C++ OCR API for Adding OCR Capabilities

Getting Started with Aspose.OCR for C++

The recommend way to install Aspose.OCR for C++ is using NuGet. Please use the following command for a smooth installation.

 NuGet\Install-Package Aspose.Ocr.Cpp -Version 23.4.0

Efficient Text Extraction in C++ Apps

Aspose.OCR for C++ provides a reliable and efficient approach for extracting text from a wide variety of file formats, including scanned documents, images, PDF files, multi-page TIFF, pixel array, receipts and so on. It utilizes sophisticated OCR algorithms to recognize and extract text with high accuracy, preserving the original formatting and structure. The library supports a wide range of languages, making it suitable for multilingual applications and enabling developers to extract text from diverse sources effortlessly. The following example shows how to extract text from TIFF image via C++ API.

std::string image_path = "source.tiff";
const size_t len = 4096;
wchar_t buffer[len] = { 0 };
RecognitionSettings settings;
settings.language_alphabet = language::ukr;
size_t res_len = aspose::ocr::page_tiff("1.tif;2.tif", buffer, len, set);
std::wcout << buffer;

Read Certain Areas of Images via C++

Aspose.OCR for C++ makes it easy for software developers read a particular area of an image and extract text from that areas or regions inside C++ applications. This feature is particularly useful when you only need to extract text from specific sections of an image and want to exclude irrelevant content. The library offers a simple and efficient method to achieve this. Below is an example code snippet demonstrating how to read certain areas of images using Aspose.OCR for C++.

// Load the image
System::SharedPtr imageStream = System::MakeObject(new System::IO::FileStream(u"image.jpg", System::IO::FileMode::Open));
// Initialize OCR engine
System::SharedPtr ocrEngine = System::MakeObject();
// Set the image for OCR
ocrEngine->Image = imageStream;
// Set the rectangle coordinates for the specific area to read
System::SharedPtr areaRect = System::MakeObject(10, 10, 200, 100);
ocrEngine->Config->SetArea(areaRect);
// Perform OCR on the specified area
ocrEngine->Process();
// Retrieve the extracted text from the specific area
System::String extractedText = ocrEngine->Text;
// Display the extracted text
std::cout << "Extracted Text: " << extractedText.ToUtf8String() << std::endl;

Image Preprocessing via C++ API

Aspose.OCR for C++ has provided a standardized way to prepare your content for OCR and achieve accurate OCR results. The library offers a range of advanced image preprocessing techniques. These techniques enhance image quality, correct perspective distortion, remove noise, and optimize the text recognition process. By employing image preprocessing, developers can significantly improve OCR accuracy, especially when dealing with challenging images or documents with complex layouts. Multiple preprocessing filters can be applied to the same image to further improve the recognition quality.

 // Recognition settings
std::string image_path = "source.png";
const size_t len = 4096;
wchar_t buffer[len] = { 0 };
RecognitionSettings settings;
settings.auto_denoising = true;
size_t res_len = aspose::ocr::page_settings(image_path.c_str(), buffer, len, settings);
std::wcout << buffer;

// apply Preprocessing filter

std::string image_path = "source.png";
custom_preprocessing_filters filters_;
filters_.filter_1 = OCR_IMG_PREPROCESS_AUTODENOISING;
asposeocr_preprocess_page_and_save(image_path.c_str(), "result.png", filters_);
 

Save Recognition Results in Other Formats

Aspose.OCR for C++ enables software developers to recognize text from numerous popular file formats, such as PDF, JPEG, PNG, TIFF, BMP and more. The API allows developers to save recognition results in multiple formats so they can be shared, stored in a database, displayed, or analyzed. Software developers can save recognition results as file, text, JSON or XML. The library allows setting recognition confidence thresholds, enabling software developers to filter out text with low confidence levels. This feature proves invaluable when dealing with large volumes of text, ensuring that only reliable and accurate results are extracted. The following examples shows how to save recognition results as file using C++ commands.

directory dir(".");
const string current_dir = dir.full_name();
const string image = current_dir + "p.png";
const size_t len = 6000;
wchar_t buffer[len] = { 0 };
RecognitionSettings settings;
settings.save_format = file_format::docx;
aspose::ocr::page_save(image.c_str(), "result.docx", settings);