Build Simple OCR

hackmd-github-sync-badge

tags: public,build,ocr,Opencv

We use Pytesseract + OpenCV to build a simple ocr system on colab

Env

This is for colab's simple ocr sample,please copy it or put it in your Google Drive than you can open it.

How to use?

Clone from my github

git clone https://github.com/tingruikp0925/simpleocr.git

After you upload the picture for this sample You can Read/Resize/Flip/Rotate or cvtColor via use cv2 package.

Follow this information and change the config you need.

Read：cv2.imread(filename, flags)
Show：cv2.imshow(window_name, image)
Save：cv2.imwrite(save_filename, image)
Resize：cv2.resize(image, size, interpolation)
Flip：cv2.flip(image, flipCode)
Rotate： M = cv2.getRotationMatrix2D(rotate_center, rotate_angle, ratio) rot_img = cv2.warpAffine(image, M, size)
CvtGray: cv2.cvtColor(image, colorspace_num=7)

Source Code:https://github.com/tingruikp0925/simpleocr

Select a repo