optimize: on aggressive settings try JPG to PNG transcoding

If the color count of an image is low such as when black and white
documents are scanned in color, PNG with lossy quantization may
produce a superior encoding to JPEG. This is expensive to test however.

src/ocrmypdf/exec/pngquant.py

@@ -17,6 +17,9 @@
 
 from functools import lru_cache
 from subprocess import run
+from tempfile import NamedTemporaryFile
+
+from PIL import Image
 
 from . import get_version
 from ..exceptions import MissingDependencyError
@@ -36,16 +39,32 @@ def available():
 
 
 def quantize(input_file, output_file, quality_min, quality_max):
-    args = [
-        'pngquant',
-        '--force',
-        '--skip-if-larger',
-        '--output',
-        output_file,
-        '--quality',
-        f'{quality_min}-{quality_max}',
-        '--',
-        input_file,
-    ]
-    proc = run(args)
-    proc.check_returncode()
+    if input_file.endswith('.jpg'):
+        im = Image.open(input_file)
+        with NamedTemporaryFile(suffix='.png') as tmp:
+            im.save(tmp)
+            args = [
+                'pngquant',
+                '--force',
+                '--skip-if-larger',
+                '--output',
+                output_file,
+                '--quality',
+                f'{quality_min}-{quality_max}',
+                '--',
+                tmp.name,
+            ]
+            run(args)
+    else:
+        args = [
+            'pngquant',
+            '--force',
+            '--skip-if-larger',
+            '--output',
+            output_file,
+            '--quality',
+            f'{quality_min}-{quality_max}',
+            '--',
+            input_file,
+        ]
+        run(args)

src/ocrmypdf/optimize.py

@@ -420,6 +420,9 @@ def optimize(input_file, output_file, log, context):
 
     jpegs, pngs = extract_images_generic(pike, root, log, options)
     transcode_jpegs(pike, jpegs, root, log, options)
+    if options.optimize >= 2:
+        # Try pngifying the jpegs
+        transcode_pngs(pike, jpegs, jpg_name, root, log, options)
     transcode_pngs(pike, pngs, png_name, root, log, options)
 
     jbig2_groups = extract_images_jbig2(pike, root, log, options)