feat: Make predictions by uploading a image

Project/Backend/Server_Python/.gitignore

@@ -12,3 +12,6 @@
 *.vob
 *.m4v
 *.ts
+
+files/*
+!files/
\ No newline at end of file

Project/Backend/Server_Python/controllers/translate_controler.py

+import base64
+import os
+import cv2
 from fastapi import APIRouter, File, HTTPException,UploadFile
+import numpy as np
+from pydantic import BaseModel
+import tensorflow as tf
 
 from core import setup_logger
 
@@ -6,19 +12,53 @@ from core import setup_logger
 router = APIRouter()
 logger = setup_logger()
 
+class ImageRequest(BaseModel):
+    image: UploadFile
+
+# Load your Keras model
+model = tf.keras.models.load_model('D:\RP\SL-Detection-Action-Recognition\models\model')
+CLASSES = os.listdir('D:\RP\SL-Detection-Action-Recognition\data\Sn_sign_language_dataset')  # list of classes
+NUM_CLASSES = len(CLASSES)  # number of classes
+IMG_SIZE = 224  # image size
+
 @router.post("/upload/video")
 async def upload_video(video: UploadFile = File(...)):
     try:
-        # Save the uploaded video to the "uploads" folder
-        logger.info("Received request at root endpoint.")
+  
         file_location = f"files/{video.filename}"
         with open(file_location, "wb") as file:
             file.write(video.file.read())
 
-        return {"filename": video.filename}
+        return {"text": "ආආආආආආආ"}
     except Exception as e:
         logger.info(f"Failed to upload file. {e}")
         raise HTTPException(
             status_code=500,
             detail="Failed to upload the video"
         )
+    
+@router.post('/predict-sign-language')
+def predict(image_request:  UploadFile = File(...)):
+    try:
+
+        file_location = f"files/{image_request.filename}"
+        with open(file_location, "wb") as file:
+                file.write(image_request.file.read())
+
+        # Load the saved image using OpenCV
+        img = cv2.imread(file_location)
+        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+        img = cv2.resize(img, (IMG_SIZE, IMG_SIZE))
+        img = np.array([img], dtype=np.float32) / 255.0
+
+        # Make prediction
+        prediction = model.predict(img)
+        class_name = CLASSES[np.argmax(prediction)]
+
+        return {'class_name': class_name}
+    except Exception as e:
+        logger.info(f"Failed to make predictions. {e}")
+        raise HTTPException(
+            status_code=500,
+            detail="Failed to make predictions"
+        )
\ No newline at end of file

Project/Backend/Server_Python/requirements.txt

 fastapi
 uvicorn
 python-multipart==0.0.6
+tensorflow==2.12.0
+opencv-python==4.7.0.72
\ No newline at end of file