Spaces:
Sleeping
Sleeping
| import pandas as pd | |
| import numpy as np | |
| import streamlit as st | |
| import easyocr | |
| import PIL | |
| from PIL import Image, ImageDraw | |
| from utils import construct_possible_nickname, text_dict | |
| file = None | |
| col1, col2, col3 = st.columns(3) | |
| with col1: | |
| st.image('ubahn.png', width=100) | |
| with col2: | |
| st.title("WIENER UBAHN / VIENNA SUBWAY - GEOCACHE") | |
| with col3: | |
| st.image('gc_logo_grey.png', width=100) | |
| language = st.selectbox( | |
| 'Sprache / Language:', | |
| ('Deutsch', 'English')) | |
| # subtitle | |
| st.markdown(f"#### {text_dict['step1'][language]}") | |
| nickname = st.text_input("Nickname", value="") | |
| if nickname is not None and nickname != "": | |
| with st.spinner(f"{text_dict['wait'][language]}"): | |
| stations, possible_nickname = construct_possible_nickname(nickname) | |
| df_list = [] | |
| for key in stations: | |
| df_list.append([key, ", ".join(stations[key])]) | |
| possible_stations_table = pd.DataFrame(df_list, columns=[text_dict['letter'][language], text_dict['stations'][language]]) | |
| st.divider() | |
| st.markdown(f"#### {text_dict['step2'][language]}") | |
| st.markdown(f"<h3 style='text-align: center;'>{possible_nickname}</3>", | |
| unsafe_allow_html=True) | |
| with st.expander(f"{text_dict['explanation'][language]}"): | |
| st.table(possible_stations_table) | |
| st.divider() | |
| # subtitle | |
| st.markdown(f"#### {text_dict['step3'][language]}") | |
| # upload image file | |
| file = st.file_uploader(label = "Upload", type=['png', 'jpg', 'jpeg']) | |
| #read the csv file and display the dataframe | |
| if file is not None: | |
| with st.spinner(f"{text_dict['wait'][language]}"): | |
| image = Image.open(file) # read image with PIL library | |
| st.image(image) #display | |
| with st.spinner(f"{text_dict['wait'][language]}"): | |
| # it will only detect the English and Turkish part of the image as text | |
| reader = easyocr.Reader(['en','de'], gpu=False) | |
| result = reader.readtext(np.array(image)) # turn image to numpy array | |
| with st.spinner(f"{text_dict['wait'][language]}"): | |
| # collect the results in the dictionary: | |
| textdic_easyocr = {} | |
| for idx in range(len(result)): | |
| pred_coor = result[idx][0] | |
| pred_text = result[idx][1] | |
| pred_confidence = result[idx][2] | |
| textdic_easyocr[pred_text] = {} | |
| textdic_easyocr[pred_text]['Text confidence'] = pred_confidence | |
| # create a data frame which shows the predicted text and prediction confidence | |
| df = pd.DataFrame.from_dict(textdic_easyocr).T | |
| st.markdown(f"### {text_dict['detect'][language]}") | |
| st.table(df) | |
| if nickname is not None and nickname != "" and file is not None: | |
| detected_text = [key.lower() for key in textdic_easyocr.keys()] | |
| if possible_nickname.lower() in detected_text: | |
| st.markdown(f"<h3 style='text-align: center; color: green'>{text_dict['correct'][language]}</h3>", | |
| unsafe_allow_html=True) | |
| else: | |
| st.markdown(f"<h3 style='text-align: center; color: red'>{text_dict['wrong'][language]}</h3>", | |
| unsafe_allow_html=True) | |