使用Python与OpenCV的cv2.matchTemplate时触发断言错误

机器人截图

运行用于Runescape的bot.py脚本时触发了断言错误，优化移动逻辑时遇到此问题，试过Stack Overflow上的解决方案但无效。本人是编程和Stack Overflow新手。

相关代码 #

def __init__(self, window_offset, window_size):
    self.lock = Lock()

    self.window_offset = window_offset
    self.window_w = window_size[0]
    self.window_h = window_size[1]

    self.limestone_tooltip = cv.imread('limestone_tooltip.jpg', cv.IMREAD_UNCHANGED)

    self.state = BotState.INITIALIZING
    self.timestamp = time()

def click_next_target(self):
    targets = self.targets_ordered_by_distance(self.targets)

    target_i = 0
    found_limestone = False
    while not found_limestone and target_i < len(targets):
        # if we stopped our script, exit this loop
        if self.stopped:
            break

        target_pos = targets[target_i]
        screen_x, screen_y = self.get_screen_position(target_pos)
        print('Moving mouse to x:{} y:{}'.format(screen_x, screen_y))
        if self.confirm_tooltip(target_pos):
            print('Click on confirmed target at x:{} y:{}'.format(screen_x, screen_y))
            found_limestone = True
            pyautogui.click()
            # save this position to the click history
            self.click_history.append(target_pos)
        target_i += 1

    return found_limestone

def confirm_tooltip(self, target_position):
    result = cv.matchTemplate(self.screenshot, self.limestone_tooltip, cv.TM_CCOEFF_NORMED)
    min_val, max_val, min_loc, max_loc = cv.minMaxLoc(result)
# object found
    if max_val >= self.TOOLTIP_MATCH_THRESHOLD:
        return True
    # print('Tooltip not found.')
    return False

def get_screen_position(self, pos):
    return (pos[0] + self.window_offset[0], pos[1] + self.window_offset[1])

错误信息 #

Traceback (most recent call last):
  File "C:\Python311\Lib\threading.py", line 1038, in _bootstrap_inner
    self.run()
  File "C:\Python311\Lib\threading.py", line 975, in run
    self._target(*self._args, **self._kwargs)
  File "D:\DEV\OpenCvTeste\teste_de_fogo\bot.py", line 227, in run
    success = self.click_next_target()
              ^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\DEV\OpenCvTeste\teste_de_fogo\bot.py", line 90, in click_next_target
    if self.confirm_tooltip(target_pos):
       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\DEV\OpenCvTeste\teste_de_fogo\bot.py", line 149, in confirm_tooltip
    result = cv.matchTemplate(self.screenshot, self.limestone_tooltip, cv.TM_CCOEFF_NORMED)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
cv2.error: OpenCV(4.9.0) D:\a\opencv-python\opencv-python\opencv\modules\imgproc\src\templmatch.cpp:1164: error: (-215:Assertion failed) (depth == CV_8U || depth == CV_32F) && type == _templ.type() && _img.dims() <= 2 in function 'cv::matchTemplate'

问题原因与解决方法 #

原因 #

OpenCV的matchTemplate函数有三个硬性要求：

1. 源图像和模板图像的深度必须是CV_8U（8位无符号整数）或CV_32F（32位浮点数）
2. 两者的图像类型（通道数、数据类型）必须完全一致
3. 图像维度不能超过2（不能是3D数组）

你的代码中，用cv.IMREAD_UNCHANGED读取的模板图可能包含Alpha通道（4通道），而self.screenshot大概率是3通道的BGR图像，导致类型不匹配；也可能存在图像深度不一致的情况。

解决步骤 #

1. 统一图像通道数：修改模板图读取方式，去掉Alpha通道，改用cv.IMREAD_COLOR读取：

   self.limestone_tooltip = cv.imread('limestone_tooltip.jpg', cv.IMREAD_COLOR)
   

2. 确保截图格式正确：如果self.screenshot是通过pyautogui.screenshot()获取的，需要转换成OpenCV的BGR格式（pyautogui返回的是RGB）：

   # 假设截图代码示例，需转换格式
   screenshot_pil = pyautogui.screenshot(region=...)
   self.screenshot = cv.cvtColor(np.array(screenshot_pil), cv.COLOR_RGB2BGR)
   

3. 检查图像尺寸：确保模板图的宽高小于等于源图像（截图）的宽高，否则matchTemplate也会触发断言错误。

内容的提问来源于stack exchange，提问作者Rupelio