TypeError: only size-1 arrays can be converted to Python scalars

[bhavyabpk]

Traceback (most recent call last):
File "yolodetect2.py", line 98, in
draw_prediction(image, class_ids[i], confidences[i], round(x), round(y), round(x+w), round(y+h))
File "yolodetect2.py", line 39, in draw_prediction
cv2.rectangle(img, (x,y), (x_plus_w,y_plus_h), color, 2)
TypeError: only size-1 arrays can be converted to Python scalars

ubuntu version 16.04
python version 2.7.12
opencv version 4.1.0

[arunponnusamy]

Hello @bhavyabpk, Can you try with Python 3 ? I didn't really test with Python 2.7.
Or you can replace round( ) with int( ).

[bhavyabpk]

Yes it is working with python 3 but before the class name which is being printed on the box, there are 3 question mark coming. Like my class name is Tank, so on the bounding box its written ???Tank. I am unable to figure it out why its is so. I think confidence must be coming at its place but there are question marks. I haven't done any changes in the code.

…

On Tue 9 Jul, 2019, 12:28 PM Arun Ponnusamy, ***@***.***> wrote: Hello @bhavyabpk <https://github.com/bhavyabpk>, Can you try with Python 3 ? I didn't really test with Python 2.7. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#18?email_source=notifications&email_token=AH4G2LEOPBKBZQMGZPVWXI3P6QZIXA5CNFSM4H64KVC2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODZPJPHA#issuecomment-509515676>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AH4G2LGAPBHLRDUITPFBXR3P6QZIXANCNFSM4H64KVCQ> .

[arunponnusamy]

The issue could be in reading the class labels from the text file. Try printing the labels before writing on image. Or you can try the example below with cvlib.
https://github.com/arunponnusamy/cvlib/blob/master/examples/object_detection.py

[bhavyabpk]

My issue was resolved when I created new classes.names file and wrote the classes using the keyboard again. For me, it was because there was an unknown character present in my previous classes.names file which resulted in "??"

…

On Thu, Mar 19, 2020 at 10:27 PM jayeshbhole ***@***.***> wrote: Hello @bhavyabpk <https://github.com/bhavyabpk>, Can you try with Python 3 ? I didn't really test with Python 2.7. Or you can replace round( ) with int( ). Sir, I'm getting the same error with Python 3.7. The change is that I've take the image with webcam and not from the system itself. Should I change the type of the image? Code Snippets: image = cv2.VideoCapture(0) image.set(3, 640) image.set(4, 480) while True: _ , img = image.read() #everything else is the same draw_prediction(image, confidences[i], int(x), int(y), int(x+w), int(y+h)) plate = img[(y+20):(y+h-20), (x+20):(x+w-20)] #I've removed the class id as only one object is to be detected Exact Error: File "platedet.py", line 100, in draw_prediction(image, confidences[i], int(x), int(y), int(x+w), int(y+h)) File "platedet.py", line 29, in draw_prediction cv2.rectangle(img, (x,y), (x_plus_w,y_plus_h), color, 15) TypeError: only size-1 arrays can be converted to Python scalars — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#18 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AH4G2LHMIOCPZSWTBL5AAITRIJFG3ANCNFSM4H64KVCQ> .