Tess4J内存访问异常：setDatapath与实例管理深度解析

import net.sourceforge.tess4j.ITesseract;
import net.sourceforge.tess4j.Tesseract;
import net.sourceforge.tess4j.TesseractException;

import java.awt.image.BufferedImage;
import java.io.File;
import javax.imageio.ImageIO;

public class TesseractConfigExample {

    private static ITesseract tesseractInstance; // 声明为静态或成员变量

    public static void initializeTesseract() {
        if (tesseractInstance == null) {
            tesseractInstance = new Tesseract();
            // 设置语言数据路径，仅调用一次
            // 确保 'path/to/your/tessdata' 指向包含 'eng.traineddata' 等文件的目录
            tesseractInstance.setDatapath("path/to/your/tessdata"); 
            tesseractInstance.setLanguage("eng"); // 设置识别语言
            // 其他配置，如设置PSM模式等
            // tesseractInstance.setPageSegMode(ITesseract.DEFAULT_PAGE_SEG_MODE);
            // tesseractInstance.setOcrEngineMode(ITesseract.DEFAULT_OCR_ENGINE_MODE);
        }
    }

    public String performOcr(BufferedImage image) throws TesseractException {
        if (tesseractInstance == null) {
            throw new IllegalStateException("Tesseract instance not initialized. Call initializeTesseract() first.");
        }
        return tesseractInstance.doOCR(image);
    }

    public static void main(String[] args) {
        initializeTesseract(); // 在程序启动时或首次使用前调用一次

        try {
            // 模拟读取图片
            BufferedImage image = ImageIO.read(new File("path/to/your/image.png"));

            TesseractConfigExample ocrProcessor = new TesseractConfigExample();
            String result = ocrProcessor.performOcr(image);
            System.out.println("OCR Result: " + result);

            // 再次进行OCR操作，无需重复调用setDatapath
            // BufferedImage anotherImage = ImageIO.read(new File("path/to/another/image.png"));
            // String anotherResult = ocrProcessor.performOcr(anotherImage);
            // System.out.println("Another OCR Result: " + anotherResult);

        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}

import net.sourceforge.tess4j.ITesseract;
import net.sourceforge.tess4j.Tesseract;
import net.sourceforge.tess4j.TesseractException;

import java.awt.image.BufferedImage;
import java.io.File;
import javax.imageio.ImageIO;
import java.util.concurrent.atomic.AtomicInteger;

public class TesseractInstanceManager {

    private ITesseract tesseractInstance;
    private final String datapath;
    private final String language;
    private final int reinitializeThreshold; // 设定一个阈值，例如100次操作
    private AtomicInteger operationCount;

    public TesseractInstanceManager(String datapath, String language, int threshold) {
        this.datapath = datapath;
        this.language = language;
        this.reinitializeThreshold = threshold;
        this.operationCount = new AtomicInteger(0);
        initializeTesseract(); // 首次初始化
    }

    private synchronized void initializeTesseract() {
        // 销毁旧实例（如果存在）
        if (tesseractInstance != null) {
            // Tess4J没有直接的dispose方法来释放原生资源，
            // 销毁引用并等待GC是间接手段，但不足以立即释放原生内存。
            // 重新创建实例是更可靠的刷新原生资源的方式。
            tesseractInstance = null; 
            System.gc(); // 提示JVM进行垃圾回收，但这并不能保证原生内存立即释放
            System.out.println("Tesseract instance disposed and re-initialized.");
        }
        tesseractInstance = new Tesseract();
        tesseractInstance.setDatapath(datapath);
        tesseractInstance.setLanguage(language);
        operationCount.set(0); // 重置操作计数
    }

    public String performOcr(BufferedImage image) throws TesseractException {
        // 检查是否达到重新初始化阈值
        if (operationCount.incrementAndGet() >= reinitializeThreshold) {
            initializeTesseract(); // 重新初始化实例
        }
        return tesseractInstance.doOCR(image);
    }

    public static void main(String[] args) {
        String tessdataPath = "path/to/your/tessdata"; // 替换为你的tessdata路径
        String lang = "eng";
        int threshold = 5; // 为了演示，设置为5次操作后重新初始化

        TesseractInstanceManager manager = new TesseractInstanceManager(tessdataPath, lang, threshold);

        try {
            for (int i = 0; i < 15; i++) { // 模拟15次OCR操作
                // 模拟读取图片
                BufferedImage image = ImageIO.read(new File("path/to/your/image.png")); // 确保图片存在
                String result = manager.performOcr(image);
                System.out.println("Operation " + (i + 1) + ", OCR Result: " + result);
                Thread.sleep(100); // 模拟处理时间
            }
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
}