File size: 12,553 Bytes
441a971 14b7025 441a971 d896324 441a971 d896324 441a971 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 |
import gradio as gr
from PIL import Image
import os
from tensorflow import keras
from keras.models import load_model
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import Normalize
from io import BytesIO
# Images path
path_main = 'Data/'
images_file = path_main + 'Scennarios init/Scennarios W'
# Load DL models
modelo_1ap = load_model(path_main + 'Models/modelo_1ap_app.keras')
modelo_2ap = load_model(path_main + 'Models/modelo_2ap_app.keras')
plt.rc('font', family='Times New Roman')
fontsize_t = 15
def coordinates_process(texto):
coordinates = texto.split("), ")
resultado = []
for coord in coordinates:
try:
coord = coord.replace("(", "").replace(")", "")
x, y = map(int, coord.split(","))
# Validate range
if 0 <= x <= 255 and 0 <= y <= 255:
resultado.append((x, y))
else:
return False
except ValueError:
return False
while len(resultado) < 3:
resultado.append((0, 0))
return resultado
# plan images path
def plan_images_list():
return [file_ for file_ in os.listdir(images_file) if file_.endswith((".JPG", ".jpg", ".jpeg", ".png"))]
# Valdate inputs
def validate_input(value):
if value == "" or value is None:
return 0
elif value >= 0 or value <= 2:
return value
# MAIN FUNCTION ****************************************************************
def main_function(plan_name, apch1, apch6, apch11, coord1, coord6, coord11):
image_plan_path = os.path.join(images_file, plan_name)
imagen1 = Image.open(image_plan_path)
# No negative number as input
if not (0 <= apch1 <= 2):
return False
if not (0 <= apch6 <= 2):
return False
if not (0 <= apch11 <= 2):
return False
# Some variables init
deep_count = 0
deep_coverage = []
channels = [1, 6, 11]
num_APs = np.zeros(len(channels), dtype=int)
num_APs[0] = apch1
num_APs[1] = apch6
num_APs[2] = apch11
dimension = 256
aps_chs = np.zeros((dimension, dimension, len(channels)))
# Load plan
numero = plan_name[:1]
plan_in = np.array(Image.open(f"{path_main}Scennarios init/Scennarios B/{numero}.png")) / 255
coords = [coord1, coord6, coord11]
for att, channel in enumerate(channels):
if num_APs[att] > 0:
coordinates = coordinates_process(coords[att])
for x, y in coordinates:
if x != 0 and y != 0:
aps_chs[int(y), int(x), att] = 1
# Coverage process
deep_coverage = []
ap_images = []
layer_indices = []
imagencober = {}
for k in range(len(channels)):
capa = aps_chs[:, :, k]
filas, columnas = np.where(capa == 1)
if len(filas) == 2:
# For 2 AP
deep_count += 1
layer_1 = np.zeros_like(capa)
layer_2 = np.zeros_like(capa)
layer_1[filas[0], columnas[0]] = 1
layer_2[filas[1], columnas[1]] = 1
datos_entrada = np.stack([plan_in, layer_1, layer_2], axis=-1)
prediction = modelo_2ap.predict(datos_entrada[np.newaxis, ...])[0]
elif len(filas) == 1:
# For 1 AP
deep_count += 1
layer_1 = np.zeros_like(capa)
layer_1[filas[0], columnas[0]] = 1
datos_entrada = np.stack([plan_in, layer_1], axis=-1)
prediction = modelo_1ap.predict(datos_entrada[np.newaxis, ...])[0]
else:
# Whitout AP
prediction = np.zeros((dimension,dimension,1))
# print(prediction.shape)
deep_coverage.append(prediction)
prediction_rgb = np.squeeze((Normalize()(prediction)))
ap_images.append(prediction_rgb) # Guardar la imagen de cobertura del AP
if np.all(prediction == 0):
plt.imshow(prediction_rgb)
plt.title('No coverage', fontsize=fontsize_t + 2, family='Times New Roman')
plt.axis("off")
else:
plt.imshow(prediction_rgb, cmap='jet')
cbar = plt.colorbar(ticks=np.linspace(0, 1, num=6),)
cbar.set_label('SINR [dB]', fontsize=fontsize_t, fontname='Times New Roman')
cbar.set_ticklabels(['-3.01', '20.29', '43.60', '66.90', '90.20', '113.51'])
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagencober[k] = Image.open(buf)
# Cell map estimation
layer_indices.append(np.argmax(prediction, axis=0))
# Final coverage
if deep_coverage:
deep_coverage = np.array(deep_coverage)
nor_matrix = np.max(deep_coverage, axis=0)
celdas = np.argmax(deep_coverage, axis=0)
resultado_rgb = np.squeeze((Normalize()(nor_matrix)))
plt.imshow(resultado_rgb, cmap='jet')
cbar = plt.colorbar(ticks=np.linspace(0, 1, num=6))
cbar.set_label('SINR [dB]', fontsize=fontsize_t, fontname='Times New Roman')
cbar.set_ticklabels(['-3.01', '20.29', '43.60', '66.90', '90.20', '113.51'])
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagen3 = Image.open(buf)
if num_APs[0] > 0 and num_APs[1] > 0 and num_APs[2] > 0:
cmap = plt.cm.colors.ListedColormap(['blue', 'red', 'green'])
plt.imshow(celdas, cmap=cmap)
cbar = plt.colorbar()
cbar.set_ticks([0, 1, 2])
cbar.set_ticklabels(['1', '6', '11'])
cbar.set_label('Cell ID', fontsize=fontsize_t, fontname='Times New Roman')
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagen4 = Image.open(buf)
elif num_APs[0] > 0 and num_APs[1] > 0:
cmap = plt.cm.colors.ListedColormap(['blue', 'red'])
plt.imshow(celdas, cmap=cmap)
cbar = plt.colorbar()
cbar.set_ticks([0, 1])
cbar.set_ticklabels(['1', '6'])
cbar.set_label('Cell ID', fontsize=fontsize_t, fontname='Times New Roman')
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagen4 = Image.open(buf)
elif num_APs[0] > 0 and num_APs[2] > 0:
cmap = plt.cm.colors.ListedColormap(['blue', 'red'])
plt.imshow(celdas, cmap=cmap)
cbar = plt.colorbar()
cbar.set_ticks([0, 1])
cbar.set_ticklabels(['1', '11'])
cbar.set_label('Cell ID', fontsize=fontsize_t, fontname='Times New Roman')
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagen4 = Image.open(buf)
elif num_APs[1] > 0 and num_APs[2] > 0:
cmap = plt.cm.colors.ListedColormap(['blue', 'red'])
plt.imshow(celdas, cmap=cmap)
cbar = plt.colorbar()
cbar.set_ticks([0, 1])
cbar.set_ticklabels(['6', '11'])
cbar.set_label('Cell ID', fontsize=fontsize_t, fontname='Times New Roman')
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagen4 = Image.open(buf)
else:
cmap = plt.cm.colors.ListedColormap(['blue'])
plt.imshow(celdas, cmap=cmap)
cbar = plt.colorbar()
cbar.set_ticks([0])
cbar.set_ticklabels(['1'])
cbar.set_label('Cell ID', fontsize=fontsize_t, fontname='Times New Roman')
cbar.ax.tick_params(labelsize=fontsize_t, labelfontfamily = 'Times New Roman')
plt.axis("off")
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
imagen4 = Image.open(buf)
return [imagencober[0], imagencober[1], imagencober[2], imagen3, imagen4]
# plan visualization
def load_plan_vi(mapa_seleccionado):
image_plan_path1 = os.path.join(images_file, mapa_seleccionado)
plan_image = Image.open(image_plan_path1)
plan_n = np.array(plan_image.convert('RGB'))
plt.figure(figsize=(3, 3))
plt.imshow(plan_n)
plt.xticks(np.arange(0, 256, 50))
plt.yticks(np.arange(0, 256, 50))
# Save the plot to a buffer
buf = BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an image
plan_im = Image.open(buf)
return plan_im
with gr.Blocks() as demo:
gr.Markdown("""
## Fast Radio Propagation Prediction in WLANs Using Deep Learning
This app use deep learning models to radio map estimation (RME). RME entails estimating the received RF power based on spatial information maps.
Instructions for use:
- A predefined list of indoor floor plans is available for use.
- A maximum of (255,255) pixels is allowed for image size.
- Negative numbers are not allowed.
- The established format for the coordinates of each access point (AP) must be maintained.
- A maximum of 2 APs are allowed per channel.
""")
with gr.Row():
# Input left panel
with gr.Column(scale=1): # Scale to resize
map_dropdown = gr.Dropdown(choices=plan_images_list(), label="Select indoor plan")
ch1_input = gr.Number(label="APs CH 1")
ch6_input = gr.Number(label="APs CH 6")
ch11_input = gr.Number(label="APs CH 11")
coords_ch1_input = gr.Textbox(label="Coordinate CH 1", placeholder="Format: (x1, y1), (x2, y2)")
coords_ch6_input = gr.Textbox(label="Coordinate CH 6", placeholder="Format: (x1, y1), (x2, y2)")
coords_ch11_input = gr.Textbox(label="Coordinate CH 11", placeholder="Format: (x1, y1), (x2, y2)")
button1 = gr.Button("Load plan")
button2 = gr.Button("Predict coverage")
# Rigth panel
a_images = 320 # Size putput images
with gr.Column(scale=3):
first_image_output = gr.Image(label="plan image", height=a_images, width=a_images)
# with gr.Row():
with gr.Row():
image_ch1 = gr.Image(label="CH 1 coverage", height=a_images, width=a_images)
image_ch6 = gr.Image(label="CH 6 coverage", height=a_images, width=a_images)
image_ch11 = gr.Image(label="CH 11 coverage", height=a_images, width=a_images)
with gr.Row():
image_cover_final = gr.Image(label="Final coverage", height=a_images, width=a_images)
image_cells = gr.Image(label="Cells coverage", height=a_images, width=a_images)
# Buttons
button1.click(load_plan_vi, inputs=[map_dropdown], outputs=first_image_output)
button2.click(
lambda map_dropdown, ch1, ch6, ch11, coords1, coords6, coords11: main_function(
map_dropdown,
validate_input(ch1),
validate_input(ch6),
validate_input(ch11),
coords1,
coords6,
coords11,
),
inputs=[map_dropdown, ch1_input, ch6_input, ch11_input, coords_ch1_input, coords_ch6_input, coords_ch11_input],
outputs=[image_ch1, image_ch6, image_ch11, image_cover_final, image_cells]
)
demo.launch() |