基于图像处理和tensorflow实现GTA5的车辆自动驾驶——第二节获取游戏窗口

代码已放到码云

https://gitee.com/photographer_adam/Based-on-image-processing-and-tensorflow-to-realize-GTA5-vehicle-automatic-driving

直接上代码了

import numpy as np
from PIL import ImageGrab
import cv2
import time
def screen_record():
    last_time = time.time()
    while True:
        # 800x600 windowed mode for GTA 5, at the top left position of your main screen.
        # 40 px accounts for title bar. 
        printscreen = np.array(ImageGrab.grab(bbox=(0, 40, 800, 640)))
        print('loop took {} seconds'.format(time.time() - last_time))
        last_time = time.time()
        cv2.imshow('window', cv2.cvtColor(printscreen, cv2.COLOR_BGR2RGB))
        if cv2.waitKey(25) & 0xFF == ord('q'):
            cv2.destroyAllWindows()
            break
screen_record()

实现效果

需要把游戏调整窗口化分辨率为：800*600.放到屏幕左上角，当然也可以不放(需要修改代码，printscreen = np.array(ImageGrab.grab(bbox=(0, 40, 800, 640)))这个部分)

左边为游戏界面，右边为opencv获取到的界面，帧数还是比较高的(1帧能玩，2帧流畅)