app file uploading

Flask/app.py

+import cv2
+import numpy as np
+import base64
+from io import BytesIO
+from PIL import Image
+import tensorflow as tf
+from flask_cors import CORS
+from flask import Flask, request, jsonify
+
+print(tf.__version__)
+print(tf.keras.__version__)
+
+app = Flask(__name__)
+CORS(app)
+
+class_dict_defect = {
+    'defective': 0,
+    'qualified': 1
+}
+
+class_dict_defect_rev = {
+    0: 'defective',
+    1: 'qualified'
+}
+
+model_defects = tf.keras.models.load_model('models/courier-defect-detector.h5')
+
+@app.route('/defects', methods=['POST'])
+def defects():
+    if request.method == 'POST':
+        data = request.json
+        base64_image = data['image']
+        image_bytes = base64.b64decode(base64_image)
+        image = np.array(Image.open(BytesIO(image_bytes)))
+        filename = "uploads/image.jpg"
+        cv2.imwrite(filename, image)
+        result = inference_model(filename)
+        return jsonify({
+            "status": "success",
+            "defectiveness": result,
+        }), 200
+    
+    return jsonify({
+        "status": "unsuccess",
+        "defectiveness": None,
+    }), 400
+
+def inference_model(filename):
+    try:
+        img = cv2.imread(filename)
+        img = cv2.resize(img, (224, 224))
+        if img is not None and not img.size == 0:
+            img = tf.keras.applications.xception.preprocess_input(img)
+            img = np.expand_dims(img, axis=0)
+            pred = model_defects.predict(img)
+            pred = pred.squeeze() > 0.5
+            pred = pred.squeeze()
+            return class_dict_defect_rev[int(pred)]
+        
+        else:
+            print("Invalid or empty image:", filename)
+            return None
+    except Exception as e:
+        print(f"Error processing image: {str(e)}")
+        return None
+
+if __name__ == '__main__':
+    app.run(debug=True)