直接上源码，这是一个收费的且不开源的库，测试效果也不太理想
亲测效果：

1. 对于相同字体。非倾斜的，比如http://dz.bjjtgl.gov.cn/service/checkCode.do，识别率还是挺高的；
2. 对于斜体或者其他变异的，如下代码识别率就很低了，可能需要调整识别引擎的参数了，大家自己查找测试下。
3. 这个测试版本的，会有弹出对话框 框框～～

package com.example;

import java.awt.image.BufferedImage;
import java.io.File;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.util.Date;

import javax.imageio.ImageIO;

import org.apache.commons.httpclient.HttpClient;
import org.apache.commons.httpclient.HttpStatus;
import org.apache.commons.httpclient.methods.GetMethod;
import org.apache.commons.io.IOUtils;
import com.asprise.ocr.Ocr;

public class ReadImg {
    public static void main(String[] args) throws IOException {
        HttpClient httpClient = new HttpClient();
        GetMethod getMethod = new GetMethod("http://dz.bjjtgl.gov.cn/service/checkCode.do");
//      GetMethod getMethod = new GetMethod("https://dynamic.12306.cn/otsweb/passCodeAction.do?rand=sjrand");
        int statusCode = httpClient.executeMethod(getMethod);
        if (statusCode != HttpStatus.SC_OK) {
            System.err.println("Method failed: " + getMethod.getStatusLine());
            return ;
        }
        String picName = "F:\\img\\";
        File filepic=new File(picName);
        if(!filepic.exists())
            filepic.mkdir();
        File filepicF=new File(picName+new Date().getTime() + ".jpg");
        InputStream inputStream = getMethod.getResponseBodyAsStream();
        OutputStream outStream = new FileOutputStream(filepicF);
        IOUtils.copy(inputStream, outStream);
        outStream.close();

        Ocr.setUp(); // one time setup
        Ocr ocr = new Ocr(); // create a new OCR engine
        ocr.startEngine("eng", Ocr.SPEED_FASTEST); // English
        String s = ocr.recognize(new File[] {filepicF},Ocr.RECOGNIZE_TYPE_TEXT, Ocr.OUTPUT_FORMAT_PLAINTEXT);
        System.out.println("Result: " + s);
        System.out.println("num is:" + s.replace(",", "").replace("i", "1").replace(" ", "").replace("'", "").replace("o", "0").replace("O", "0").replace("g", "6").replace("B", "8").replace("s", "5").replace("z", "2"));
        // ocr more images here ...
        ocr.stopEngine();
    }
}

需要的java包

注意：主要的jar包

1. aocr.jar -- 去Asprise官网下载最新jar包
2. commons-codec.jar
3. commons-httpclient-3.1.jar
4. commons-io.jar
5. commons-logging-1.0.4.jar

参考链接

1. Java 识别图片验证码