Getting Bad image data
when trying to detect text in a part (numpy array rectangle out) of an opencv image. When I try to convert the resulting base64 string online it works without issues (here for example)
im = Image.fromarray(narray.astype("uint8"))
rawBytes = io.BytesIO()
im.save(rawBytes, "PNG")
rawBytes.seek(0) # return to the start of the file
image = vision.types.Image(content = b64encode(rawBytes.read()))
resp = client.text_detection(image = image)
I found this on the topic but why would I need to do that?
Any ideas what I could be doing wrong?