Add: Feature segmentation

app.py

@@ -2,6 +2,7 @@ from fastapi import FastAPI, UploadFile, File
 import tensorflow as tf
 import numpy as np
 from PIL import Image
+import cv2
 import io
 import uvicorn
 
@@ -26,15 +27,81 @@ labels = [
 def home():
     return {"message": "Aksara Lontara API is running"}
 
+def preprocess_image(image: np.array):
+    """Mengolah gambar untuk segmentasi huruf sebelum prediksi"""
+    
+    # **1️⃣ Edge Detection (Canny)**
+    edges = cv2.Canny(image, 50, 150)  # Menangkap batas huruf
+    
+    # **2️⃣ Morphological Operations untuk membersihkan noise**
+    kernel = np.ones((3,3), np.uint8)
+    edges_cleaned = cv2.morphologyEx(edges, cv2.MORPH_CLOSE, kernel, iterations=2)
+
+    # **3️⃣ Connected Component Analysis (CCA)**
+    num_labels, labels, stats, _ = cv2.connectedComponentsWithStats(edges_cleaned, connectivity=8)
+
+    # **4️⃣ Filter Huruf Berdasarkan Area**
+    min_area = 500  # Sesuaikan berdasarkan ukuran huruf
+    bounding_boxes = []
+
+    for i in range(1, num_labels):  # Skip background
+        x, y, w, h, area = stats[i]
+        if area > min_area:  # Filter noise kecil
+            bounding_boxes.append((x, y, w, h))
+
+    # **5️⃣ Urutkan huruf berdasarkan posisi X**
+    bounding_boxes.sort(key=lambda b: b[0])
+
+    # **6️⃣ Gabungkan Bounding Box yang Berdekatan**
+    merged_boxes = []
+    merge_threshold = 20  # Jika jarak antar bounding box < 20 piksel, gabungkan
+
+    for i in range(len(bounding_boxes)):
+        x, y, w, h = bounding_boxes[i]
+
+        if merged_boxes and (x - (merged_boxes[-1][0] + merged_boxes[-1][2])) < merge_threshold:
+            # Gabungkan bounding box terakhir dengan yang sekarang
+            x_prev, y_prev, w_prev, h_prev = merged_boxes.pop()
+            x_new = min(x_prev, x)
+            y_new = min(y_prev, y)
+            w_new = max(x_prev + w_prev, x + w) - x_new
+            h_new = max(y_prev + h_prev, y + h) - y_new
+            merged_boxes.append((x_new, y_new, w_new, h_new))
+        else:
+            # Tambahkan sebagai bounding box baru
+            merged_boxes.append((x, y, w, h))
+
+    # **7️⃣ Potong karakter hasil segmentasi**
+    segmented_chars = []
+    for (x, y, w, h) in merged_boxes:
+        char_segment = image[y:y+h, x:x+w]  # Potong area karakter
+        segmented_chars.append(char_segment)
+
+    return segmented_chars
+
 @app.post("/predict")
 async def predict(file: UploadFile = File(...)):
     # Baca gambar
-    image = Image.open(io.BytesIO(await file.read())).convert("L").resize((128, 128))
-    image = np.array(image) / 255.0
-    image = image.reshape(1, 128, 128, 1)
+    image = Image.open(io.BytesIO(await file.read())).convert("L")  # Konversi ke grayscale
+    image = np.array(image)
+
+    # Preprocessing: Segmentasi huruf
+    segmented_chars = preprocess_image(image)
+
+    predictions = []
+    for char_img in segmented_chars:
+        # Resize ke ukuran yang diharapkan oleh model
+        char_img_resized = cv2.resize(char_img, (128, 128))
+        char_img_resized = char_img_resized / 255.0  # Normalisasi
+        char_img_resized = char_img_resized.reshape(1, 128, 128, 1)  # Ubah ke bentuk input model
+
+        # Prediksi
+        prediction = model.predict(char_img_resized)
+        label = labels[np.argmax(prediction)]
+
+        predictions.append(label)
 
-    # Prediksi
-    prediction = model.predict(image)
-    label = labels[np.argmax(prediction)]
+    return {"predictions": predictions}
 
-    return {"prediction": label}
+if __name__ == "__main__":
+    uvicorn.run(app, host="0.0.0.0", port=8000)