预测预训练的移动网络模型的实时视频源

问题描述

我能够为图像上传预测相同的模型，但无法为实时视频提要预测相同的模型。我正在尝试从实时视频源中提取一帧并使用 mobilenet 模型对其进行预处理以进行类别预测

import cv2


#importing objects from flask
from flask import Flask,request,render_template,send_from_directory
app = Flask(__name__)  #needs this because flask sets up some paths behind the scene : name is the special variable holds the model


# Flask
from flask import Flask,redirect,url_for,Response,jsonify,redirect
from werkzeug.utils import secure_filename
from gevent.pywsgi import WsgiServer

# TensorFlow and tf.keras
import tensorflow as tf
from tensorflow import keras

from tensorflow.keras.applications.imagenet_utils import preprocess_input,decode_predictions
from tensorflow.keras.models import load_model
from tensorflow.keras.preprocessing import image

# Some utilites
import numpy as np


#importing objects from flask
from flask import Flask,send_from_directory
app = Flask(__name__)  #needs this because flask sets up some paths behind the scene : name is the special variable holds the model

APP_ROOT = os.path.dirname(os.path.abspath(__file__))   #file variable having file name: absolute path func / relative path

classes = ['Class1','Class2','Class3','Class4']

new_model = load_model('model.h5')    #create var new_model for the file model.h5
new_model.summary()

cap = cv2.VideoCapture(0)
if not cap.isOpened():
        raise IOError("We cannot open webcam")

while True:
        frame = cap.read()
        
        frame = cv2.flip(frame,cv2.COLOR_BGR2RGB)
        test_image = image.load_img(frame,target_size=(224,224))

        # Preprocessing the image
        test_image = image.img_to_array(test_image)
        # x = np.true_divide(x,255)
        test_image = np.expand_dims(test_image,axis=0)

        # Be careful how your trained model deals with the input
        # otherwise,it won't make correct prediction!
        test_image = tf.keras.applications.mobilenet.preprocess_input(test_image)

        result = new_model.predict(test_image)
        prediction = classes[np.argmax(result)]
        score = float("%0.2f" % (max(result[0]) * 100))
        #score = float("{.2f}".format((classes) * 100))
        print(f'Result: {prediction},score: {score}')
        cv2.imshow("frame",frame)


        if cv2.waitKey(25) & 0xFF == ord('q'):
            break

cv2.destroyAllWindows()
cap.release()

我在执行代码时遇到以下错误。 v2.imshow("frames",frame) 类型错误：参数“mat”的预期 Ptr<:umat>

解决方法

暂无找到可以解决该程序问题的有效方法，小编努力寻找整理中！

如果你已经找到好的解决方法，欢迎将解决方案带上本链接一起发送给小编。

小编邮箱:dio#foxmail.com (将#修改为@）