Create app.py

app.py

@@ -0,0 +1,122 @@
+import os
+import json
+import cv2
+import gradio as gr
+import numpy as np
+from PIL import Image
+
+# Path to images folder
+image_folder = "./olmocr_images"
+output_folder = "./olmocr_annotations"
+
+# Ensure output folder exists
+os.makedirs(output_folder, exist_ok=True)
+
+# Get all image files
+image_files = sorted([f for f in os.listdir(image_folder) if f.lower().endswith(('.png', '.jpg', '.jpeg'))])
+image_index = 0  # Start at first image
+
+# JSON template for annotation
+default_json_template = {
+    "image_path": "",
+    "text": {
+        "TITLE": "",
+        "TEMPAT-TANGGAL": "",
+        "NOMOR REGISTRASI": "",
+        "No.": "",
+        "NAMA PEMILIK": "",
+        "ALAMAT": "",
+        "MERK": "",
+        "TYPE": "",
+        "JENIS": "",
+        "MODEL": "",
+        "TAHUN PEMBUATAN": "",
+        "ISI SILINDER/DAYA LISTRIK": "",
+        "NOMOR RANGKA": "",
+        "NOMOR MESIN": "",
+        "NIK/TDP/NIB/KITAS/KITAP": "",
+        "WARNA": "",
+        "BAHAN BAKAR": "",
+        "WARNA TNKB": "",
+        "TAHUN REGISTRASI": "",
+        "NOMOR BPKB": "",
+        "KODE LOKASI": "",
+        "NO URUT PENDAFTARAN": "",
+        "BERLAKU SAMPAI": ""
+    }
+}
+
+def load_image():
+    global image_index
+    if 0 <= image_index < len(image_files):
+        image_path = os.path.join(image_folder, image_files[image_index])
+        image = Image.open(image_path)
+        
+        json_filename = f"license_{image_index+1:03d}.json"
+        json_path = os.path.join(output_folder, json_filename)
+        annotation_data = default_json_template.copy()
+
+        if os.path.exists(json_path):
+            with open(json_path, "r") as f:
+                annotation_data = json.load(f)
+
+        return image, annotation_data["text"], image_path
+    return None, {}, ""
+
+def save_annotation(image_path, annotation_text):
+    global image_index
+    if 0 <= image_index < len(image_files):
+        json_filename = f"license_{image_index+1:03d}.json"
+        json_path = os.path.join(output_folder, json_filename)
+        
+        annotation_data = {"image_path": image_path, "text": annotation_text}
+        
+        with open(json_path, "w") as f:
+            json.dump(annotation_data, f, indent=4)
+        
+        return f"✅ Saved: {json_filename}"
+    return "❌ Error saving annotation"
+
+def next_image():
+    global image_index
+    if image_index < len(image_files) - 1:
+        image_index += 1
+    return load_image()
+
+def prev_image():
+    global image_index
+    if image_index > 0:
+        image_index -= 1
+    return load_image()
+
+# Gradio Interface
+with gr.Blocks() as demo:
+    with gr.Row():
+        image_display = gr.Image()
+        
+        with gr.Column():
+            text_inputs = {key: gr.Textbox(label=key) for key in default_json_template["text"]}
+            save_button = gr.Button("Save")
+    
+    status = gr.Textbox(label="Status", interactive=False)
+    
+    with gr.Row():
+        prev_button = gr.Button("Previous")
+        next_button = gr.Button("Next")
+    
+    # Load first image on startup
+    image, annotation, img_path = load_image()
+    image_display.value = image
+    for key, textbox in text_inputs.items():
+        textbox.value = annotation.get(key, "")
+    
+    # Button Actions
+    def save_action():
+        annotation_text = {key: text_inputs[key].value for key in text_inputs}
+        return save_annotation(img_path, annotation_text)
+    
+    save_button.click(save_action, outputs=status)
+    next_button.click(lambda: next_image(), outputs=[image_display, *text_inputs.values()])
+    prev_button.click(lambda: prev_image(), outputs=[image_display, *text_inputs.values()])
+
+demo.launch()