Nowosad
commented
8 months ago Hi @spono!
Thanks for opening the issue. In short, you use minarea = 900, which means that superpixels smaller that 900 will be removed (merged with a larger ones, to be more precise). However, in you case, you also specify step = 20, which means that initial centers of superpixels will be 20 cells apart. As the result, majority of created superpixels are smaller than 900 cells and thus the code does not work (see the attached code below). Let me know if you have any suggestions of changes here.
library(terra)
#> terra 1.7.69
library(supercells)
r = rast("CHM_1m.tif")
r
#> class : SpatRaster
#> dimensions : 416, 509, 1 (nrow, ncol, nlyr)
#> resolution : 1, 1 (x, y)
#> extent : 691296, 691805, 5077224, 5077640 (xmin, xmax, ymin, ymax)
#> coord. ref. : WGS 84 / UTM zone 32N (EPSG:32632)
#> source : CHM_1m.tif
#> name : Z
#> min value : 0.001
#> max value : 41.305
plot(r)
vol_slic = supercells::supercells(r, step = 20, compactness = 2, dist_fun = "euclidean",
avg_fun = "median", clean = TRUE, minarea = 900, iter = 10,
verbose = 1)
#> Step: 20
#> Initialization: Completed
#> Iteration: 1/10Iteration: 2/10Iteration: 3/10Iteration: 4/10Iteration: 5/10Iteration: 6/10Iteration: 7/10Iteration: 8/10Iteration: 9/10Iteration: 10/10
#> Cleaning connectivity: Completed
#> Error: I cannot return supercells. This may be due to a large number of missing values in the 'x' object. Try to either trim your data to the non-NA area (e.g., with 'terra::trim()') or increase the number of expected supercells.vol_slic2 = supercells::supercells(r, step = 20, compactness = 2, dist_fun = "euclidean",
avg_fun = "median", clean = FALSE, iter = 10,
verbose = 1)
#> Step: 20
#> Initialization: Completed
#> Iteration: 1/10Iteration: 2/10Iteration: 3/10Iteration: 4/10Iteration: 5/10Iteration: 6/10Iteration: 7/10Iteration: 8/10Iteration: 9/10Iteration: 10/10
r = extract(r, vol_slic2)
superpixels_sizes = sort(table(r$ID))
superpixels_sizes
#>
#> 1 2 296 4 402 12 10 470 353 162 356 227 321 5 59 3 346 11 9 67
#> 96 97 104 108 113 136 137 140 153 165 166 173 177 181 183 186 195 198 211 211
#> 367 101 304 460 41 233 154 370 223 33 211 441 244 489 300 383 6 161 503 328
#> 213 219 222 222 223 224 230 231 235 237 237 239 240 244 245 245 247 250 252 253
#> 36 111 112 301 237 347 354 14 478 173 204 455 116 7 445 400 198 369 379 128
#> 255 255 256 256 258 260 261 262 262 263 263 264 269 272 273 275 277 278 278 279
#> 44 219 405 420 365 69 466 446 450 226 92 392 258 429 397 342 453 350 91 451
#> 280 284 284 286 288 289 289 291 291 294 296 296 297 297 298 299 301 303 305 305
#> 262 62 448 129 216 261 18 305 373 147 136 343 440 278 54 95 362 426 313 272
#> 307 308 308 309 309 310 311 311 311 312 313 313 313 314 315 316 316 316 318 319
#> 17 45 270 181 327 138 323 22 443 93 485 175 210 351 439 349 436 96 201 477
#> 321 321 321 322 322 323 323 324 324 326 326 327 327 328 328 330 330 333 335 335
#> 306 387 182 318 143 303 374 386 239 331 377 72 215 447 77 487 248 366 61 235
#> 336 336 337 337 338 338 338 338 341 341 341 342 343 343 344 344 345 345 347 347
#> 238 260 414 133 196 8 384 412 398 454 23 177 29 106 292 380 207 150 188 407
#> 347 347 347 348 348 349 349 349 350 350 351 352 353 353 353 353 354 355 355 355
#> 87 316 37 287 411 425 482 423 46 197 271 322 452 53 110 148 246 348 480 236
#> 356 357 359 360 360 360 360 361 363 363 364 365 365 366 366 366 366 366 367 368
#> 419 220 360 169 195 461 396 71 256 189 488 43 141 168 212 208 444 51 395 249
#> 368 369 369 371 371 372 373 374 374 375 375 378 378 378 378 379 379 380 380 381
#> 274 372 476 358 190 291 378 49 94 245 462 224 385 491 30 424 15 125 338 413
#> 381 381 381 382 383 383 383 384 384 384 384 385 385 385 386 386 387 388 388 388
#> 449 403 492 28 134 268 364 471 83 113 457 483 166 332 309 421 42 130 100 252
#> 388 390 390 391 391 391 392 392 393 393 393 393 394 394 395 395 396 396 397 397
#> 334 151 228 326 382 21 267 458 63 76 142 203 98 280 363 371 145 266 297 155
#> 397 398 398 398 398 400 400 400 401 401 401 401 402 402 402 402 403 403 403 404
#> 344 107 194 13 137 225 247 312 118 355 417 495 438 102 250 185 265 299 340 473
#> 404 405 405 406 406 406 406 406 407 407 407 407 408 409 410 411 411 411 411 411
#> 119 167 368 279 284 105 391 24 123 264 50 293 320 415 75 178 180 202 221 57
#> 412 412 412 413 413 414 414 415 415 415 416 416 416 417 419 419 419 419 420 421
#> 242 263 68 179 393 469 38 90 222 481 259 275 335 406 16 19 254 285 314 81
#> 421 421 423 423 423 424 425 425 426 426 427 427 427 427 428 428 428 428 429 431
#> 135 253 257 310 375 251 269 103 475 410 114 337 388 205 288 330 325 35 126 214
#> 431 431 431 431 431 432 432 433 433 434 436 436 436 437 437 437 438 441 441 441
#> 117 192 442 186 394 172 191 206 352 64 170 295 357 48 65 122 232 329 336 474
#> 442 442 442 443 443 444 444 444 445 446 447 447 449 450 450 450 451 451 451 451
#> 82 84 499 104 115 55 88 231 317 401 501 73 27 158 183 484 505 146 311 361
#> 452 452 452 454 455 456 456 456 456 456 456 457 459 459 459 460 460 461 461 461
#> 39 381 408 479 193 409 40 290 144 157 213 243 431 496 281 289 124 187 467 60
#> 462 462 462 463 464 464 465 465 466 466 466 466 466 466 468 468 470 470 470 471
#> 390 89 163 422 70 153 286 399 437 97 217 241 121 493 127 184 25 490 240 298
#> 471 472 472 473 475 475 475 475 476 477 478 479 480 481 482 484 486 486 487 487
#> 504 66 108 26 78 230 432 502 463 174 34 132 79 85 282 120 200 32 31 389
#> 488 489 489 490 490 490 491 492 494 495 496 496 497 497 499 502 502 503 504 504
#> 434 52 468 139 152 359 427 465 277 74 140 131 464 494 171 176 273 319 255 56
#> 504 505 506 507 508 509 510 511 512 514 518 519 519 520 521 522 522 522 524 526
#> 433 58 229 435 472 86 276 341 283 333 209 456 149 109 376 156 416 302 459 315
#> 526 527 529 529 529 532 532 532 533 533 536 536 537 539 539 546 550 553 554 557
#> 486 308 165 418 345 294 500 234 80 47 99 164 339 404 307 430 159 506 428 160
#> 557 559 567 570 571 573 576 577 588 592 592 594 594 596 600 604 609 616 626 627
#> 497 199 20 324 498 218
#> 656 667 682 744 757 813
hist(superpixels_sizes)
Created on 2025-03-08 with reprex v2.0.2
spono
Hi, first of all: thanks a lot for this great package (and the others) that I'm discovering only now! I think they'll fill a big gap in R.
Running the following lines:
I'm having a bit of troubles with the
Cleaning connectivitystep which seems to fail when theminareaparameter is used.I made few tests getting to this:
1000I get the following error message:Error: I cannot return supercells. This may be due to a large number of missing values in the 'x' object. Try to either trim your data to the non-NA area (e.g., with 'terra::trim()') or increase the number of expected supercells.It seems strange because all no-data values were first convert to 0 before running the function.
with values rangin from 750 to 900 I get:
Error in data.frame(...) : arguments imply differing number of rows: 1, 2while with values below 750 it seems to work fine.
Am I missing something?
Here the file I'm using for testing test_supercells.zip