File size: 7,091 Bytes
02b08d8
 
 
 
 
 
dcfa000
02b08d8
 
 
 
 
 
 
 
 
 
 
 
 
 
dcfa000
 
 
 
 
02b08d8
dcfa000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
02b08d8
dcfa000
02b08d8
 
dcfa000
 
 
 
 
 
6007cdb
02b08d8
 
dcfa000
02b08d8
fdf64e2
dcfa000
40b4fe1
 
 
 
dcfa000
 
 
 
 
 
 
 
 
 
 
 
 
 
867c191
dcfa000
 
 
 
 
 
 
 
 
02b08d8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
dcfa000
fdf64e2
02b08d8
 
 
 
 
 
 
 
 
dcfa000
02b08d8
 
867c191
02b08d8
dcfa000
02b08d8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cf88914
02b08d8
 
 
 
 
 
65ee0df
02b08d8
 
 
 
 
 
 
 
 
dcfa000
ffa6155
02b08d8
 
 
 
 
 
 
867c191
02b08d8
867c191
 
02b08d8
dcfa000
 
02b08d8
 
 
 
 
 
 
 
 
dcfa000
02b08d8
 
 
 
 
 
 
 
 
 
 
 
 
dcfa000
02b08d8
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
# app.py
import streamlit as st
import cv2
import numpy as np
from PIL import Image
import easyocr
import os
from streamlit_paste_button import paste_image_button as pbutton

color_ranges = {
    'fire': 'B50B0E',
    'water': '015AB6',
    'wind': '1F6A0B',
    'earth': '623F23',
    'light': 'DA8D09',
    'dark': '502181'
}

def hex_to_rgb(hex_code):
    return tuple(int(hex_code[i:i+2], 16) for i in (0, 2, 4))

# # ์บ์‹ฑ์œผ๋กœ Reader๋ฅผ ํ•œ ๋ฒˆ๋งŒ ๋กœ๋“œ
# @st.cache_resource
# def load_reader():
#     return easyocr.Reader(['en'], gpu=False, verbose=False)

@st.cache_resource
def load_number_templates():
    templates = {}
    
    template_dir = './templates'
    
    for num in range(1, 16):
        template_path = os.path.join(template_dir, f'{num}.png')
        
        if os.path.exists(template_path):
            # ๊ทธ๋ ˆ์ด์Šค์ผ€์ผ๋กœ ๋กœ๋“œ
            template = cv2.imread(template_path, cv2.IMREAD_GRAYSCALE)
            
            if template is not None:
                # ์ „์ฒ˜๋ฆฌ: ์ด์ง„ํ™”
                _, template = cv2.threshold(template, 127, 255, cv2.THRESH_BINARY)
                templates[num] = template
            else:
                st.warning(f"ํ…œํ”Œ๋ฆฟ {num}.png๋ฅผ ๋กœ๋“œํ•  ์ˆ˜ ์—†์Šต๋‹ˆ๋‹ค.")
    if not templates:
        st.error("ํ…œํ”Œ๋ฆฟ ์ด๋ฏธ์ง€๋ฅผ ์ฐพ์„ ์ˆ˜ ์—†์Šต๋‹ˆ๋‹ค. templates/ ํด๋”์— 1.png ~ 15.png ํŒŒ์ผ์„ ์ถ”๊ฐ€ํ•ด์ฃผ์„ธ์š”.")

    return templates

def extract_region(region):
    try:
        gray = cv2.cvtColor(region, cv2.COLOR_BGR2GRAY)
        _, binary = cv2.threshold(gray, 215, 255, cv2.THRESH_BINARY)
                
        # ๋…ธ์ด์ฆˆ ์ œ๊ฑฐ
        kernel = np.ones((2, 2), np.uint8)
        cleaned = cv2.morphologyEx(binary, cv2.MORPH_CLOSE, kernel)
        cleaned = cv2.morphologyEx(cleaned, cv2.MORPH_OPEN, kernel)
        return cleaned

    except Exception as e:
        return None

def template_matching(region, templates):
    try:
        h, w = region.shape
        aspect_ratio = w / h
        new_width = int(25 * aspect_ratio)
        region = cv2.resize(region, (new_width, 25))
        best_match = 0
        best_score = -1
        for num, template in templates.items():
            try:
                result = cv2.matchTemplate(region, template, cv2.TM_CCOEFF_NORMED)
                _, max_val, _, _ = cv2.minMaxLoc(result)
                print(max_val)
                if max_val > best_score:
                    best_score = max_val
                    best_match = num
            except Exception as e:
                print(f"Error matching template {num}: {e}")
                continue

        if best_score > 0.65:
            return best_match
        else:
            return 0
    except Exception as e:
        print(f"Error matching template: {e}")
        return 0


def find_items(img_array, color_range, templates):
    img_rgb = cv2.cvtColor(img_array, cv2.COLOR_BGR2RGB)
    results = {}

    for type, hex_color in color_range.items():
        target_rgb = hex_to_rgb(hex_color)
        lower_c = np.array([max(0, c - 10) for c in target_rgb])
        upper_c = np.array([min(255, c + 10) for c in target_rgb])
        
        mask = cv2.inRange(img_rgb, lower_c, upper_c)
        contours, _ = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

        if contours:
            largest = max(contours, key=cv2.contourArea)
            if cv2.contourArea(largest) > 100:
                x, y, w, h = cv2.boundingRect(largest)
                number_region = img_rgb[y+int(h*1.2) :y+ int(h*1.7), x+int(w*1.4):x+int(w*2.2)]
                count = template_matching(extract_region(number_region), templates)
                results[type] = count
            else:
                results[type] = 0
        else:
            results[type] = 0

    return results


def process_image(img, templates):
    st.image(cv2.cvtColor(img, cv2.COLOR_BGR2RGB), caption='์—…๋กœ๋“œ๋œ ์ด๋ฏธ์ง€', use_container_width=True)
    
    with st.spinner('์•„์ดํ…œ ๊ฐœ์ˆ˜๋ฅผ ์„ธ๋Š” ์ค‘...'):
        try:
            results = find_items(img, color_ranges, templates)
            
            # ๊ฒฐ๊ณผ ํ‘œ์‹œ
            st.success('โœ… ๋ถ„์„ ์™„๋ฃŒ!')
            
            col1, col2, col3 = st.columns(3)
            
            emoji_map = {
                'fire': '๐Ÿ”ฅ',
                'water': '๐Ÿ’ง',
                'wind': '๐Ÿ’จ',
                'earth': '๐ŸŒ',
                'light': 'โœจ',
                'dark': '๐ŸŒ™'
            }
            
            korean_map = {
                'fire': '๋ถˆ',
                'water': '๋ฌผ',
                'wind': '๋ฐ”๋žŒ',
                'earth': '๋Œ€์ง€',
                'light': '๋น›',
                'dark': '์–ด๋‘ '
            }

            counts_str = ''
            # ๋‹จ์ผ ์—ด๋กœ ํ‘œ์‹œ
            for type, count in results.items():
                st.metric(
                    label=f"{emoji_map.get(type, '')} {korean_map.get(type, type)}", 
                    value=f"{count}๊ฐœ"
                )

            # cols = [col1, col2, col3]
            # counts_str = ''
            # for idx, (type, count) in enumerate(results.items()):
            #     col = cols[idx % 3]
            #     with col:
            #         st.metric(
            #             label=f"{emoji_map.get(type, '')} {korean_map.get(type, type)}", 
            #             value=f"{count}๊ฐœ"
            #         )
                counts_str += f"{count % 10 if count >= 10 else count}/"
            st.markdown(f'{"/".join(korean_map.values())}')
            st.markdown(f'{counts_str}')
        except Exception as e:
            st.error(f"๋ถ„์„ ์ค‘ ์˜ค๋ฅ˜ ๋ฐœ์ƒ: {e}")
            st.info("EasyOCR ๋ชจ๋ธ ๋กœ๋”ฉ์— ์‹คํŒจํ–ˆ์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. Streamlit Cloud์˜ ๋ฉ”๋ชจ๋ฆฌ ์ œํ•œ ๋•Œ๋ฌธ์ผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.")


# Streamlit UI
st.set_page_config(page_title="์†์„ฑ ์ž ์žฌ๋ ฅ ์ฃผ๋ฌธ์„œ ์นด์šดํ„ฐ", page_icon="๐ŸŽฎ")

st.title('๐ŸŽฎ ์†์„ฑ ์ž ์žฌ๋ ฅ ์ฃผ๋ฌธ์„œ ์นด์šดํ„ฐ')
st.write('ํš๋“ํ•œ ์†์„ฑ ์ž ์žฌ๋ ฅ ์ฃผ๋ฌธ์„œ ๊ฐœ์ˆ˜๋ฅผ ์ž๋™์œผ๋กœ ์„ธ์–ด๋“œ๋ฆฝ๋‹ˆ๋‹ค!')

templates = load_number_templates()

tab1, tab2 = st.tabs(["๐Ÿ“ ํŒŒ์ผ ์—…๋กœ๋“œ", "๐Ÿ“‹ ๋ถ™์—ฌ๋„ฃ๊ธฐ"])

with tab1:
    st.write('์ด๋ฏธ์ง€ ํŒŒ์ผ์„ ์—…๋กœ๋“œํ•˜์„ธ์š”')
    uploaded_file = st.file_uploader("์Šคํฌ๋ฆฐ์ƒท์„ ์—…๋กœ๋“œํ•˜์„ธ์š”", type=['png', 'jpg', 'jpeg'])

    if uploaded_file is not None:
        file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
        img = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
        process_image(img, templates)

with tab2:
    st.write('์ด๋ฏธ์ง€๋ฅผ ๋ถ™์—ฌ๋„ฃ์œผ์„ธ์š”')
    paste_result = pbutton(
        label="๐Ÿ“‹ ์—ฌ๊ธฐ๋ฅผ ํด๋ฆญํ•˜๊ณ  Ctrl+V",
        background_color="#FF4B4B",
        hover_background_color="#FF6B6B",
    )

    if paste_result.image_data is not None:
        pil_image = paste_result.image_data
        # st.markdown(f'{type(pil_image)}')
        img = cv2.cvtColor(np.array(pil_image), cv2.COLOR_RGB2BGR)
        process_image(img, templates)


st.markdown('---')
st.caption('Made by โค๏ธsseong')