OCR text detection with google APIs C#, implement text detection using google API with C#,
Read more on my blog :
After searching more than 3 hrs to know how to implement text detection using google API i decided to write this post, hope it help c# developer to know how to use and implement Vision API,
Source Code :
1- Create server key on google console
- Go to http://console.developers.google.com/ and create new project or use existing project
- From Overview page enable vision API
-
You need to create new bill information to use vision API, it's fully free for 1st 1000 request/ month you can check https://cloud.google.com/vision/docs/pricing
-
After Enable Google Cloud Vision API go to Credentials page and create new server key
-
Create key by fill all requested data and download Json file
2- Create Visual studio project I'll use vs 2015 and .NET framework 4.5
- Add google Nuget packages : Google.Apis.Vision.v1 by right click on project and then manage nuget package .
- Register the JSON file in the environment variable.
- Environment variable name is "GOOGLE_APPLICATION_CREDENTIALS"
private VisionService CreateAuthorizedClient()
{
GoogleCredential credential =
GoogleCredential.GetApplicationDefaultAsync().Result;
// Inject the Cloud Vision scopes
if (credential.IsCreateScopedRequired)
{
credential = credential.CreateScoped(new[]
{
VisionService.Scope.CloudPlatform
});
}
return new VisionService(new BaseClientService.Initializer
{
HttpClientInitializer = credential,
GZipEnabled = false
});
}
- Set it to the environment variable and make a request to the Cloud Vision API using the service account key.
- Create batch request to upload image and get result :
private int DetectTextWord(
VisionService vision, byte[] getImage, ref string FullText)
{
int result = 1;
Console.WriteLine("Detecting image to texts...");
// Convert image to Base64 encoded for JSON ASCII text based request
string imageContent = Convert.ToBase64String(getImage);
try
{
// Post label detection request to the Vision API
var responses = vision.Images.Annotate(
new BatchAnnotateImagesRequest()
{
Requests = new[] {
new AnnotateImageRequest() {
Features = new [] { new Feature() { Type =
"TEXT_DETECTION"}},
Image = new Image() { Content = imageContent }
}
}
}).Execute();
if (responses.Responses != null)
{
FullText = responses.Responses[0].TextAnnotations[0].Description;
Console.WriteLine("SUCCESS：Cloud Vision API Access.");
result = 0;
}
else
{
FullText = "";
Console.WriteLine("ERROR : No text found.");
result = -1;
}
}
catch
{
FullText = "";
Console.WriteLine("ERROR : Not Access Cloud Vision API.");
result = -1;
}
return result;
}
- Vision api have a lot of type like face detection and other types I use text_detiction https://cloud.google.com/vision/reference/rest/v1/images/annotate#Type
- As a precaution, try-catch describes exception handling. However, it is not a smart description.
- Excute request to get result
public int getTextAndRoi(byte[] getImage, ref string FullText)
{
int result = 1;
GCPVisonAPI sample = new GCPVisonAPI();
// Create a new Cloud Vision clieuthorint azed via Application
// Default Credentials
VisionService vision = sample.CreateAuthorizedClient();
// Use the client to get label annotations for the given image
string getFullText = "";
result = sample.DetectTextWord(vision, getImage, ref getFullText);
FullText = getFullText;
return result;
}
- As a result of the API, you can get the full text of OCR, each word list, region list of each word ·
- This time we will receive the full text of the OCR result.
Google reference :