团队冲刺第九天
今天上课展示了第一阶段冲刺的成果,可以做到读取传输的图片上的表格,并输出到web页面,但是要保存到本地形成excel表格还有点问题,
还需要解决,观看了其他组的展示后,我看到了他们的优点,比如页面好看,简洁,功能多多,也认识到了自己的不足,还有许多地方需要改进。
package com.example.demo;
import com.sun.org.slf4j.internal.Logger;
import com.sun.org.slf4j.internal.LoggerFactory;
import net.sourceforge.tess4j.ITesseract;
import net.sourceforge.tess4j.Tesseract;
import net.sourceforge.tess4j.TesseractException;
import org.opencv.core.*;
import org.opencv.core.Rect;
import org.opencv.highgui.HighGui;
import org.opencv.imgcodecs.Imgcodecs;
import org.opencv.imgproc.Imgproc;
import org.opencv.objdetect.Objdetect;
import javax.imageio.ImageIO;
import java.awt.image.BufferedImage;
import java.io.ByteArrayInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
import java.util.List;
public class TableScanner {
private static final Logger logger = LoggerFactory.getLogger(TableScanner.class);
private static final String TESSERACT_DATA_PATH = "/usr/share/tesseract-ocr/4.00/tessdata"; // Tesseract OCR数据路径
private String imagePath; // 图像文件路径
private String tablePath; // 表格文件保存路径
public TableScanner(String imagePath, String tablePath) {
this.imagePath = imagePath;
this.tablePath = tablePath;
}
public void scanTable() throws IOException {
// 加载OpenCV库
System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
// 读取图像文件
Mat image = Imgcodecs.imread(imagePath);
// 转换为灰度图像
Mat gray = new Mat();
Imgproc.cvtColor(image, gray, Imgproc.COLOR_BGR2GRAY);
// 对图像进行二值化处理
Mat binary = new Mat();
Imgproc.threshold(gray, binary, 0, 255, Imgproc.THRESH_BINARY_INV | Imgproc.THRESH_OTSU);
// 进行表格检测
Rect tableRect = detectTable(binary);
if (tableRect == null) {
logger.warn("Failed to detect table in image: {}", imagePath);
return;
}
// 提取表格区域
Mat table = new Mat(image, tableRect);
// 进行表格识别
ITesseract tesseract = new Tesseract();
tesseract.setDatapath(TESSERACT_DATA_PATH);
tesseract.setLanguage("eng"); // 使用英文语言库
String text = null;
BufferedImage bufferedImage = null;
try {
MatOfByte matOfByte = new MatOfByte();
Imgcodecs.imencode(".jpg", table, matOfByte);
byte[] byteArray = matOfByte.toArray();
InputStream in = new ByteArrayInputStream(byteArray);
bufferedImage = ImageIO.read(in);
} catch (IOException e) {
logger.error("Failed to convert Mat to BufferedImage: {}", e.getMessage());
return;
}
try {
text = tesseract.doOCR(bufferedImage);
} catch (TesseractException e) {
logger.error("Failed to recognize table in image: {}", imagePath, e);
return;
}
// try {
// text = tesseract.doOCR(table);
// } catch (TesseractException e) {
// logger.error("Failed to recognize table in image: {}", imagePath, e);
// return;
// }
// 保存为CSV文件
Path tableFile = Paths.get(tablePath);
Files.write(tableFile, text.getBytes());
}
/**
* 检测图像中的表格区域
*/
private Rect detectTable(Mat binary) {
// 进行轮廓检测
List<MatOfPoint> contours = Lists.newArrayList();
Mat hierarchy = new Mat();
Imgproc.findContours(binary, contours, hierarchy, Imgproc.RETR_EXTERNAL, Imgproc.CHAIN_APPROX_SIMPLE);
// 查找最大的矩形轮廓
Rect maxRect = null;
double maxArea = 0;
for (MatOfPoint contour : contours) {
Rect rect = (Rect) Imgproc.boundingRect(contour);
double area = rect.width * rect.height;
if (area > maxArea && isTable(rect)) {
maxRect = rect;
maxArea = area;
}
}
return maxRect;
}
/**
* 判断矩形是否为表格
*/
private boolean isTable(Rect rect) {
return rect.width > 50 && rect.height > 50;
}
}

浙公网安备 33010602011771号