pixels, the shortest side should not be less than 15 pixels, and the aspect ratio of the image should be within the common horizontal 5:1 (please refer to the API documentation of each service for specific requirements).
+
How Can I Improve Recognition Speed?
+The recognition speed is related to the image size, which affects the time required for network transmission and image Base64 decoding. To speed up image processing by the OCR API, it is recommended that you compress the images while maintaining the required level of resolution before processing them. It is recommended to upload images in JPG format.
+
Based on practical experience, it is generally recommended that small images of documents (with less text) should be below 1 MB, and large images of dense A4-sized documents should be below 2 MB.
+
Refer to the following code for how to compress images:
+
import cv2
+def resize_image(image, max_size):
+ """
+ This code is used to expand or downsize an image proportionally. It compares the long side of the image with the input parameter max_size. If the long side of the image exceeds max_size, the image is downsized proportionally. Otherwise, the original image is returned.
+ :param max_size: maximum length of the long side of an image. (Set this parameter based on site requirements. You are advised to set this parameter to a value as small as possible as long as the resolution requirement is met.)
+ :return: returns the downsized image or the original image.
+ """
+
+ height, width = image.shape[:2]
+ max_side = max(height, width)
+ if max_side > max_size:
+ scale = max_size / max_side
+ image = cv2.resize(image, None, fx=scale, fy=scale)
+
+ return image
+
+image = cv2.imread('test.png')
+image = resize_image(image, max_size=1024)
+
How Do I Convert a PDF to an Image?
+# -*- coding: utf-8 -*-
+import os
+import base64
+import fitz
+import io
+from PIL import Image
+from glob import glob
+
+class CovertPdfToJpg:
+ def __init__(self, file_path, save_root):
+ self.file_path = file_path
+ self.save_root = save_root
+
+ @staticmethod
+ def open_pdf(file):
+ return fitz.open(file)
+
+ @staticmethod
+ def get_trans(doc, page, min_side=0, max_side=0, rotate=0.0):
+ """ Create a scale object. """
+ region = doc[page].rect
+ scale = 1
+ if max_side > min_side > 0:
+ scale = min_side / min(region.width, region.height)
+ if max(region.width, region.height) * scale > max_side:
+ scale = max_side / max(region.width, region.height)
+ trans = fitz.Matrix(scale, scale).preRotate(rotate)
+ return trans
+
+ def page2pix(self, doc, page, trans):
+ """ Parse the current page as image data based on given parameters."""
+ # Obtain the PDF format of a specified page. Note that page parameters need to be pre-parsed to avoid any issues.
+ return doc[page].getPixmap(matrix=trans, alpha=False)
+
+ def pdf_to_jpg(self, width=1024, height=1400):
+ doc = self.open_pdf(self.file_path)
+ save_dir = os.path.join(self.save_root)
+ if not os.path.exists(save_dir):
+ os.makedirs(save_dir)
+ print("document", len(doc), doc.pageCount)
+ for i in range(len(doc)):
+ trans = self.get_trans(doc, i, width, height, rotate=0)
+ try:
+ pdf = self.page2pix(doc, i, trans)
+ except:
+ continue
+ image = pdf.getPNGData()
+ image = Image.open(io.BytesIO(image))
+ print(os.path.join(
+ save_dir, os.path.basename(self.file_path).replace('.pdf', '') + '_' + str(i + 1) + '.jpg'))
+ image.save(
+ os.path.join(save_dir, os.path.basename(self.file_path).replace('.pdf', '') + '_' + str(i + 1) + '.jpg'))
+ return
+
Why Is the Header Overflowing When I Call an OCR API Using Postman?
+If the error message "Error: Header overflow" is displayed when calling an API using Postman to obtain an authentication token, it means that the header has exceeded its limit. To resolve this issue, follow these steps:
+
Modify the environment variables of the operating system. In Windows 10, right click This PC and choose Properties. Click Advanced system settings. On the Advanced tab, click Environment Variables.... In the Environment Variables dialog box, click New... in the System variables area. Configure Variable name and Variable value as follows:
+
- Variable name: NODE_OPTIONS
- Variable value: --max-http-header-size=16384
+
How Do I Handle the Error APIG.0106?
+If error message and error code "error_msg":"Orchestration error.","error_code":"APIG.0106" are returned when an API is called, check whether the frontend and backend parameters configured for the API are correct.
+
This error is reported when the verification rule configured for frontend parameters of APIs is not met during API calling.
+
Rectify the fault using either of the following methods:
+
- Check whether all mandatory parameters are configured.
- Check whether parameter configuration rules are met. For example, the parameter value must be a number.
+
Getting Started
+Use Postman to call the General Text OCR API.
+
To call an OCR API, perform the following steps:
+
Step 1: Subscribing to a Service
+
Step 2: Configuring the Environment
+
Step 3: Using a Token for Authentication
+
Step 4: Calling the Service
+
Step 1: Subscribing to a Service
- Log in to the OCR management console.
Select a region based on service requirements. For details about the region where each service is deployed, see Regions and Endpoints.
+ - On the page displayed, select and subscribe to your desired APIs.
For this example, subscribe to the General Text OCR API.
+
+
+
Step 2: Configuring the Environment
Download and install Postman.
+
+
Step 3: Using a Token for Authentication
Tokens are used for identity authentication and permission management when calling an OCR API.
+
Before calling an OCR API, you need first use the "Obtaining a Token" API to obtain the token value. Then, pass the token value into the request header parameter of the OCR API to authenticate the user's API request and enable the OCR service to verify their identity.
+
The token is valid for 24 hours.
+
+
To obtain the token, perform the following steps:
+
- Log in to the cloud, hover your cursor over the username in the upper right corner, and choose My Credentials. On the API Credentials page displayed, obtain the username, domain name, and project ID.
+
Figure 1 Obtaining the username, domain name, and project ID
+ - Start Postman and create a POST request. For example, to obtain a token in the eu-de region, enter the following URL and request header parameter:
- URL: https://iam.eu-de.otc.t-systems.com/v3/auth/tokens
- Request header parameter: Content-Type; parameter value: application/json
+Figure 2 Entering the URL and request header parameter
+ - Enter the request body of the API for obtaining a token. Click Body, select raw, copy and enter the following code by referring to Figure 3, and enter the username, domain name, and password.
{
+ "auth": {
+ "identity": {
+ "methods": [
+ "password"
+ ],
+ "password": {
+ "user": {
+ "name": "username", // IAM username
+ "password": "********", // User password
+ "domain": {
+ "name": "domainname" // Domain name
+ }
+ }
+ }
+ },
+ "scope": {
+ "project": {
+ "name": "eu-de"
+ }
+ }
+ }
+}
+Figure 3 Request body
+ - Click Send to send the request. If the status code 201 is returned, the API is successfully called. In this case, click Headers, find and copy the X-Subject-Token value, which is the token.
Figure 4 Obtaining a token
+
+
+
Step 4: Calling the Service
- Create a POST request in Postman and enter the API request address. For details, see "APIs" in Optical Character Recognition API Reference.
Example: https://ocr.eu-de.otc.t-systems.com/v2/{project_id}/ocr/general-text
+ - Set two request header parameters by referring to Figure 5.
- KEY: Content-Type; VALUE: application/json
- KEY: X-Auth-Token; VALUE: the token value obtained in Figure 4
Figure 5 Request header
+
+ - Click Body, select raw, copy and enter the following code by referring to Figure 6, and enter the request body.
{
+ "image":"/9j/4AAQSkZJRgABAgEASABIAAD/4RFZRXhpZgAATU0AKgAAAA...",
+ "detect_direction":false,
+ "quick_mode":false
+ }
+Figure 6 Request body
+ - Click Send to send the request. If the status code 200 is returned, the API is successfully called and you can view the returned information in Postman.
Figure 7 Obtaining the calling result
+
+
- If you encounter an error message in Postman while calling APIs that indicates an invalid SSL certificate, such as self signed certificate, certificate has expired, or unable to verify the first certificate, you can resolve the issue by disabling SSL certificate verification in Postman's settings.
- For details about the request and response parameters of OCR APIs, see Optical Character API Reference.
+
+
+
+