Download zip file with example files: :Tiepoint Transform
There is a bug in the Visual Studio compiler (still) and the program may not run - you may need the Visual Studio 2008 Redistributable Package (also included in the zip)
There are two versions of the transform: the batch version, where images are processed
The transform is run from the command line with only one parameter - the name of the configuration file. If no parameter/configuration file specified, the executable will assume the configuration file is named config.ini. Use the command prompt to navigate to the directory where the executable is located and type in the name of the executable followed by the configuration file name, for example:
Transform_v0.93_x64_non-export.exe config.ini
The program will read 3 paths from the configuration file:
[FILES] ; List of images with image IDs: img1.jpg 1 ; Default value assumed by program: "imagelist.txt" IMAGE_LIST = imagelist.txt ; List of seed points in x y format: 100 200 SEED_LIST = winrock_5_beam.txt ; Directory where the images are located - this could be either a direct path to the directory, such as "C:/Documents And Settings/images" ; or a folder that is in the same location as the tiepoint executable, such as "images". Note the absence of backslash (/) at the end IMAGES_DIR = images ; Solution file SOLUTION_FILE = AFFINE_SOLUTION.txt ; ERDAS solution file ERDAS_SOLUTION = ERDAS_SOLUTION.txt [SETTINGS] ; Number of CPU cores available for multithreaded support ; Word of warning - make sure you have enough RAM to support multiple images being open CPU_CORES = 2 ; Size of batches in which images are processed as to not use up all the memory - do not change this unless you're experiencing problems BATCH_SIZE = 10 ; If you only need to recompute the affine and have the matches stored on disk ; (from a previous run of the program), then set RECOMPUTE_MATCHES to 0 RECOMPUTE_MATCHES = 1 ; If you only need to recompute the affine & the matches and have the features stored on disk ; (from a previous run of the program), then set RECOMPUTE_FEATURES to 0 RECOMPUTE_FEATURES = 1 ; Number of neighbors around each seed to be used in the computation of the affine ; transformation - less points will result in a less accurate answer, more points ; will result in a better answer - the trade off is that as the number of points ; increases, the chance that the points do not lie on the same plane also increases ; If points don't lie on the same plane, the answer will be less accurate. ; Value of 20 works well. 6 is minimum (program will make sure) NUM_NEIGHBORS = 20 ; Number of iterations for the least squares algorithm for each image. Typically, ; you won't have to changes this. Larger number may provide more accuracy but may take ; a long time NUM_LM_ITERATIONS = 150 ; Tolerance for the least squares approximation of the residuals. Algorithm quits whenever ; this or NUM_LM_ITERATIONS is achieved RESIDUAL_TOLERANCE = 0.0001
Kingston 447.jpg 001 Kingston 448.jpg 002 Kingston 449.jpg 003 Kingston 450.jpg 004 Kingston 451.jpg 005
558 554 940 542 1366 569 768 537 1151 534
0 558 554 0 images\158.tif 0 0 940 542 0 images\158.tif 0 0 1366 569 0 images\158.tif 0 0 768 537 0 images\158.tif 0 0 1151 534 0 images\158.tif 0 1 348.786 544.245 2.06931 images\158.tif 0 1 738.093 528.984 2.0815 images\158.tif 0 1 1167.87 555.561 2.10245 images\158.tif 0 1 562.296 525.771 2.12172 images\158.tif 0 1 951.247 519.57 2.06795 images\158.tif 0 0 766.11 564.724 1.46684 images\159.tif 1 0 1148.72 554.365 1.45945 images\159.tif 1 0 1574.22 582.372 1.45771 images\159.tif 1 0 976.764 548.604 1.45743 images\159.tif 1 0 1360.24 547.096 1.45789 images\159.tif 1 1 558 554 0 images\159.tif 1 1 940 542 0 images\159.tif 1 1 1366 569 0 images\159.tif 1 1 768 537 0 images\159.tif 1 1 1151 534 0 images\159.tif 1 2 357.366 548.797 1.69782 images\159.tif 1 2 738.431 531.113 1.69601 images\159.tif 1 2 1168.99 551.029 1.73057 images\159.tif 1 2 566.26 528.584 1.69835 images\159.tif 1 2 947.677 521.777 1.7119 images\159.tif 1 1 764.09 562.681 1.20763 images\160.tif 2 1 1161.19 556.457 1.20387 images\160.tif 2 1 1604.14 591.153 1.21085 images\160.tif 2 1 983.322 548.571 1.20905 images\160.tif 2 1 1381.25 552.041 1.20444 images\160.tif 2 2 558 554 0 images\160.tif 2 2 940 542 0 images\160.tif 2 2 1366 569 0 images\160.tif 2 2 768 537 0 images\160.tif 2 2 1151 534 0 images\160.tif 2 3 344.303 558.671 1.58824 images\160.tif 2 3 713.044 544.282 1.58429 images\160.tif 2 3 1125.63 571.903 1.58253 images\160.tif 2 3 547.523 539.707 1.60561 images\160.tif 2 3 917.442 533.338 1.56034 images\160.tif 2 2 772.66 551.581 1.30877 images\161.tif 3 2 1155.74 545.038 1.30232 images\161.tif 3 2 1578.62 578.174 1.29186 images\161.tif 3 2 983.265 537.495 1.298 images\161.tif 3 2 1366.13 540.175 1.29371 images\161.tif 3 3 558 554 0 images\161.tif 3