Getting recognition results as text

Recognition results in plain text with line breaks are returned by default from any recognition method, unless you directly specify another output format in recognition settings.

std::string image_path = "source.png";
// prepare buffer for result (in symbols, len_byte = len * sizeof(wchar_t))
const size_t len = 4096;
wchar_t buffer[len] = { 0 };
// extract text
size_t size = aspose::ocr::page(image_path.c_str(), buffer, len);
// print result
std::wcout << buffer << L"\n";

You can also extract text blocks from specific image areas or get individual lines.