import gradio as gr import requests import tensorflow as tf import keras_ocr import cv2 import os import csv import numpy as np import pandas as pd import huggingface_hub from huggingface_hub import Repository from datetime import datetime import scipy.ndimage.interpolation as inter import easyocr from datasets import load_dataset, Image, Features, Array3D from PIL import Image from paddleocr import PaddleOCR import socket # from send_email_user import send_user_email from huggingface_hub import HfApi import smtplib HF_TOKEN = os.environ.get("HF_TOKEN") # mydataset_name = "pragnakalp/OCR-img-to-text" # print("$$$$$$$$$$$$$$$$$$$$$$$$$$$$",type(mydataset_name)) # hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN,mydataset_name) DATASET_REPO_URL = "https://huggingface.co/datasets/pragnakalp/OCR-img-to-text" DATA_FILENAME = "ocr_data.csv" DATA_FILE = os.path.join("ocr_data", DATA_FILENAME) DATA_FILENAME2 = "ocr_image.csv" DATA_FILE2 = os.path.join("ocr_image", DATA_FILENAME2) HF_TOKEN = os.environ.get("HF_TOKEN") DATASET_REPO_ID = "pragnakalp/OCR-img-to-text" print("is none?", HF_TOKEN is None) try: hf_hub_download( repo_id=DATASET_REPO_ID, filename=DATA_FILENAME, cache_dir=DATA_DIRNAME, force_filename=DATA_FILENAME ) except: print("file not found") repo = Repository( local_dir="ocr_data", clone_from=DATASET_REPO_URL, use_auth_token=HF_TOKEN ) def get_device_ip_address(): if os.name == "nt": result = "Running on Windows" hostname = socket.gethostname() result += "\nHostname: " + hostname host = socket.gethostbyname(hostname) result += "\nHost-IP-Address:" + host return result elif os.name == "posix": gw = os.popen("ip -4 route show default").read().split() s = socket.socket(socket.AF_INET, socket.SOCK_DGRAM) s.connect((gw[2], 0)) ipaddr = s.getsockname()[0] gateway = gw[2] host = socket.gethostname() result = "\nIP address:\t\t" + ipaddr + "\r\nHost:\t\t" + host return result else: result = os.name + " not supported yet." return result """ Paddle OCR """ def ocr_with_paddle(img): finaltext = '' ocr = PaddleOCR(lang='en', use_angle_cls=True) # img_path = 'exp.jpeg' result = ocr.ocr(img) for i in range(len(result[0])): text = result[0][i][1][0] finaltext += ' '+ text return finaltext """ Keras OCR """ def ocr_with_keras(img): output_text = '' pipeline=keras_ocr.pipeline.Pipeline() images=[keras_ocr.tools.read(img)] predictions=pipeline.recognize(images) first=predictions[0] for text,box in first: output_text += ' '+ text return output_text """ easy OCR """ # gray scale image def get_grayscale(image): return cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) # Thresholding or Binarization def thresholding(src): return cv2.threshold(src,127,255, cv2.THRESH_TOZERO)[1] def ocr_with_easy(img): gray_scale_image=get_grayscale(img) thresholding(gray_scale_image) cv2.imwrite('image.png',gray_scale_image) reader = easyocr.Reader(['th','en']) bounds = reader.readtext('image.png',paragraph="False",detail = 0) bounds = ''.join(bounds) return bounds # def store_single_disk(image, image_id, label): # """ Stores a single image as a .png file on disk. # Parameters: # --------------- # image image array, (32, 32, 3) to be stored # image_id integer unique ID for image # label image label # """ # Image.fromarray(image).save(disk_dir / f"{image_id}.png") # with open(disk_dir / f"{image_id}.csv", "wt") as csvfile: # writer = csv.writer( # csvfile, delimiter=" ", quotechar="|", quoting=csv.QUOTE_MINIMAL # ) # writer.writerow([label]) """ Generate OCR """ def generate_ocr(Method,img): try: text_output = '' add_csv = [] image_id = 1 print("Method___________________",Method) if Method == 'EasyOCR': text_output = ocr_with_easy(img) if Method == 'KerasOCR': text_output = ocr_with_keras(img) if Method == 'PaddleOCR': text_output = ocr_with_paddle(img) new_data=img.reshape(img.shape) imge = Image.fromarray(new_data.astype(np.uint8),'RGB') add_csv = [Method,imge,text_output] with open(DATA_FILE, "a") as f: writer = csv.writer(f) # write the data writer.writerow(add_csv) commit_url = repo.push_to_hub() print(commit_url) Image.fromarray(image).save(DATA_FILE2 / f"{image_id}.png") with open(DATA_FILE2, "wt") as csvfile: writer = csv.writer( csvfile, delimiter=" ", quotechar="|", quoting=csv.QUOTE_MINIMAL ) writer.writerow([0]) # try: # dataset = load_dataset("pragnakalp/OCR-img-to-text", streaming=True) # print(dataset.features) # except Exception as e: # print("error in loading data",e) # with open(DATA_FILE, "a") as csvfile: # writer = csv.Writer(csvfile) # writer.writerow(add_csv) # commit_url = repo.push_to_hub() # print(commit_url) # save_details(Method,text_output,img) # sender="pragnakalp.dev33@gmail.com" # password="httscgatatbbxxur" # reciever="pragnakalp.dev35@gmail.com" # s = smtplib.SMTP('smtp.gmail.com', 587) # s.starttls() # s.ehlo() # s.login(sender,password) # message = """Subject : Appointment Booking\n\n # Hello, # Your OCR generated successfully""" # s.sendmail(sender, reciever, message) # s.quit() # mailsend=1 # print("Send mail successfully") return text_output except Exception as e: print("Error in ocr generation ==>",e) text_output = "Something went wrong" return text_output """ Save generated details """ def save_details(Method,text_output,img): # print("//////////") hostname = get_device_ip_address() # url = 'https://pragnakalpdev33.pythonanywhere.com/HF_space_image_to_text' # url = 'http://pragnakalpdev35.pythonanywhere.com/HF_space_image_to_text' # myobj = {'Method': Method,'text_output':text_output,'img':img.tolist(),'hostname':hostname} # x = requests.post(url, json = myobj) # method = [] # img_path = [] # text = [] # input_img = '' # hostname = '' # picture_path = "image.jpg" # curr_datetime = datetime.now().strftime('%Y-%m-%d %H-%M-%S') # if text_output: # splitted_path = os.path.splitext(picture_path) # modified_picture_path = splitted_path[0] + curr_datetime + splitted_path[1] # cv2.imwrite("myimage.jpg", img) # with open('savedata.txt', 'w') as f: # print("write test") # f.write("testdata") # print("write Successfully") # # img = Image.open(r"/home/user/app/") # # img.save(modified_picture_path) # input_img = modified_picture_path # try: # df = pd.read_csv("AllDetails.csv") # df2 = {'method': Method, 'input_img': input_img, 'generated_text': text_output} # df = df.append(df2, ignore_index = True) # df.to_csv("AllDetails.csv", index=False) # except: # method.append(Method) # img_path.append(input_img) # text.append(text_output) # dict = {'method': method, 'input_img': img_path, 'generated_text': text} # df = pd.DataFrame(dict,index=None) # df.to_csv("AllDetails.csv") return send_user_email() # return x """ Create user interface for OCR demo """ image = gr.Image(shape=(224, 224),elem_id="img_div") method = gr.Radio(["EasyOCR", "KerasOCR", "PaddleOCR"],value="PaddleOCR",elem_id="radio_div") output = gr.Textbox(label="Output") demo = gr.Interface( generate_ocr, [method,image], output, title="Optical Character Recognition", description="Try OCR with different methods", css=".gradio-container {background-color: lightgray} #radio_div {background-color: #FFD8B4; font-size: 40px;}", allow_flagging = "manual" # flagging_dir = "flagged", # flagging_callback=hf_writer ) demo.launch(enable_queue = False)