Optical Character Recognition (OCR) has become an indispensable technology in modern software development, enabling computers to convert different types of documents, such as images or scanned documents, into editable and searchable text. In the Java ecosystem, integrating OCR capabilities enables powerful document processing, data extraction, and accessibility features. This guide explores how to implement OCR in Java using the Spire.OCR for Java library, leveraging its advanced features to extract text from images with or without positional coordinates.

Table of Contents:

• Why Perform OCR in Java?
• Environment Setup & Configuration
• Extract Text from an Image in Java
• Extract Text with Coordinates from an Image in Java
• Advanced OCR Techniques
  • 1. Enable Auto-Rotation of Images
  • 2. Preserve the Original Layout
• FAQs (Supported Languages and Image Formats)
• Conclusion & Free License

Why Perform OCR in Java?

OCR technology transforms images into machine-readable text. Java developers leverage OCR for:

• Automating invoice/receipt processing
• Digitizing printed records and forms
• Enabling text search in scanned documents
• Extracting structured data with spatial coordinates

Spire.OCR for Java stands out with its:

• Advanced OCR algorithms ensure accurate text recognition.
• Support for multiple image formats and languages.
• Batch processing of multiple images, improving efficiency.
• Ease of integration with Java applications

Environment Setup & Configuration

Step 1: Add Spire.OCR to Your Project

Add the following to your pom.xml:

<repositories>
    <repository>
        <id>com.e-iceblue</id>
        <name>e-iceblue</name>
        <url>https://repo.e-iceblue.com/nexus/content/groups/public/</url>
    </repository>
</repositories>
<dependency>
    <groupId>e-iceblue</groupId>
    <artifactId>spire.ocr</artifactId>
    <version>2.1.1</version>
</dependency>

Alternatively, you can download the Spire.OCR for Java library, and then add it to your Java project.

Step 2: Download the OCR Model Files

Spire.OCR for Java relies on pre-trained models to extract image text. Download the OCR model files for your OS:

• Windows x64
• Linux x64
• macOS

After downloading, unzip the files to a directory (e.g., F:\win-x64)

Extract Text from an Image in Java

The following is a basic text extraction example, i.e., reading text from an image and saving it to .txt files.

Use Cases:

• Archiving printed materials (e.g., books, newspapers) as digital text.
• Converting images (e.g., screenshots, memes, signs) into shareable/editable text.

Java code to get text from an image:

import com.spire.ocr.*;
import java.io.BufferedWriter;
import java.io.FileWriter;
import java.io.IOException;

public class ocrJava {
    public static void main(String[] args) throws Exception {
            // Create an instance of the OcrScanner class
            OcrScanner scanner = new OcrScanner();

            // Create an instance of the ConfigureOptions class
            ConfigureOptions configureOptions = new ConfigureOptions();

            // Set the path to the OCR model
            configureOptions.setModelPath("F:\\win-x64");

            // Set the language for text recognition
            configureOptions.setLanguage("English");

            // Apply the configuration options to the scanner
            scanner.ConfigureDependencies(configureOptions);

            // Extract text from an image
            scanner.scan("sample.png");
            String text = scanner.getText().toString();

            // Save the extracted text to a text file
            try (BufferedWriter writer = new BufferedWriter(new FileWriter("output.txt"))) {
                writer.write(text);
            } catch (IOException e) {
            e.printStackTrace();
        }
    }
}

Key Steps:

1. Initialize Scanner: OcrScanner handles OCR operations

2. Configure Settings:

   • setModelPath(): Location of OCR model files
   • setLanguage(): Supports multiple languages (Chinese, Spanish, etc.)

3. Process Image: scan() performs OCR on the image file

4. Export Text: getText() retrieves results as a string

Output:

Extract Text with Coordinates from an Image in Java

In some applications, knowing the position of text within the image is crucial, for example, when processing structured documents like invoices or forms. The Spire.OCR library supports this through its block-based text extraction feature.

Use Cases:

• Automated invoice processing (locate amounts, dates)
• Form data extraction (identify fields by position)

Java code to extract text with coordinates:

import com.spire.ocr.*;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Paths;
import java.util.ArrayList;
import java.util.List;

public class ExtractWithCoordinates {
    public static void main(String[] args) throws Exception {
            // Create an instance of the OcrScanner class
            OcrScanner scanner = new OcrScanner();

            // Create an instance of the ConfigureOptions class
            ConfigureOptions configureOptions = new ConfigureOptions();

            // Set the path to the OCR model
            configureOptions.setModelPath("F:\\win-x64");

            // Set the language for text recognition
            configureOptions.setLanguage("English");

            // Apply the configuration options to the scanner
            scanner.ConfigureDependencies(configureOptions);

            // Extract text from an image
            scanner.scan("invoice.png");
            IOCRText text = scanner.getText();

            // Create a list to store information
            List<String> results = new ArrayList<>();

            // Iterate through each detected text block
            for (IOCRTextBlock block : text.getBlocks()) {
                // Add the extracted text and coordinates to the list
                results.add("Block Text: " + block.getText());
                results.add("Coordinates: " + block.getBox());
                results.add("---------");
            }

            // Save to a text file
            try {
                Files.write(Paths.get("ExtractWithCoordinates.txt"), results);
            } catch (IOException e) {
                e.printStackTrace();
            }
        }
    }

Key Features:

1. Text & Coordinate Extraction:

   • Iterates over text blocks detected in the image.

   • getText(): Returns an IOCRText object containing recognized text.

   • getBox(): Returns bounding box coordinates [x, y, width, height]

2. Efficient File Writing:

   • Uses Java NIO for efficient file operations.

   • Outputs human-readable results for easy parsing.

Output:

Advanced OCR Techniques

1. Enable Auto-Rotation of Images

For accurate processing of skewed or rotated images, enable the SetAutoRotate() method to rotate the image to the correct upright position automatically:

ConfigureOptions configureOptions = new ConfigureOptions();
configureOptions.SetAutoRotate(true);

2. Preserve the Original Layout

If you need to preserve the original visual layout in the image (e.g., table, multi-column layout), initialize the VisualTextAligner class to enhance the formatting of the extracted text.

// Align the recognized text (for better formatting)
VisualTextAligner visualTextAligner = new VisualTextAligner(scanner.getText());
String scannedText = visualTextAligner.toString();

FAQs (Supported Languages and Image Formats)

Q1: What image formats does Spire.OCR for Java support?

A: Spire.OCR for Java supports all common formats:

• PNG
• JPEG/JPG
• BMP
• TIFF
• GIF

Q2: What languages does Spire.OCR for Java support?

A: Multiple languages are supported:

• English (default)
• Chinese (Simplified and Traditional)
• Japanese
• Korean
• German
• French

Q3: How to improve OCR accuracy?

A: To boost accuracy:

• Use high-quality images (300+ DPI, well-lit, low noise).
• Preprocess images (adjust contrast, remove artifacts) before scanning.
• Specify the correct language(s) for the text in the image.

Q4: Can Spire.OCR for Java extract text from scanned PDFs?

A: This task requires the Spire.PDF for Java integration to convert PDFs to images or extract images from scanned PDFs first, and then use the above Java examples to get text from the images.

Conclusion & Free License

This guide provides a comprehensive roadmap for mastering Spire.OCR for Java, equipping developers with the knowledge to seamlessly integrate powerful OCR capabilities into their Java applications. From step-by-step installation to implementing basic and advanced text extraction, every critical aspect of recognizing text from images has been covered. Whether you're new to OCR or experienced, you now have the tools to convert images to text simply and effectively.

Request a 30-day free trial license here to enjoy unlimited OCR processing in Java.