Code:
//Initialize an instance of OcrEngine
OcrEngine ocrEngine = new OcrEngine();
//Clear notifier list
ocrEngine.clearNotifies();
//Clear recognition blocks
ocrEngine.getConfig().clearRecognitionBlocks();
//Add 3 rectangle blocks to user defined recognition blocks
ocrEngine.getConfig().addRecognitionBlock(RecognitionBlock.createTextBlock(450, 1420, 1700, 120));
//Ignore everything else on the image other than the user defined recognition blocks
ocrEngine.getConfig().setDetectTextRegions(false);
//Set Image property by loading an image from file path
ocrEngine.setImage(ImageStream.fromFile(“Converted_Image.jpg”));
//Run recognition process
if (ocrEngine.process())
{
//Retrieve user defined blocks that determines the page layout
List blocks = ocrEngine.getConfig().getRecognitionBlocks();
//Loop over the list of blocks
for (IRecognitionBlock block : blocks)
{
//Check if block has recognition data
if (block.getRecognitionData() == null)
{
System.out.println(“Null”);
continue;
}
//Display dimension & size of rectangle that defines the recognition block
System.out.println(“Block” + block.getRectangle());
//Display the recognition results
IRecognizedTextPartInfo textPartInfo = (IRecognizedTextPartInfo)block.getRecognitionData();
System.out.println(“Text” + textPartInfo.getText());
}
}
Output:
The aee AcrobacReader is easy zo dowload md can be freely dismibured by
myone.
Attached the source image.
Although now the text blocks are in the right place, OCR is not reading the characters correctly. Can you please guide us on how to improve the output?