Upload 2 files

app.py

@@ -0,0 +1,64 @@
+from typing import List
+import pytesseract
+from PIL import Image
+import gradio as gr
+import cv2
+import numpy as np
+
+def tesseract_ocr_with_selection(filepath: str, languages: List[str], coordinates: List[int] = None):
+    # Görseli yükle
+    image = Image.open(filepath)
+    
+    if coordinates:
+        # Koordinatlara göre kırp
+        x1, y1, x2, y2 = coordinates
+        image = image.crop((x1, y1, x2, y2))
+    
+    # OCR işlemi
+    return pytesseract.image_to_string(image=image, lang=', '.join(languages))
+
+def draw_selection_box(image):
+    # Görseli numpy formatına çevir
+    image = np.array(image)
+    # Kullanıcıdan alan seçmesini iste
+    coordinates = cv2.selectROI("Alanı Seçin (ESC ile çıkın)", image, showCrosshair=True)
+    cv2.destroyAllWindows()
+    return list(coordinates)
+
+# Gradio UI ayarları
+title = "Tesseract OCR with Selection"
+description = "Gradio demo for Tesseract OCR with region selection."
+article = "<p style='text-align: center'><a href='https://tesseract-ocr.github.io/' target='_blank'>Tesseract documentation</a> | <a href='https://github.com/tesseract-ocr/tesseract' target='_blank'>Github Repo</a></p>"
+
+examples = [
+    ['examples/eurotext.png', ['eng'], [50, 50, 200, 200]],
+    ['examples/tesseract_sample.png', ['jpn', 'eng'], [30, 40, 150, 120]],
+]
+
+language_choices = pytesseract.get_languages()
+
+with gr.Blocks() as demo:
+    with gr.Row():
+        gr.Markdown("# Tesseract OCR with Selection")
+    with gr.Row():
+        img_input = gr.Image(type="filepath", label="Input Image")
+        lang_input = gr.CheckboxGroup(language_choices, type="value", value=['eng'], label='Language')
+        coords_input = gr.Textbox(label="Selection Coordinates (x1, y1, x2, y2)", placeholder="50, 50, 200, 200")
+    with gr.Row():
+        ocr_button = gr.Button("Run OCR with Selection")
+    with gr.Row():
+        ocr_output = gr.Textbox(label="OCR Result")
+    
+    def run_with_selection(image_path, languages, coordinates):
+        if coordinates:
+            coordinates = [int(coord) for coord in coordinates.split(",")]
+        return tesseract_ocr_with_selection(image_path, languages, coordinates)
+    
+    ocr_button.click(
+        run_with_selection,
+        inputs=[img_input, lang_input, coords_input],
+        outputs=[ocr_output]
+    )
+
+if __name__ == '__main__':
+    demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+gradio==3.41.1
+pytesseract==0.3.10
+Pillow==10.0.0
+opencv-python==4.8.0.74