Inability to recognize text on images (attached)

[Dmitry Klimenkov]

Hi! I am in stuck with recognizing text these images

My preprocessing pipeline (these are not original images as you see) is:

1. Increasing size of images (x8), increased images are attached

2. Grayscaling

3. Binarization (

image = cv2.threshold(image, 0, 255, cv2.THRESH_TOZERO | cv2.THRESH_OTSU)[1]

Is there any way to solve this problem?

Thank you!