app.py

from flask import Flask,render_template
from flask_socketio import SocketIO,emit
import base64
import numpy as np
import cv2 
import numpy as np
from keras.models import load_model


app = Flask(__name__)
app.config['SECRET_KEY'] = 'secret!'
socket = SocketIO(app,async_mode="eventlet")


# load model and labels
np.set_printoptions(suppress=True)
model = load_model(r"keras_model.h5", compile=False)
class_names = open(r"labels.txt", "r").readlines()

def base64_to_image(base64_string):
    # Extract the base64 encoded binary data from the input string
    base64_data = base64_string.split(",")[1]
    # Decode the base64 data to bytes
    image_bytes = base64.b64decode(base64_data)
    # Convert the bytes to numpy array
    image_array = np.frombuffer(image_bytes, dtype=np.uint8)
    # Decode the numpy array as an image using OpenCV
    image = cv2.imdecode(image_array, cv2.IMREAD_COLOR)
    return image

@socket.on("connect")
def test_connect():
    print("Connected")
    emit("my response", {"data": "Connected"})

@socket.on("image")
def receive_image(image):
    # Decode the base64-encoded image data
    image = base64_to_image(image)
    image = cv2.resize(image, (224, 224), interpolation=cv2.INTER_AREA)
    # emit("processed_image", image)
    # Make the image a numpy array and reshape it to the models input shape.
    image = np.asarray(image, dtype=np.float32).reshape(1, 224, 224, 3)
    image = (image / 127.5) - 1
    # Predicts the model
    prediction = model.predict(image)
    index = np.argmax(prediction)
    class_name = class_names[index]
    confidence_score = prediction[0][index]
    emit("result",{"name":str(class_name),"score":str(confidence_score)})

@app.route("/")
def home():
    return render_template("index.html")

if __name__ == '__main__':
    # app.run(debug=True)
    socket.run(app,host="0.0.0.0", port=7860)