pytesseract - enhance image before OCR

[kerzol81]

Hi,

I'd like to get pytesseract correctly recognise characters from a security camera's OSD.

Here is a sample:
sample OSD image

And here is the code that doesn't recognise that misses the 0 characters:

def getDate(osd):
    img = Image.open(osd)
    text = pytesseract.image_to_string(img, lang='eng', config='--psm 10 --oem 1 -c tessedit_char_whitelist=-:0123456789')
    print(text)

As I run it on this sample picture, it gives me this:
2618-11-69 16:21:57

Do you guys have any idea, how can I get pytessearct do the job?

Thanks,