Roboflow cv2.imshow 错误不断发生，即使昨天还有效。怎么解决？

Question

我不断收到此错误：

cv2.error: OpenCV(4.9.0) D:\a\opencv-python\opencv-python\opencv\modules\highgui\src\window.cpp:971: error: (-215:Assertion failed) size.width>0 && size.height>0 in function 'cv::imshow'

我已经完成了所有其他修复，这些修复实际上工作了一天，然后没有（例如将版本更新为 name/version#）。但现在，即使没有更改代码，我又再次收到错误。

这是我的代码部分：

upload_url = "".join([
    "https://detect.roboflow.com/",
    "basketball-cv/3",
    "?api_key=",
    "*************",
    "&format=image",
    "&stroke=5"
])

async def infer(requests):
    # Get the current image from the webcam
    ret, img = video.read()

    # Resize (while maintaining the aspect ratio) to improve speed and save bandwidth
    height, width, channels = img.shape
    scale = ROBOFLOW_SIZE / max(height, width)
    img = cv2.resize(img, (round(scale * width), round(scale * height)))

    # Encode image to base64 string
    retval, buffer = cv2.imencode('.jpg', img)
    img_str = base64.b64encode(buffer)

    # Get prediction from Roboflow Infer API
    resp = await requests.post(upload_url, data=img_str, headers={
        "Content-Type": "application/x-www-form-urlencoded"
    })

    # Parse result image
    image = np.asarray(bytearray(resp.content), dtype="uint8")
    image = cv2.imdecode(image, cv2.IMREAD_COLOR)

    return image
...

 cv2.imshow('image', image)

问题是什么？

我尝试更改版本号、更新大小和 API 密钥

Answer 1

@ChristophRackwitz 和 @DanMašek 已经提供了一些提示：

1. 您没有检查

   video.read()

   是否返回帧，假设

   video

   是

   cv2.VideoCapture

   的实例，您应该在假设方法返回帧之前检查

   ret

   ，正如OpenCV文档建议的那样

2. 您没有检查

   requests.post

   返回的响应

我还要补充一点，很难推理您提供的示例代码，似乎

video

是一个全局变量，并且在脚本执行到达

infer

之前，它可能会在代码中的其他位置关闭。

您可以考虑稍微重构一下您的代码，这样您就可以更轻松地推断出问题所在：

import cv2
import numpy as np
import requests


class ReadingFrameFailed(Exception):
    pass


class RequestsPostFailed(Exception):
    pass


class MalformedResponse(Exception):
    pass


# (...)


async def infer(
    video: cv2.VideoCapture,
    model_id,
    roboflowapi_key,
    requests: ????,
    upload_url_template: str = "https://detect.roboflow.com/{model_id}?api_key={roboflow_api_key}&format=image&stroke=5",
) -> np.array:
    ret, img = video.read()

    if not ret:
        raise ReadingFrameFailed("Failed to grab frame")

    # logic for resizing and serializing frame
    # (...)

    upload_url = upload_url_template.format(**{
        "model_id": model_id,
        "roboflow_api_key": roboflow_api_key,
    })

    try:
        resp = await requests.post(upload_url, data=img_str,
            headers={"Content-Type": "application/x-www-form-urlencoded"})
    except Exception as exc:
        # ugly but since we don't know what `requests` object is...
        raise RequestsPostFailed(str(exc))

    image = np.asarray(bytearray(resp.content), dtype="uint8")
    image = cv2.imdecode(image, cv2.IMREAD_COLOR)
    if not image:
        raise MalformedResponse("Failed to parse image, API returned unexpected result")

    return image