Update app.py

app.py

@@ -2,34 +2,96 @@ import gradio as gr
 import cv2
 import numpy as np
 from PIL import Image
-from EasyOpticalCharacterRecognition import process_image
+import pickle
+from tensorflow.keras.models import load_model
+from tensorflow.keras.preprocessing.image import img_to_array
+import easyocr
 
-# Wrapper function for Gradio interface
+# === Load Model and Label Encoder ===
+model_path = "MobileNetBest_Model.h5"
+label_path = "MobileNet_Label_Encoder.pkl"
+
+model = load_model(model_path)
+print("✅ Model loaded.")
+
+# Load label encoder
+try:
+    with open(label_path, 'rb') as f:
+        label_map = pickle.load(f)
+        index_to_label = {v: k for k, v in label_map.items()}
+    print("✅ Label encoder loaded:", index_to_label)
+except:
+    index_to_label = {0: "Handwritten", 1: "Computerized"}
+    print("⚠️ Label encoder not found. Using default:", index_to_label)
+
+# === Classify Region ===
+def classify_text_region(region_img):
+    try:
+        region_img = cv2.resize(region_img, (224, 224))
+        region_img = region_img.astype("float32") / 255.0
+        region_img = img_to_array(region_img)
+        region_img = np.expand_dims(region_img, axis=0)
+
+        preds = model.predict(region_img)
+        if preds.shape[-1] == 1:
+            return "Computerized" if preds[0][0] > 0.5 else "Handwritten"
+        else:
+            class_idx = np.argmax(preds[0])
+            return index_to_label.get(class_idx, "Unknown")
+    except Exception as e:
+        print("❌ Classification error:", e)
+        return "Unknown"
+
+# === OCR and Annotate ===
+def AnnotatedTextDetection_EasyOCR_from_array(img):
+    reader = easyocr.Reader(['en'], gpu=False)
+    results = reader.readtext(img)
+    annotated_results = []
+
+    for (bbox, text, conf) in results:
+        if conf < 0.3 or text.strip() == "":
+            continue
+
+        x1, y1 = map(int, bbox[0])
+        x2, y2 = map(int, bbox[2])
+        w, h = x2 - x1, y2 - y1
+
+        crop = img[y1:y2, x1:x2]
+        if crop.size == 0:
+            continue
+
+        label = classify_text_region(crop)
+        annotated_results.append(f"{text.strip()} → {label}")
+
+        color = (0, 255, 0) if label == "Computerized" else (255, 0, 0)
+        cv2.rectangle(img, (x1, y1), (x2, y2), color, 2)
+        cv2.putText(img, label, (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.6, color, 1)
+
+    return cv2.cvtColor(img, cv2.COLOR_BGR2RGB), "\n".join(annotated_results)
+
+# === Gradio Wrapper ===
 def infer(image):
     img = np.array(image)
-    annotated_img, result_text = process_image(img)
+    annotated_img, result_text = AnnotatedTextDetection_EasyOCR_from_array(img)
     return Image.fromarray(annotated_img), result_text
 
-# Custom CSS for light blue background and bordered elements
+# === Custom CSS ===
 custom_css = """
-body 
-{
+body {
     background-color: #e6f2ff;
 }
-.gradio-container 
-{
+.gradio-container {
     border-radius: 12px;
     padding: 20px;
     border: 2px solid #007acc;
 }
-.gr-input, .gr-output 
-{
+.gr-input, .gr-output {
     border: 1px solid #007acc;
     border-radius: 10px;
 }
 """
 
-# Gradio Interface
+# === Launch Interface ===
 demo = gr.Interface(
     fn=infer,
     inputs=gr.Image(type="pil", label="Upload Image"),
@@ -37,9 +99,10 @@ demo = gr.Interface(
         gr.Image(type="pil", label="Annotated Image"),
         gr.Textbox(label="Detected Text and Classification")
     ],
-    title="🧠 OCR & Text Type Classifier",
-    description="Application detect text from images and classifier classify text into Computerized & Handwritten Text.",
-    theme="soft",  # Optional: you can also try "default" or "huggingface"
+    title="Text Detection and Clssification",
+    description="This application detects text using EasyOCR and classifies each text region as Handwritten or Computerized using a MobileNet model.",
+    theme="soft",
     css=custom_css
 )
+
 demo.launch()