use fitz to export images of pdf

main.py

@@ -142,14 +142,18 @@ def ocr_one_pdf(pth_pdf):
     from pathlib import Path
     import os
     import json
+    
+    import fitz
+    from PIL import Image  # 需要安装Pillow库
 
     base = Path(pth_pdf).stem
     dir = os.path.dirname(pth_pdf)
 
 
-    def get_page_image(reader, page_num):
+    def get_page_image(reader, pdfDoc, page_num):
         page_num = int(page_num)
         page = reader.pages[page_num - 1]
+            
         text = page.extract_text()
         # try:
         #     for idx, image_file_object in enumerate(page.images):
@@ -157,15 +161,38 @@ def ocr_one_pdf(pth_pdf):
         # except Exception:
         #     return None
             # 提取不到图像先用 Acrobat 导出全部图片，再新建 pdf 替换原 pdf 就可以了
+            
+        page = pdfDoc[page_num - 1]
+                # 关键参数设置（保持原始尺寸和分辨率）
+        matrix = fitz.Matrix(fitz.Identity)  # 使用Identity矩阵确保不缩放[6,8](@ref)
+        pix = page.get_pixmap(
+            matrix=matrix,      # 保持原始尺寸
+            dpi=300,            # 设置输出分辨率（默认72dpi）
+            alpha=False,        # 关闭透明通道（提高兼容性）
+            colorspace="rgb"    # 使用RGB色彩空间[8](@ref)
+        )
+    
+        from io import BytesIO
+        byte_io = BytesIO()
+        # 转换到内存流
+        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
+        byte_io = BytesIO()
+        img.save(byte_io, format="JPEG", quality=100) # 使用无损压缩
+        byte_io.seek(0)
+        img_bytes = byte_io.read()
         
-        for idx, image_file_object in enumerate(page.images):
-            img_bytes = image_file_object.data
-            img_buffer_numpy = np.frombuffer(img_bytes, dtype=np.uint8)  # 将图片字节码 bytes 转换成一维的 numpy 数组到缓存中
-            ocr_frame = cv2.imdecode(img_buffer_numpy, 1)   # 从指定的内存缓存中读取一维 numpy 数据，并把数据转换(解码)成图像矩阵格式
-            ocr_frame = cv2.cvtColor(ocr_frame, cv2.COLOR_BGR2RGB)
-            # cv2.imshow('test', self.ocr_frame)
-            # cv2.waitKey(0)
-            break
+        img_buffer_numpy = np.frombuffer(img_bytes, dtype=np.uint8)  # 将图片字节码 bytes 转换成一维的 numpy 数组到缓存中
+        ocr_frame = cv2.imdecode(img_buffer_numpy, 1)   # 从指定的内存缓存中读取一维 numpy 数据，并把数据转换(解码)成图像矩阵格式
+        ocr_frame = cv2.cvtColor(ocr_frame, cv2.COLOR_BGR2RGB)
+        
+        # for idx, image_file_object in enumerate(page.images):
+        #     img_bytes = image_file_object.data
+        #     img_buffer_numpy = np.frombuffer(img_bytes, dtype=np.uint8)  # 将图片字节码 bytes 转换成一维的 numpy 数组到缓存中
+        #     ocr_frame = cv2.imdecode(img_buffer_numpy, 1)   # 从指定的内存缓存中读取一维 numpy 数据，并把数据转换(解码)成图像矩阵格式
+        #     ocr_frame = cv2.cvtColor(ocr_frame, cv2.COLOR_BGR2RGB)
+        #     # cv2.imshow('test', self.ocr_frame)
+        #     # cv2.waitKey(0)
+        #     break
         return ocr_frame
 
     def md5_file(fname):
@@ -191,7 +218,11 @@ def ocr_one_pdf(pth_pdf):
 
     from pypdf import PdfReader
     reader = PdfReader(pth_pdf)
+    pdfDoc = fitz.open(pth_pdf)
     number_of_pages = len(reader.pages)
+    number_of_pages2 = pdfDoc.page_count
+    
+    assert number_of_pages == number_of_pages2
 
     # see huggingface/PPOCRLabel/PPOCRLabel.py   for image notation
     name_pp_label = 'Label.txt'
@@ -208,10 +239,10 @@ def ocr_one_pdf(pth_pdf):
     
     for nth_page in range(1, number_of_pages+1):
 
-        if nth_page > 500:
+        if nth_page > 10:
             break
 
-        img_color = get_page_image(reader, nth_page)
+        img_color = get_page_image(reader, pdfDoc, nth_page)
         if img_color is None:
             continue
         cv2.imwrite('./tmp.jpg', cv2.cvtColor(img_color, cv2.COLOR_RGB2BGR))
@@ -339,6 +370,6 @@ if __name__ == '__main__':
     # ocr_one_pdf('pdfs/en/TIME片挑200/TIME片挑200.pdf')
     
     
-    ocr_one_pdf('pdfs/en/周美佳生活大爆炸每天说一点地道口语第6季/周美佳生活大爆炸每天说一点地道口语第6季.pdf')  # 李致雨N2词汇详解 提取的图片几乎不可见，不要了
+    ocr_one_pdf('pdfs/jp/荻原雲来梵语漢訳対照梵和大辞典/荻原雲来梵语漢訳対照梵和大辞典.pdf')  # 李致雨N2词汇详解 提取的图片几乎不可见，不要了
     
     

requirements.txt

@@ -70,3 +70,9 @@ Werkzeug==3.0.4
 ######### PaddleOCR_ali1k_det_rec_300epoch  end   ########################
 
 
+######### export images of pdf begin  ########################
+fitz==0.0.1.dev2
+pymupdf==1.25.4
+pillow==10.4.0
+    # 先装 fitz 再装 pymupdf 才行。如果出错，先卸载 pymupdf ，再装一次 pymupdf 又不错了
+######### export images of pdf end    ########################