我真的不想秃头 发表于 2021-12-23 19:28:37

车牌识别出现的问题

出现问题的源代码：
img = cv2.imread('D:/pycharm/cv-cnn-lpr-main/data/testset/set1/19.jpg')

    resize_img = img_resize(img)

    # 预处理
    pred_img = pre_process(resize_img)

    # 车牌定位
    car_plate_list = locate_plate(resize_img, pred_img)

    # CNN车牌过滤
    ret,cnn_plate = cnn_select_plate(car_plate_list, plate_model_path)
    if not ret:
      print("未检测到车牌")
    cv2.imshow('cnn_plate_img', cnn_plate)

    cv2.imwrite('./data/output/plate_cnn.jpg', cnn_plate)


问题：
Traceback (most recent call last):
File "D:/pycharm/cv-cnn-lpr-main/lpr_main.py", line 457, in <module>
    cv2.imshow('cnn_plate_img', cnn_plate)
cv2.error: OpenCV(4.5.4) :-1: error: (-5:Bad argument) in function 'imshow'
> Overload resolution failed:
>- mat is not a numpy array, neither a scalar
>- Expected Ptr<cv::cuda::GpuMat> for argument 'mat'
>- Expected Ptr<cv::UMat> for argument 'mat'

我真的不想秃头 发表于 2021-12-23 19:29:07

请问一下各位该怎么解决

hrpzcf 发表于 2021-12-23 21:01:47

else:
    cv2.imshow('cnn_plate_img', cnn_plate)
    cv2.imwrite('./data/output/plate_cnn.jpg', cnn_plate)

我真的不想秃头 发表于 2021-12-26 15:08:36

hrpzcf 发表于 2021-12-23 21:01
else:
    cv2.imshow('cnn_plate_img', cnn_plate)
    cv2.imwrite('./data/output/plate_cnn.jpg', cn ...

还是报错 但错误类型变了：
Traceback (most recent call last):
File "D:/pycharm/cv-cnn-lpr-main/lpr_main.py", line 470, in <module>
    char_img_list = extract_char(cnn_plate)
File "D:/pycharm/cv-cnn-lpr-main/lpr_main.py", line 370, in extract_char
    gray_plate = cv2.cvtColor(car_plate_extract, cv2.COLOR_BGR2GRAY)
cv2.error: OpenCV(4.5.4) :-1: error: (-5:Bad argument) in function 'cvtColor'
> Overload resolution failed:
>- src is not a numpy array, neither a scalar
>- Expected Ptr<cv::UMat> for argument 'src'
错误的两个地方代码：
1.
# 车牌定位
    car_plate_list = locate_plate(resize_img, pred_img)

    # CNN车牌过滤
    ret, cnn_plate = cnn_select_plate(car_plate_list, plate_model_path)
    if not ret:
      print("未检测到车牌")
    else:
      cv2.imshow('cnn_plate_img', cnn_plate)
      cv2.waitKey(0)

      cv2.imwrite('./data/output/plate_cnn.jpg', cnn_plate)


    # 字符提取
    char_img_list = extract_char(cnn_plate)

    # CNN字符识别
    text = cnn_recognize_char(char_img_list, char_model_path)
    print(text)

    cv2.waitKey(0)
2.
# 识别字符并排列
def extract_char(car_plate_extract):
    gray_plate = cv2.cvtColor(car_plate_extract, cv2.COLOR_BGR2GRAY)
    # retn, binary_plate = cv2.threshold(gray_plate, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)
    binary_plate = cv2.adaptiveThreshold(gray_plate, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 17, -5)
    char_img_lists = get_chars(binary_plate)
    return char_img_lists
页: [1]
查看完整版本: 车牌识别出现的问题