Базовое определение текста через easyocr

main.py

@@ -1,17 +1,15 @@
 import time
 import math
 import logging
+
 import numpy as np
-import pandas as pd
 import pygetwindow as gw
 import pyautogui
-from pytesseract import pytesseract
-from pytesseract import Output
 import cv2
+import easyocr
 
 
 win_name = "EVE - Nostrom Stone"
-pytesseract.tesseract_cmd = "C:\\Program Files\\Tesseract-OCR\\tesseract.exe"
 
 logging.basicConfig(
     level=logging.INFO,
@@ -22,8 +20,6 @@ logging.basicConfig(
 
 save_result = True
 
-pd.options.mode.use_inf_as_na = True
-
 
 def save_highlighted_screenshot(screenshot, boxes, filename):
     new_image = screenshot.copy()
@@ -90,108 +86,33 @@ def get_screenshot():
     return cv2.cvtColor(np.array(screenshot), cv2.COLOR_BGR2RGB)
 
 
-def get_boxes(screenshot):
-    inverted_screenshot = cv2.bitwise_not(
-        cv2.cvtColor(screenshot, cv2.COLOR_BGR2GRAY)
-    )
-
-    raw_boxes = pytesseract.image_to_data(
-        inverted_screenshot,
-        lang="eng",
-        output_type=Output.DATAFRAME,
-        config="--psm 3",
-    )
-
-    if save_result:
-        raw_boxes.to_excel("xlsx/0_raw_boxes.xlsx", index=False)
-
-    base_boxes = raw_boxes.loc[
-        (raw_boxes["conf"] > 30)
-        & (raw_boxes["text"].notnull())
-        & (raw_boxes["text"].str.isalnum())
-    ]
-
-    if save_result:
-        save_highlighted_screenshot(
-            screenshot, base_boxes, "1_base_highlighted_screenshot"
-        )
-        base_boxes.to_excel("xlsx/1_base_boxes.xlsx", index=False)
-
-    logging.info("Боксы получены")
-
-    return base_boxes
-
-
-def union_boxes(base_boxes):
-    result_phrases = pd.DataFrame(
-        columns=[
-            "left",
-            "top",
-            "width",
-            "height",
-            "text",
-            "block_num",
-            "line_num",
-            "par_num",
-        ]
-    )
-    for box in base_boxes["block_num"].unique():
-        paragraphs_in_box = base_boxes.loc[base_boxes["block_num"] == box][
-            "par_num"
-        ].unique()
-        for paragraph in paragraphs_in_box:
-            words_in_paragraph = base_boxes.loc[
-                (base_boxes["block_num"] == box)
-                & (base_boxes["par_num"] == paragraph),
-            ]
-
-            grouped_words = words_in_paragraph.groupby(
-                "line_num", as_index=False
-            )
-
-            box_phrases = grouped_words["width"].sum()
-            box_phrases = box_phrases.merge(
-                grouped_words["height"].max(), on="line_num", how="left"
-            )
-            box_phrases = box_phrases.merge(
-                grouped_words["left"].min(), on="line_num", how="left"
-            )
-            box_phrases = box_phrases.merge(
-                grouped_words["top"].min(), on="line_num", how="left"
-            )
-            box_phrases = box_phrases.merge(
-                grouped_words["text"].apply(" ".join),
-                on="line_num",
-                how="left",
-            )
-            box_phrases["block_num"] = box
-
-        rightest_box = words_in_paragraph.loc[
-            words_in_paragraph["left"] == words_in_paragraph["left"].max()
-        ]
-        leftest_box = words_in_paragraph.loc[
-            words_in_paragraph["left"] == words_in_paragraph["left"].min()
-        ]
-        box_phrases["width"] = (
-            rightest_box.iloc[0].left
-            + rightest_box.iloc[0].width
-            - leftest_box.iloc[0].left
-        )
-
-        result_phrases = pd.concat([result_phrases, box_phrases])
-
-    if save_result:
-        result_phrases.to_excel("xlsx/2_unioned_boxes.xlsx", index=False)
-
-    logging.info("Боксы объединены")
-
-    return result_phrases
-
-
 screenshot = get_screenshot()
-base_boxes = get_boxes(screenshot)
-unioned_boxes = union_boxes(base_boxes)
 
-save_highlighted_screenshot(
-    screenshot, unioned_boxes, "2_unioned_highlighted_screenshot"
-)
+reader = easyocr.Reader(["en"], gpu=True)
+result = reader.readtext(screenshot)
+
+
+for bbox, text, prob in result:
+    (tl, tr, br, bl) = bbox
+    tl = (int(tl[0]), int(tl[1]))
+    tr = (int(tr[0]), int(tr[1]))
+    br = (int(br[0]), int(br[1]))
+    bl = (int(bl[0]), int(bl[1]))
+
+    text = "".join([c if ord(c) < 128 else "" for c in text]).strip()
+    cv2.rectangle(screenshot, tl, br, (0, 255, 0), 2)
+    cv2.putText(
+        screenshot,
+        text,
+        (tl[0], tl[1] - 10),
+        cv2.FONT_HERSHEY_SIMPLEX,
+        0.8,
+        (0, 255, 0),
+        2,
+    )
+
+
+print(result)
+
+cv2.imshow("screenshot", screenshot)
+cv2.waitKey(0)

requirements.txt

@@ -1,19 +1,34 @@
-et-xmlfile==1.1.0
+certifi==2022.12.7
+charset-normalizer==3.1.0
+easyocr==1.6.2
+idna==3.4
+imageio==2.26.0
+lazy_loader==0.1
 MouseInfo==0.1.3
+networkx==3.0
+ninja==1.11.1
 numpy==1.24.2
 opencv-python==4.7.0.72
-openpyxl==3.1.1
 packaging==23.0
-pandas==1.5.3
 Pillow==9.4.0
 PyAutoGUI==0.9.53
+pyclipper==1.3.0.post4
 PyGetWindow==0.0.9
 PyMsgBox==1.0.9
 pyperclip==1.8.2
 PyRect==0.2.0
 PyScreeze==0.1.28
-pytesseract==0.3.10
-python-dateutil==2.8.2
+python-bidi==0.4.2
 pytweening==1.0.4
-pytz==2022.7.1
+PyWavelets==1.4.1
+PyYAML==6.0
+requests==2.28.2
+scikit-image==0.20.0
+scipy==1.10.1
+shapely==2.0.1
 six==1.16.0
+tifffile==2023.2.28
+torch==1.13.1+cu116
+torchvision==0.14.1+cu116
+typing_extensions==4.5.0
+urllib3==1.26.14