Implementing ID Card and Business License Recognition in Spring Boot Using Tesseract OCR and OpenCV

In Spring Boot, recognizing information like ID numbers and business license codes from images can be achieved through a series of steps: image preprocessing, text detection, OCR recognition, and information extraction.

Steps:

Image preprocessing: Adjust size, contrast, brightness, convert to grayscale, apply Gaussian blur, and edge detection using OpenCV to improve OCR accuracy.

Text detection: Locate text regions in the image with image‑processing algorithms.

Text recognition: Feed the detected regions to an Optical Character Recognition engine (Tesseract) to obtain raw text.

Information extraction: Use regular expressions to pull out ID numbers, business license numbers, etc., from the recognized text.

Tools and libraries:

Tesseract OCR: Open‑source OCR engine supporting multiple languages.

OpenCV: Computer‑vision library for image processing.

Spring Boot: Framework for building the backend service.

Dependency configuration (pom.xml):

<dependencies>
<dependency>
<groupId>
org.springframework.boot
</groupId>
<artifactId>
spring-boot-starter-web
</artifactId>
</dependency>
<dependency>
<groupId>
org.bytedeco
</groupId>
<artifactId>
javacv
</artifactId>
<version>
1.5.6
</version>
</dependency>
<dependency>
<groupId>
org.bytedeco
</groupId>
<artifactId>
javacpp
</artifactId>
<version>
1.5.6
</version>
</dependency>
<!-- other dependencies -->
</dependencies>

OCR service implementation:

public class OCRService {
    public String doOCR(String imagePath) {
        TessBaseAPI tessBaseAPI = new TessBaseAPI();
        // Initialize Tesseract
        if (tessBaseAPI.Init(".", "ENG") != 0) {
            System.err.println("Could not initialize Tesseract.");
            return null;
        }
        // Read image
        Mat image = opencv_imgcodecs.imread(imagePath);
        // Preprocess (e.g., convert to gray)
        cvtColor(image, image, COLOR_BGR2GRAY);
        // Set image for Tesseract
        tessBaseAPI.SetImage(image.data(), image.cols(), image.rows(), 1, image.step());
        // Perform OCR
        BytePointer outText = tessBaseAPI.GetUTF8Text();
        String result = outText.getString();
        outText.deallocate();
        tessBaseAPI.End();
        return result;
    }
}

Information extraction example:

public class InfoExtractor {
    public String extractIDNumber(String text) {
        Pattern pattern = Pattern.compile("[0-9]{18}|[0-9]{15}");
        Matcher matcher = pattern.matcher(text);
        if (matcher.find()) {
            return matcher.group();
        }
        return null;
    }
    // Additional methods for business license, etc.
}

Document processing flow:

public class DocumentProcessor {
    public String processDocument(String imagePath) {
        OCRService ocrService = new OCRService();
        String text = ocrService.doOCR(imagePath);
        if (text != null && !text.isEmpty()) {
            InfoExtractor extractor = new InfoExtractor();
            String idNumber = extractor.extractIDNumber(text);
            return idNumber; // returns extracted ID number
        }
        return null;
    }
}

Image preprocessing class (OpenCV):

public class ImagePreprocessing {
    public Mat preprocessImage(String imagePath) {
        // Read image
        Mat image = imread(imagePath);
        // Convert to gray
        Mat gray = new Mat();
        cvtColor(image, gray, COLOR_BGR2GRAY);
        // Gaussian blur
        Mat blurred = new Mat();
        GaussianBlur(gray, blurred, new Size(3, 3), 0);
        // Edge detection
        Mat edged = new Mat();
        Canny(blurred, edged, 75, 200);
        return edged;
    }
}

Finally, integrate the preprocessing step into OCRService by creating an ImagePreprocessing instance, obtaining the preprocessed Mat , and passing its data to Tesseract. This combination of OpenCV and Tesseract within a Spring Boot backend improves the accuracy of text recognition for ID cards and business licenses.