新增img2ofd img2pfd

README.md

@@ -15,6 +15,8 @@
 
 5 添加gui 工具实现上述功能
 
+6 jpg2ofd jpg2pfd
+
 关于 jb2格式图片解析 
 使用了第三方库 jbig2dec 去读取jb2格式图片 参考下面链接安装使用jbig2dec 
 https://github.com/rillian/jbig2dec 

easyofd/draw/draw_ofd.py

@@ -9,6 +9,7 @@
 from io import BytesIO
 
 import xmltodict
+import cv2
 from PIL import Image
 from .pdf_parse import DPFParser
 from .ofdtemplate import OFDTemplate,DocumentTemplate,DocumentResTemplate,PulicResTemplate,ContentTemplate,OFDStructure
@@ -52,6 +53,7 @@ def build_content_res(self,pil_img_list):
         """一张图片是一页"""
         content_res_list = []
         for idx, pil_img in enumerate(pil_img_list):
+            print(pil_img)
             print(idx,pil_img[1],pil_img[2])
             PhysicalBox = f"0 0 {pil_img[1]} {pil_img[2]}"
             ImageObject = [{
@@ -93,11 +95,14 @@ def pil_2_bytes(self, image):
         img_bytesio.close()
         return img_bytes
 
-    def __call__(self,pdf_bytes):
+    def __call__(self,pdf_bytes,CV2_img_list=None):
         # 读取 pdf 转图片 
-        pdf_obj = DPFParser()
-        img_list = pdf_obj.to_img(pdf_bytes)
-        pil_img_list = [(self.pil_2_bytes(Image.frombytes("RGB", [_img.width, _img.height], _img.samples)),_img.width,_img.height) for _img in img_list]
+        if CV2_img_list:
+            pil_img_list = [(self.pil_2_bytes(Image.fromarray(cv2.cvtColor(_img,cv2.COLOR_BGR2RGB))),_img.shape[1],_img.shape[0] ) for _img in CV2_img_list]
+        else:
+            pdf_obj = DPFParser()
+            img_list = pdf_obj.to_img(pdf_bytes)
+            pil_img_list = [(self.pil_2_bytes(Image.frombytes("RGB", [_img.width, _img.height], _img.samples)),_img.width,_img.height) for _img in img_list]
         document = DocumentTemplate()
         pulic_res = self.build_pulic_res()
         document_res = self.build_document_res(len(pil_img_list))

easyofd/ofd.py

@@ -14,6 +14,7 @@
 from typing import Any
 
 import fitz
+import cv2
 import numpy as np
 from loguru import logger
 
@@ -84,6 +85,24 @@ def pdf2img(self,pdfbytes):
         logger.info(f"to_jpg")
         return image_list
 
+    def jpg2ofd(self,imglist:list):
+        """
+        imglist: cv2 image list
+        """
+
+        ofd_byte = OFDWrite()(None,CV2_img_list=imglist)
+        return ofd_byte
+
+    def jpg2pfd(self,imglist:list ):
+        """
+        imglist: cv2 image list
+        1 构建data 
+        2 DrawPDF(self.data)()
+        """
+
+        data = OFDParser(None).img2data(imglist)
+        return DrawPDF(data)()
+
     def to_jpg(self,format="jpg"):
         """
         return numpy list

easyofd/parser_ofd/ofd_parser.py

@@ -14,6 +14,7 @@
 import traceback
 import base64
 import re
+import cv2
 from typing import Any
 
 from loguru import logger
@@ -34,6 +35,60 @@ def __init__(self, ofdb64):
         self.file_tree = None
         self.jbig2dec_path = r"C:/msys64/mingw64/bin/jbig2dec.exe"
 
+    def img2data(self,imglist):
+        """
+        imglist to ofd data
+        
+        """
+        OP = 200/25.4
+        doc_list = []
+        img_info = {}
+        page_size = []
+        font_info = {}
+        page_info_d = {}
+
+
+
+        for idx, img_numpy in enumerate(imglist):
+            h,w,_=img_numpy.shape
+            _, img_encode = cv2.imencode('.jpg', img_numpy)
+            img_bytes = img_encode.tobytes()
+            imgb64 = str(base64.b64encode(img_bytes),encoding="utf-8")
+            img_info[str(idx)] = {
+                "format":  "jpg",
+                "wrap_pos":  "",
+                "type":  "IMG",
+                "suffix":  "jpg",
+                "fileName":  f"{idx}.jpg",
+                "imgb64":  imgb64,
+
+            }
+            text_list = []
+            img_list = []
+            img_d = {}
+            img_d["CTM"] = "" # 平移矩阵换
+            img_d["ID"] = str(idx) # 图片id
+            img_d["ResourceID"] = str(idx) # 图片id
+            img_d["pos"] = [0,0,w/OP,h/OP] # 平移矩阵换
+            page_size = [0,0,w/OP,h/OP]
+            # print(page_size)
+            img_list.append(img_d)
+
+            content_d = {
+        "text_list":text_list,
+        "img_list":img_list,
+                     }
+            page_info_d[idx]=content_d
+        doc_list.append({
+            "pdf_name": "demo.pdf",
+            "doc_no":"0",
+            "images":img_info,
+            "page_size":page_size,
+            "fonts":font_info,
+            "page_info":page_info_d        
+                        })
+
+        return doc_list
 
     # 获得xml 对象
     def get_xml_obj(self, label):