Thank you for your inquiry. Aspose.OCR only takes in image type of documents to perform OCR operation. This means that in order to perform OCR operation, you have to supply only image as parameter. You can perform OCR operation on PDF files but for that you have to use Aspose.OCR in combination with Aspose.PDF libraries. Aspose.PDF will process the input PDF file and convert PDF pages to images. Those images will then can be used as input parameter to OCR function. You can read/extract only invoice number for image by using Text Recognition Block functionality. It is recommended that you may go through the following online documentation articles for detail. You can download the latest version of Aspose.OCR from NuGet.
Further to update you that we are currently working on Aspose.OCR for Cloud. Development on Aspose.OCR for .NET is temporarily suspended. You can try Aspose.OCR for Cloud. You may post your inquiry on OCR for Cloud support forum for further assistance.