瀏覽代碼

Merge branch 'release_real_202102' of https://git.malls.iformall.com/server/formallProject into release_real_202102

release_toaliyun_real
xhxu 5 年之前
父節點
當前提交
3af894e290
共有 3 個檔案被更改,包括 7 行新增6 行删除
  1. +3
    -2
      mallinkOcr/src/main/java/com/iformall/ocr/FormallTess4j.java
  2. +4
    -4
      mallinkOcr/src/main/java/com/iformall/ocr/FormallTess4jTrain.java
  3. 二進制
      mallinkOcr/src/main/resources/tessdata/test1.traineddata

+ 3
- 2
mallinkOcr/src/main/java/com/iformall/ocr/FormallTess4j.java 查看文件

@@ -125,7 +125,7 @@ public class FormallTess4j {
}
private static String ocrLocalImgFile(String languagePath,String filePath,String fontName) throws Exception {
File imageFile = new File(filePath,fontName);
File imageFile = new File(filePath);
String result = testDoOCR_File(languagePath,imageFile,fontName);
deletetempFile(imageFile);
return result;
@@ -155,7 +155,8 @@ public class FormallTess4j {
public static void main(String[] args) {
try {
System.out.println(ocrNetImgFile(testResourcesLanguagePath,"https://gimg2.baidu.com/image_search/src=http%3A%2F%2Fimg.my0832.com%2Fattachments%2Fbbs%2F20140424%2F201442413083356803_740_1186.jpg&refer=http%3A%2F%2Fimg.my0832.com&app=2002&size=f9999,10000&q=a80&n=0&g=0n&fmt=jpeg?sec=1613903848&t=985fc4c6c7b7ee82cb98f97ab5bf1459","test1"));
//System.out.println(ocrNetImgFile(testResourcesLanguagePath,"https://gimg2.baidu.com/image_search/src=http%3A%2F%2Fimg.my0832.com%2Fattachments%2Fbbs%2F20140424%2F201442413083356803_740_1186.jpg&refer=http%3A%2F%2Fimg.my0832.com&app=2002&size=f9999,10000&q=a80&n=0&g=0n&fmt=jpeg?sec=1613903848&t=985fc4c6c7b7ee82cb98f97ab5bf1459","test1"));
System.out.println(ocrLocalImgFile(testResourcesLanguagePath,"C:\\Users\\iformall\\Desktop\\01.jpg","test1"));
} catch (Exception e) {
// TODO Auto-generated catch block
e.printStackTrace();


+ 4
- 4
mallinkOcr/src/main/java/com/iformall/ocr/FormallTess4jTrain.java 查看文件

@@ -45,7 +45,7 @@ public class FormallTess4jTrain {
private static String getBoxFont(String lang,String fontName) {
String prefix = getprefix(lang,fontName);
String extractorCmdstr = "unicharset_extractor "+prefix+".box";
String extractorCmdstr = "unicharset_extractor "+prefix+".box [other .box files...]";
return extractorCmdstr;
}
@@ -101,11 +101,11 @@ public class FormallTess4jTrain {
sb.append("第四步:用box文件生成。tr文件 ").append(huanhangstr);
sb.append(createTranFile(lang,fontName)).append(huanhangstr);
sb.append("............ ").append(huanhangstr);
sb.append("@@@!!!!从此处开始,可以处理多个字库处理。如原来的字库现在做升级,需要把老的字库的。tif,.box,.tr文件重新命名也放到同目录文件夹, ").append(huanhangstr);
sb.append("............ ").append(huanhangstr);
sb.append("第五步:提取box字符 ").append(huanhangstr);
sb.append(getBoxFont(lang,fontName)).append(huanhangstr);
sb.append("............ ").append(huanhangstr);
sb.append("@@@!!!!从此处开始,可以处理多个字库处理。如原来的字库现在做升级,需要把老的字库的。tif,.box,.tr文件重新命名也放到同目录文件夹, ").append(huanhangstr);
sb.append("............ ").append(huanhangstr);
sb.append("第六步:新建font_properties(没有.txt后缀), 把所有box对应的字体特征加进去。").append(huanhangstr);
sb.append("fontname为字体名称,保持和 图片集文件 .tif 和.box文件的前缀名一致 .<italic> 、<bold> 、<fixed> 、<serif>、 <fraktur>的取值为1或0,表示字体是否具有这些属性。 ").append(huanhangstr);
sb.append("如以下内容: ").append(huanhangstr);
@@ -115,7 +115,7 @@ public class FormallTess4jTrain {
sb.append("............ ").append(huanhangstr);
sb.append("第六步:批量执行系列命令。多个文件在后面累加。 ").append(huanhangstr);
String prefix = getprefix(lang,fontName);
String shapeCmdStr = "shapeclustering -F font_properties -U unicharset "+prefix+".tr [other .tr files...] ";
String shapeCmdStr = "### shapeclustering -F font_properties -U unicharset "+prefix+".tr [other .tr files...] ";
sb.append(shapeCmdStr).append(huanhangstr);
String mftraingCmdstr = "mftraining -F font_properties -U unicharset -O unicharset "+prefix+".tr [other .tr files...] ";
sb.append(mftraingCmdstr).append(huanhangstr);


二進制
mallinkOcr/src/main/resources/tessdata/test1.traineddata 查看文件


Loading…
取消
儲存