Implement OCR in Java with Tess4j and SpringBoot in Just a Few Lines

Introduction

Ever needed to copy text from an image but had to type it manually? Java can perform OCR (optical character recognition) easily without heavyweight external tools. By using the Tess4j library, you can integrate OCR into your Java applications with just a few lines of code.

Demo

The final OCR result looks like this:

Implementation

1. Description

We will use SpringBoot together with Tess4j, a Java wrapper for the Tesseract OCR engine. Tess4j works for scanned documents, image text extraction, and automated screenshot reading, and can be exposed as a lightweight RESTful service.

2. Code

2.1 Add Dependency

<dependency>
  <groupId>net.sourceforge.tess4j</groupId>
  <artifactId>tess4j</artifactId>
</dependency>

2.2 Initialize Tesseract Engine

/**
 * TesseractOcr model loader
 */
@Slf4j
@Getter
@Component
public class TesseractOcrModelService {
    private final Tesseract tesseract = new Tesseract();
    public TesseractOcrModelService() {
        try {
            // Path to trained data (may need custom handling when packaged as a jar)
            String folderPath = new ClassPathResource("tess_data").getFile().getAbsolutePath();
            tesseract.setDatapath(folderPath);
            tesseract.setPageSegMode(OEM_TESSERACT_LSTM_COMBINED);
            tesseract.setLanguage("chi_sim"); // Simplified Chinese
        } catch (Exception e) {
            throw new RuntimeException(e);
        }
    }
}

2.3 RESTful API

/**
 * OCR controller
 */
@RestController
@RequestMapping("ocr")
@RequiredArgsConstructor
public class OcrController {
    private final TesseractOcrModelService tesseractOcrModelService;

    @PostMapping("/detection")
    public Result<String> ocrDetection(MultipartFile file) {
        try {
            Tesseract tesseract = tesseractOcrModelService.getTesseract();
            return Result.success(tesseract.doOCR(ImageIO.read(file.getInputStream())));
        } catch (Exception e) {
            throw new RuntimeException("ImageIO.read(file.getInputStream()) parsing error");
        }
    }
}

Source Code

The complete source can be found at the repository link below.

https://gitee.com/fateyifei/yf

Conclusion

Tess4j works well for IDs, phone numbers, and English words, but its free training data may struggle with Chinese. For higher accuracy, consider custom training, or use third‑party OCR services such as Google Cloud Vision, Microsoft Azure OCR, or Amazon Textract. OCR can also be applied to document digitization, automated data entry, license‑plate recognition, and handwriting conversion.