v1.3.1

rkcosmos · rkcosmos · commit bc41819c1053 · 2021-04-20T17:28:21.000+07:00
diff --git a/README.md b/README.md
@@ -11,6 +11,10 @@ Ready-to-use OCR with 80+ [supported languages](https://www.jaided.ai/easyocr) a
 [Try Demo on our website](https://www.jaided.ai/easyocr)
 
 ## What's new
+- 20 April 2021 - Version 1.3.1
+    - Add support for PIL image (thanks [@prays](https://github.com/prays))
+    - Update argument setting for command line
+    - Add `x_ths` and `y_ths` to control merging behavior when `paragraph=True`
 - 21 March 2021 - Version 1.3
     - Second-generation models: multiple times smaller size, multiple times faster inference, additional characters, comparable accuracy to the first generation models.
     EasyOCR will choose the latest model by default but you can also specify which model to use by passing `recog_network` argument when creating `Reader` instance.
diff --git a/easyocr/__init__.py b/easyocr/__init__.py
@@ -1,3 +1,3 @@
 from .easyocr import Reader
 
-__version__ = '1.3.0.1'
+__version__ = '1.3.1'
diff --git a/easyocr/cli.py b/easyocr/cli.py
@@ -12,13 +12,110 @@ def parse_args():
         type=str,
         help="for languages",
     )
+    parser.add_argument(
+        "--gpu",
+        type=bool,
+        choices=[True, False],
+        default=True,
+        help="Using GPU (default: True)",
+    )
+    parser.add_argument(
+        "--model_storage_directory",
+        type=str,
+        default=None,
+        help="Directory for model (.pth) file",
+    )
+    parser.add_argument(
+        "--user_network_directory",
+        type=str,
+        default=None,
+        help="Directory for custom network files",
+    )
+    parser.add_argument(
+        "--recog_network",
+        type=str,
+        default='standard',
+        help="Recognition networks",
+    )
+    parser.add_argument(
+        "--download_enabled",
+        type=bool,
+        choices=[True, False],
+        default=True,
+        help="Enable Download",
+    )
+    parser.add_argument(
+        "--detector",
+        type=bool,
+        choices=[True, False],
+        default=True,
+        help="Initialize text detector module",
+    )
+    parser.add_argument(
+        "--recognizer",
+        type=bool,
+        choices=[True, False],
+        default=True,
+        help="Initialize text recognizer module",
+    )
+    parser.add_argument(
+        "--verbose",
+        type=bool,
+        choices=[True, False],
+        default=True,
+        help="Print detail/warning",
+    )
+    parser.add_argument(
+        "--quantize",
+        type=bool,
+        choices=[True, False],
+        default=True,
+        help="Use dynamic quantization",
+    )
     parser.add_argument(
         "-f",
         "--file",
         required=True,
         type=str,
         help="input file",
     )
+    parser.add_argument(
+        "--decoder",
+        type=str,
+        choices=["greedy", 'beamsearch', 'wordbeamsearch'],
+        default='greedy',
+        help="decoder algorithm",
+    )
+    parser.add_argument(
+        "--beamWidth",
+        type=int,
+        default=5,
+        help="size of beam search",
+    )
+    parser.add_argument(
+        "--batch_size",
+        type=int,
+        default=1,
+        help="batch_size",
+    )
+    parser.add_argument(
+        "--workers",
+        type=int,
+        default=0,
+        help="number of processing cpu cores",
+    )
+    parser.add_argument(
+        "--allowlist",
+        type=str,
+        default=None,
+        help="Force EasyOCR to recognize only subset of characters",
+    )
+    parser.add_argument(
+        "--blocklist",
+        type=str,
+        default=None,
+        help="Block subset of character. This argument will be ignored if allowlist is given.",
+    )
     parser.add_argument(
         "--detail",
         type=int,
@@ -27,20 +124,150 @@ def parse_args():
         help="simple output (default: 1)",
     )
     parser.add_argument(
-        "--gpu",
+        "--rotation_info",
+        type=list,
+        default=None,
+        help="Allow EasyOCR to rotate each text box and return the one with the best confident score. Eligible values are 90, 180 and 270. For example, try [90, 180 ,270] for all possible text orientations.",
+    )
+    parser.add_argument(
+        "--paragraph",
         type=bool,
         choices=[True, False],
-        default=True,
-        help="Using GPU (default: True)",
+        default=False,
+        help="Combine result into paragraph",
+    )
+    parser.add_argument(
+        "--min_size",
+        type=int,
+        default=20,
+        help="Filter text box smaller than minimum value in pixel",
+    )
+    parser.add_argument(
+        "--contrast_ths",
+        type=float,
+        default=0.1,
+        help="Text box with contrast lower than this value will be passed into model 2 times. First is with original image and second with contrast adjusted to 'adjust_contrast' value. The one with more confident level will be returned as a result.",
+    )
+    parser.add_argument(
+        "--adjust_contrast",
+        type=float,
+        default=0.5,
+        help="target contrast level for low contrast text box",
+    )
+    parser.add_argument(
+        "--text_threshold",
+        type=float,
+        default=0.7,
+        help="Text confidence threshold",
     )
+    parser.add_argument(
+        "--low_text",
+        type=float,
+        default=0.4,
+        help="Text low-bound score",
+    )
+    parser.add_argument(
+        "--link_threshold",
+        type=float,
+        default=0.4,
+        help="Link confidence threshold",
+    )
+    parser.add_argument(
+        "--canvas_size",
+        type=int,
+        default=2560,
+        help="Maximum image size. Image bigger than this value will be resized down.",
+    )
+    parser.add_argument(
+        "--mag_ratio",
+        type=float,
+        default=1.,
+        help="Image magnification ratio",
+    )
+    parser.add_argument(
+        "--slope_ths",
+        type=float,
+        default=0.1,
+        help="Maximum slope (delta y/delta x) to considered merging. Low value means tiled boxes will not be merged.",
+    )
+    parser.add_argument(
+        "--ycenter_ths",
+        type=float,
+        default=0.5,
+        help="Maximum shift in y direction. Boxes with different level should not be merged.",
+    )
+    parser.add_argument(
+        "--height_ths",
+        type=float,
+        default=0.5,
+        help="Maximum different in box height. Boxes with very different text size should not be merged. ",
+    )
+    parser.add_argument(
+        "--width_ths",
+        type=float,
+        default=0.5,
+        help="Maximum horizontal distance to merge boxes.",
+    )
+    parser.add_argument(
+        "--y_ths",
+        type=float,
+        default=0.5,
+        help="Maximum horizontal distance to merge boxes (when paragraph = True).",
+    )
+    parser.add_argument(
+        "--x_ths",
+        type=float,
+        default=1.0,
+        help="Maximum horizontal distance to merge boxes (when paragraph = True).",
+    )
+    parser.add_argument(
+        "--add_margin",
+        type=float,
+        default=0.1,
+        help="Extend bounding boxes in all direction by certain value. This is important for language with complex script (E.g. Thai).",
+    )
+
     args = parser.parse_args()
     return args
 
 
 def main():
     args = parse_args()
-    reader = easyocr.Reader(lang_list=args.lang, gpu=args.gpu)
-    for line in reader.readtext(args.file, detail=args.detail):
+    reader = easyocr.Reader(lang_list=args.lang,\
+                            gpu=args.gpu,\
+                            model_storage_directory=args.model_storage_directory,\
+                            user_network_directory=args.user_network_directory,\
+                            recog_network=args.recog_network,\
+                            download_enabled=args.download_enabled,\
+                            detector=args.detector,\
+                            recognizer=args.recognizer,\
+                            verbose=args.verbose,\
+                            quantize=args.quantize)
+    for line in reader.readtext(args.file,\
+                                decoder=args.decoder,\
+                                beamWidth=args.beamWidth,\
+                                batch_size=args.batch_size,\
+                                workers=args.workers,\
+                                allowlist=args.allowlist,\
+                                blocklist=args.blocklist,\
+                                detail=args.detail,\
+                                rotation_info=args.rotation_info,\
+                                paragraph=args.paragraph,\
+                                min_size=args.min_size,\
+                                contrast_ths=args.contrast_ths,\
+                                adjust_contrast=args.adjust_contrast,\
+                                text_threshold=args.text_threshold,\
+                                low_text=args.low_text,\
+                                link_threshold=args.link_threshold,\
+                                canvas_size=args.canvas_size,\
+                                mag_ratio=args.mag_ratio,\
+                                slope_ths=args.slope_ths,\
+                                ycenter_ths=args.ycenter_ths,\
+                                height_ths=args.height_ths,\
+                                width_ths=args.width_ths,\
+                                y_ths=args.y_ths,\
+                                x_ths=args.x_ths,\
+                                add_margin=args.add_margin):
         print(line)
 
 
diff --git a/easyocr/easyocr.py b/easyocr/easyocr.py
@@ -285,10 +285,9 @@ def detect(self, img, min_size = 20, text_threshold = 0.7, low_text = 0.4,\
     def recognize(self, img_cv_grey, horizontal_list=None, free_list=None,\
                   decoder = 'greedy', beamWidth= 5, batch_size = 1,\
                   workers = 0, allowlist = None, blocklist = None, detail = 1,\
-                  rotation_info = None,\
-                  paragraph = False,\
+                  rotation_info = None,paragraph = False,\
                   contrast_ths = 0.1,adjust_contrast = 0.5, filter_ths = 0.003,\
-                  reformat=True):
+                  y_ths = 0.5, x_ths = 1.0, reformat=True):
 
         if reformat:
             img, img_cv_grey = reformat_input(img_cv_grey)
@@ -350,7 +349,7 @@ def recognize(self, img_cv_grey, horizontal_list=None, free_list=None,\
             direction_mode = 'ltr'
 
         if paragraph:
-            result = get_paragraph(result, mode = direction_mode)
+            result = get_paragraph(result, x_ths=x_ths, y_ths=y_ths, mode = direction_mode)
 
         if detail == 0:
             return [item[1] for item in result]
@@ -364,7 +363,7 @@ def readtext(self, image, decoder = 'greedy', beamWidth= 5, batch_size = 1,\
                  text_threshold = 0.7, low_text = 0.4, link_threshold = 0.4,\
                  canvas_size = 2560, mag_ratio = 1.,\
                  slope_ths = 0.1, ycenter_ths = 0.5, height_ths = 0.5,\
-                 width_ths = 0.5, add_margin = 0.1):
+                 width_ths = 0.5, y_ths = 0.5, x_ths = 1.0, add_margin = 0.1):
         '''
         Parameters:
         image: file path or numpy-array or a byte stream object
@@ -382,6 +381,6 @@ def readtext(self, image, decoder = 'greedy', beamWidth= 5, batch_size = 1,\
                                 decoder, beamWidth, batch_size,\
                                 workers, allowlist, blocklist, detail, rotation_info,\
                                 paragraph, contrast_ths, adjust_contrast,\
-                                filter_ths, False)
+                                filter_ths, y_ths, x_ths, False)
 
         return result
diff --git a/releasenotes.md b/releasenotes.md
@@ -1,3 +1,7 @@
+- 20 April 2021 - Version 1.3.1
+    - Add support for PIL image (thanks [@prays](https://github.com/prays))
+    - Update argument setting for command line
+    - Add `x_ths` and `y_ths` to control merging behavior when `paragraph=True`
 - 21 March 2021 - Version 1.3
     - Second-generation models: multiple times smaller size, multiple times faster inference, additional characters, comparable accuracy to the first generation models.
     EasyOCR will choose the latest model by default but you can also specify which model to use by passing `recog_network` argument when creating `Reader` instance.
diff --git a/setup.py b/setup.py
@@ -17,7 +17,7 @@ def readme():
     name='easyocr',
     packages=['easyocr'],
     include_package_data=True,
-    version='1.3.0.1',
+    version='1.3.1',
     install_requires=requirements,
     entry_points={"console_scripts": ["easyocr= easyocr.cli:main"]},
     license='Apache License 2.0',

Original file line number	Diff line number	Diff line change
`@@ -1,3 +1,3 @@`
`1`	`1`	`from .easyocr import Reader`
`2`	`2`
`3`		`-__version__ = '1.3.0.1'`
	`3`	`+__version__ = '1.3.1'`