Spaces:
Runtime error
Runtime error
| import cv2 | |
| import numpy as np | |
| import gradio as gr | |
| import pytesseract | |
| def text_read_tesseract(img): | |
| ''' | |
| To read wriiten texts on images | |
| Required: | |
| pip install pytesseract | |
| sudo apt install tesseract-ocr | |
| sudo apt install libtesseract-dev | |
| API: https://pypi.org/project/pytesseract/ | |
| ''' | |
| pytesseract.pytesseract.tesseract_cmd='/usr/bin/tesseract' | |
| imgWords = img.copy() | |
| imgDigits = img.copy() | |
| imgRGB = cv2.cvtColor(src=img, code=cv2.COLOR_BGR2RGB) | |
| # Character detection | |
| texts = pytesseract.image_to_string(imgRGB) | |
| boxes = pytesseract.image_to_boxes(imgRGB) | |
| # print(boxes) | |
| himg, wimg, _ = img.shape | |
| for b in boxes.splitlines(): | |
| b = b.split(' ') | |
| if b[0] != '~': | |
| x1, y1, x2, y2 = int(b[1]), int(b[2]), int(b[3]), int(b[4]) | |
| cv2.rectangle(img, pt1=(x1, himg-y1), pt2=(x2, himg-y2), color=(0,0,255), thickness=1) | |
| # cv2.imshow("Image char", img) | |
| # cv2.waitKey(0) | |
| # cv2.destroyAllWindows() | |
| return texts, img | |
| import easyocr | |
| def text_read_easyocr(img, text_threshold): | |
| # [[189, 75], [469, 75], [469, 165], [189, 165]] | |
| easyocr_reader = easyocr.Reader(lang_list=['en'], gpu=False) | |
| result = easyocr_reader.readtext(img, text_threshold=text_threshold) | |
| himg, wimg, _ = img.shape | |
| text = "" | |
| for box, word, confidence in result: | |
| text += word + "\n" | |
| # x2, y2, x1, y1 = box[0][0], box[0][1], box[2][0], box[2][1] | |
| # cv2.rectangle(img, pt1=(x1, himg-y1), pt2=(x2, himg-y2), color=(0,0,255), thickness=1) | |
| tl, tr, br, bl = box | |
| tl = (int(tl[0]), int(tl[1])) | |
| tr = (int(tr[0]), int(tr[1])) | |
| br = (int(br[0]), int(br[1])) | |
| bl = (int(bl[0]), int(bl[1])) | |
| cv2.rectangle(img, tl, br, (0, 255, 0), 2) | |
| # print(text) | |
| # cv2.imshow("Image char", img) | |
| # cv2.waitKey(0) | |
| return text, img | |
| def read_text(img, choice,text_threshold): | |
| if choice == 0: | |
| return text_read_tesseract(img) | |
| if choice == 1: | |
| return text_read_easyocr(img, text_threshold) | |
| # img = cv2.imread("../images/license.jpeg") | |
| # text_read_tesseract(img) | |
| # text_read_easyocr(img) | |
| # demo = gr.Interface(fn=text_read_tesseract, inputs="image", outputs=[gr.Textbox(lines=2, placeholder="Detected txts here..."), "image"]) | |
| # demo = gr.Interface(fn=text_read_easyocr, inputs="image", outputs=[gr.Textbox(lines=2, placeholder="Detected txts here..."), "image"]) | |
| demo = gr.Interface(fn=read_text, | |
| title="Optical character recognition", | |
| description="Author: Rituraj Kaushik", | |
| examples=[["images/bill.png", "Easy OCR", 0.6], | |
| ["images/passport.jpg", "Easy OCR", 0.6]], | |
| inputs=[gr.Image(label="Input image"), | |
| gr.Radio(choices=["Tesseract", "Easy OCR"], | |
| value= "Easy OCR", | |
| type="index"), | |
| gr.Slider(minimum=0, maximum=1, label="Text confidence threshold", value=0.6, step=0.01) | |
| ], | |
| outputs=[gr.Textbox(lines=2, placeholder="Detected txts here..."), "image"], | |
| allow_flagging="never") | |
| demo.launch() |