Enhanced PDF to Images: ultra-high DPI up to 600, sharper text rendering

server.py

@@ -1309,11 +1309,19 @@ async def api_split_pdf(
 async def api_pdf_to_images(
     file: UploadFile = File(...),
     format: str = Form("png"),
-    dpi: int = Form(150),
+    dpi: int = Form(200),
     pages: str = Form(""),
     output_name: str = Form("pages"),
 ):
-    """Convert PDF pages to images (PNG or JPG)"""
+    """Convert PDF pages to high-quality images (PNG or JPG)
+    
+    DPI Guide:
+    - 150: Fast, small files (web preview)
+    - 200: Good quality (default)
+    - 300: Print quality
+    - 400: Ultra sharp (presentations)
+    - 600: Maximum quality (OCR/archive)
+    """
     data = await file.read()
     validate_pdf_upload(data)
     
@@ -1321,7 +1329,9 @@ async def api_pdf_to_images(
     format = format.lower() if format.lower() in ["png", "jpg", "jpeg"] else "png"
     if format == "jpeg":
         format = "jpg"
-    dpi = max(72, min(300, dpi))
+    
+    # Allow higher DPI for ultra-sharp output (up to 600)
+    dpi = max(72, min(600, dpi))
     
     try:
         src_doc = fitz.open(stream=data, filetype="pdf")
@@ -1339,13 +1349,15 @@ async def api_pdf_to_images(
         job_id = generate_job_id()
         zip_path = OUT_DIR / f"{job_id}_{output_name}.zip"
         
+        # Vector render at specified DPI - handles portrait/landscape automatically
         zoom = dpi / 72.0
         matrix = fitz.Matrix(zoom, zoom)
         
         with zipfile.ZipFile(zip_path, 'w', zipfile.ZIP_DEFLATED) as zf:
             for i in sorted(page_set):
                 page = src_doc.load_page(i)
-                pix = page.get_pixmap(matrix=matrix)
+                # alpha=False removes transparency for sharper text
+                pix = page.get_pixmap(matrix=matrix, alpha=False)
                 
                 if format == "png":
                     img_bytes = pix.tobytes("png")
@@ -1353,13 +1365,13 @@ async def api_pdf_to_images(
                     # Convert to JPG via PIL
                     img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
                     buffer = io.BytesIO()
-                    img.save(buffer, format="JPEG", quality=90)
+                    img.save(buffer, format="JPEG", quality=95, optimize=True)
                     img_bytes = buffer.getvalue()
                 
                 zf.writestr(f"{output_name}_page_{i+1}.{format}", img_bytes)
         
         src_doc.close()
-        logger.info(f"Converted {len(page_set)} pages to {format.upper()}")
+        logger.info(f"Converted {len(page_set)} pages to {format.upper()} at {dpi} DPI")
         
         return FileResponse(
             path=str(zip_path),

static/index.html

@@ -980,12 +980,16 @@
               </select>
             </div>
             <div class="form-group">
-              <label class="form-label">Quality (DPI)</label>
+              <label class="form-label">
+                Quality (DPI)
+                <span class="help-tip tooltip" data-tooltip="Higher DPI = sharper images but larger files">?</span>
+              </label>
               <select id="pdf2img_dpi" class="form-select">
-                <option value="72">72 DPI (Fast)</option>
-                <option value="150" selected>150 DPI (Balanced)</option>
-                <option value="200">200 DPI (Quality)</option>
-                <option value="300">300 DPI (Best)</option>
+                <option value="150">150 DPI (Web preview)</option>
+                <option value="200" selected>200 DPI (Good quality)</option>
+                <option value="300">300 DPI (Print quality)</option>
+                <option value="400">400 DPI (Ultra sharp)</option>
+                <option value="600">600 DPI (Maximum - OCR/Archive)</option>
               </select>
             </div>
             <div class="form-group">