Browse Forums

help

68 Subscribers

help > Overwrite libtiff in Matlab to run IO toolbox

Showing 1-19 of 19 posts

Overwrite libtiff in Matlab to run IO toolbox

Hi,
I installed the V3D Matlab IO toolbox and compiled the MEX files. However, when I try to run it Matlab complains of two libraries clashing: libtiff from Matlab itself and libtiff I installed in /usr/local/lib

I am using Mac OS 64 bit. I would like to know how to tell Matlab to llok for libtiff in /usr/local/lib instead of th eother one.

Thanks,
Fernando

RE: Overwrite libtiff in Matlab to run IO toolbox

Let's use the Mac/Linux version(s) of Matlab as the example. What you can do is to disable the default libtiff in Matlab, and instead make it point to the good libtiff of the correct version.

Assume you have already installed libtiff under /usr/local/lib.

Then what you can do is to go to /Path-to-matlab/bin/maci (or maci64 or similar folder for a linux system), and list all *libtiff* files. And then rename or remove the original "libtiff.3.dylib" file. Then you can create a symbolic link to the correct one in the /usr/local/lib folder.

Then it should work (w/o need to re-start Matlab).

-hanchuan

--
====================================

Hanchuan Peng, Ph.D.,
Janelia Farm Research Campus
Howard Hughes Medical Institute
19700 Helix Drive,
Ashburn, VA, 20147, USA.
Tel: (+1) 571-209-4156
Fax: (+1) 571-209-4933
Email: pengh@janelia.hhmi.org

* Peng Lab: http://penglab.janelia.org

* V3D lets you explore & understand your
large 3D/4D/5D images a lot easier!
- fast 3D+ image visualization and analysis,
WYSIWYG, and easy plugin interface:
http://penglab.janelia.org/proj/v3d

Originally posted by Fernando Amat:

RE: Overwrite libtiff in Matlab to run IO toolbox

Hello,

I'd like also to use the V3D IO toolbox and have the libtiff problem (windows 7)

I get the message when compiling the mex files :

>> makeosmex_rawfile_io
loadRaw2Stack_c.cpp
loadRaw2Stack_c.cpp(20) : fatal error C1083: Cannot open include file: 'strings.h': No such file or directory

C:\PROGRA~1\MATLAB\R2010B\BIN\MEX.PL: Error: Compile of 'loadRaw2Stack_c.cpp' failed.

??? Error using ==> mex at 208
Unable to complete successfully.

Error in ==> makeosmex_rawfile_io at 8
mex loadRaw2Stack_c.cpp mg_image_lib.cpp mg_utilities.cpp -ltiff

The libtiff version I have is :

>> versionString = Tiff.getVersion()

versionString =

LIBTIFF, Version 3.7.1
Copyright (c) 1988-1996 Sam Leffler
Copyright (c) 1991-1996 Silicon Graphics, Inc.

So I've downloaded the 3.8.2 from sourceforge and installed it. http://sourceforge.net/projects/gnuwin32...

In my matlab/bin/ I have the file libtiff.dll (so no libtiff.3.dylib)
In the libtiff I've just installed I have libtiff3.dll

I've tried to remove the old and replace with the old : no success.
I've tried to remove the old and set the path of matlab to the new one : no success.

Could someone tell me what I should do ? (I have read the explanation but I have no experience with symbolic links on windows, but there is another way with setting the path in matlab)

Thank You
Best Regards
Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

on Windows you can use mingw or some other compilers.

mingw itself is 32bit, we have not tested any 64bit mingw variants so far.

why not simply use Matlab on a Linux or Mac system? since we don't release BrainAligner for Windows, and you could produce the displacement files, it seems you do have such OSs.

-hanchuan

Originally posted by Thomas Templier:

Hello,

I'd like also to use the V3D IO toolbox and have the libtiff problem (windows 7)

I get the message when compiling the mex files :

The libtiff version I have is :

>> versionString = Tiff.getVersion()

versionString =

LIBTIFF, Version 3.7.1
Copyright (c) 1988-1996 Sam Leffler
Copyright (c) 1991-1996 Silicon Graphics, Inc.

RE: Overwrite libtiff in Matlab to run IO toolbox

I have only brainaligner on my Fedora machine and I don't have network connection on it. I would like to do as less as possible on this isolated computer.

So I think I will use ImageJ and write in python to deal with the DF. I have a small problem with the header. What I see on the saveStack2Raw_2byte function is :

long nwrite = fwrite(formatkey, 1, lenkey, fid);
nwrite = fwrite(&endianCodeMachine, 1, 1, fid);
nwrite=fwrite(&dcode, 2, 1, fid);
nwrite = fwrite(mysz, 2, 4, fid);

nwrite = fwrite(img, unitSize, totalUnit, fid);

So the sizes in the header are : 24, 1, 2, 2, 2, 2, 2. Total : 35.
For the loadRaw2Stack I read : 24, 1, 2, 4, 4, 4, 4 Total 43

On a post you mentionned 43 for the header also.

But I do this in python :

try:
    byte=f.read(24)
    print ("a",byte)

    byte = f.read(1)
    print ("b",byte)

    byte = f.read(2)
    print ("c",byte)

    for i in range(4):
        byte = f.read(4)
        print (i, byte)
finally:
    f.close()

And I get :

('a', 'raw_image_stack_by_hpeng')
('b', 'L')
('c', '\x04\x00')
('0', '\x00\x04\x00\x00')
('1', ' \x03\x00\x00')
('2', '`\x00\x00\x00')
('3', '\x01\x00\x00\x00')

Am I missing something ?

Then to read the sx values in the file, I suppose they are placed line after line, slice after slice. So that for a 5 x 5 x 2 image, we have the following correspondences :

(0,0,0) -> 0
(1,0,0) -> 1
(2,0,0) -> 2
(0,1,0) -> 5
(0,0,1) -> 25
(4,4,1) -> 49

Is it right ?

Thank you
Best Regards

Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

I've opened several DF in V3D and they are all the same. Rather dark at the beginning of the stack and bright at the end, and a jump in the last slices. It seems that there might be a small offset somewhere in the header maybe. What do you think of it ?

You said that it is saved in 32 bit float, it means packed in packets of 32 bytes right ?

Best Regards

Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

the displacement fields for separate direction cannpt be all the same (unless the input data do not have any deformation).

You can tryto add a "-F test1" string in my example program on the web page. You should have three output files. If you load them in V3D, you will see them all different.

-hanchuan

Originally posted by Thomas Templier:

RE: Overwrite libtiff in Matlab to run IO toolbox

Hum, it's annoying. I've tested the images on the website with the option -F test1 and I notice the same things. It outputed these files :

total 128452
-rwxrwxr-x. 1 templier templier 2449948 Aug 10 14:44 brainaligner
-rwxrwxr-x. 1 templier templier     1589 Aug 10 14:45 test1_256.marker
-rwxrwxr-x. 1 templier templier 7098416 Aug 10 14:45 test1_256.tif
-rwxrwxr-x. 1 templier templier 7098416 Aug 10 14:45 test2_256.tif
-rw-rw-r--. 1 templier templier 7077931 Aug 10 15:09 test1target.raw
-rw-rw-r--. 1 templier templier      336 Aug 10 15:10 t21_256.tif_target.csv
-rw-rw-r--. 1 templier templier      333 Aug 10 15:10 t21_256.tif_subject.csv
-rw-rw-r--. 1 templier templier       54 Aug 10 15:10 t21_256.tif_matching_quality.csv
-rw-rw-r--. 1 templier templier 33554475 Aug 10 15:10 test1_999_DF_sx.v3draw
-rw-rw-r--. 1 templier templier 33554475 Aug 10 15:10 test1_999_DF_sy.v3draw
-rw-rw-r--. 1 templier templier 33554475 Aug 10 15:10 test1_999_DF_sz.v3draw
-rw-rw-r--. 1 templier templier 7098416 Aug 10 15:10 t21_256.tif

If I drag&drop the DF files into V3D, it says "The file does not exist!" Do nothing. So I make File>Import>Import general image series with the default options (1,1,1, no pattern and Pack images in Z dimension). It gives me the screenshot on the attached file. They have exactly the same values.

Strange things : in the log, it says that the files are not in tif format, but they are, and it outputs the file test1target.raw. The names of the DF files seem weird also (why 999 ?)

The command I've used is :

./brainaligner -t test1_256.tif -s test2_256.tif -L test1_256.marker -w 10 -o t21_256.tif -F test1

Is it normal or there is something wrong ?
Best Regards and thanks for the help.

Thomas

The log of BrainAligner is :

The current input file has the surfix [tif]
#original slice=108 original rez=1.00000 -> #output slices=108 new rez=1.00000
The current input file has the surfix [tif]
#original slice=108 original rez=1.00000 -> #output slices=108 new rez=1.00000
After preprocessing:
Size of subject_image = [256 256 108 1], size of target_image = [256 256 108 1].

-------------------------------------------------
***BrainAligner [version=0.942] parameters:
Target   file: [test1_256.tif]
Target    rez: [1.0000][1.0000][1.0000]
Ref chanl tgt: [0]
Landmark tgt: [test1_256.marker]
Subject file: [test2_256.tif]
Subject   rez: [1.0000][1.0000][1.0000]
Warp res file: [t21_256.tif]
Warp img size: [512]
Ref chanl sbj: [0]
Landmark sbj: [(null)]
Align G/L type: [Local]
Hierarchical landmark search level: [1]
Landmark matching method: [1]
Displacement field producing method: [3]
Use nearest neighbor interpolation method: [0]
Displacement field file: [test1]
Convert to UINT8 for processing: [No]
Resample to pixel size 0.58um for processing: [No]
Volume change map file: [(null)]
Target copy channel (for attaching mask): [-1]
Subject paste channel (for attaching mask): [-1]
Verbose print: [0]
---------------------------------------------------

********************* target=uint8 subject=uint8
in setInitData(), the target image data normalizer=255.000
The current input file has the surfix [raw]
The data is not with a TIF surfix, -- now this program assumes it is RAW format defined by Hanchuan Peng.
size of [V3DLONG]=[8], [V3DLONG]=[8] [int]=[4], [short int]=[2], [double]=[8], [float]=[4]
in setInitData(), the subject image data normalizer=255.000
Done: data initilization.
Read 65 predefined control points from the file. Thus will not invoke the automatic control point detection.
target coord[0]=[87,    138,    43] around subject [87,    138,    43]:     target [43][138][87] -> subject[49][122][87]
target coord[1]=[2,    148,    31] around subject [2,    148,    31]:     ...... not added due to inconsistency of various matching scores.
target coord[2]=[6,    80,    33] around subject [6,    80,    33]:     ...... not added due to inconsistency of various matching scores.
target coord[3]=[68,    58,    56] around subject [68,    58,    56]:     target [56][58][68] -> subject[66][58][72]
target coord[4]=[121,    203,    67] around subject [121,    203,    67]:     target [67][203][121] -> subject[71][198][111]
target coord[5]=[226,    145,    67] around subject [226,    145,    67]:     ...... not added due to inconsistency of various matching scores.
target coord[6]=[87,    184,    20] around subject [87,    184,    20]:     ...... not added due to inconsistency of various matching scores.
target coord[7]=[151,    96,    72] around subject [151,    96,    72]:     target [72][96][151] -> subject[77][103][159]
target coord[8]=[23,    108,    25] around subject [23,    108,    25]:     ...... not added due to inconsistency of various matching scores.
target coord[9]=[49,    198,    40] around subject [49,    198,    40]:     target [40][198][49] -> subject[40][179][43]
target coord[10]=[121,    145,    26] around subject [121,    145,    26]:     ...... not added due to inconsistency of various matching scores.
target coord[11]=[31,    112,    40] around subject [31,    112,    40]:     ...... not added due to inconsistency of various matching scores.
target coord[12]=[29,    70,    63] around subject [29,    70,    63]:     target [63][70][29] -> subject[69][60][27]
target coord[13]=[86,    230,    58] around subject [86,    230,    58]:     target [58][230][86] -> subject[57][219][85]
target coord[14]=[6,    142,    77] around subject [6,    142,    77]:     target [77][142][6] -> subject[81][136][1]
target coord[15]=[166,    197,    34] around subject [166,    197,    34]:     target [34][197][166] -> subject[35][188][162]
target coord[16]=[78,    85,    34] around subject [78,    85,    34]:     target [34][85][78] -> subject[43][84][84]
target coord[17]=[198,    81,    69] around subject [198,    81,    69]:     target [69][81][198] -> subject[73][79][201]
target coord[18]=[13,    240,    55] around subject [13,    240,    55]:     ...... not added due to inconsistency of various matching scores.
target coord[19]=[89,    170,    44] around subject [89,    170,    44]:     target [44][170][89] -> subject[48][152][88]
target coord[20]=[69,    123,    46] around subject [69,    123,    46]:     target [46][123][69] -> subject[52][108][71]
target coord[21]=[76,    125,    36] around subject [76,    125,    36]:     ...... not added due to inconsistency of various matching scores.
target coord[22]=[31,    116,    39] around subject [31,    116,    39]:     ...... not added due to inconsistency of various matching scores.
target coord[23]=[170,    200,    36] around subject [170,    200,    36]:     ...... not added due to inconsistency of various matching scores.
target coord[24]=[77,    84,    36] around subject [77,    84,    36]:     ...... not added due to inconsistency of various matching scores.
target coord[25]=[74,    82,    74] around subject [74,    82,    74]:     target [74][82][74] -> subject[84][81][77]
target coord[26]=[147,    208,    74] around subject [147,    208,    74]:     target [74][208][147] -> subject[77][204][138]
target coord[27]=[60,    113,    74] around subject [60,    113,    74]:     target [74][113][60] -> subject[83][114][60]
target coord[28]=[66,    63,    53] around subject [66,    63,    53]:     target [53][63][66] -> subject[63][62][71]
target coord[29]=[197,    195,    59] around subject [197,    195,    59]:     target [59][195][197] -> subject[61][188][190]
target coord[30]=[137,    206,    59] around subject [137,    206,    59]:     target [59][206][137] -> subject[64][199][128]
target coord[31]=[172,    202,    59] around subject [172,    202,    59]:     target [59][202][172] -> subject[63][197][168]
target coord[32]=[104,    191,    77] around subject [104,    191,    77]:     ...... not added due to inconsistency of various matching scores.
target coord[33]=[83,    152,    81] around subject [83,    152,    81]:     ...... not added due to inconsistency of various matching scores.
target coord[34]=[76,    169,    81] around subject [76,    169,    81]:     target [81][169][76] -> subject[85][164][67]
target coord[35]=[46,    179,    76] around subject [46,    179,    76]:     ...... not added due to inconsistency of various matching scores.
target coord[36]=[89,    145,    49] around subject [89,    145,    49]:     target [49][145][89] -> subject[55][125][87]
target coord[37]=[73,    146,    49] around subject [73,    146,    49]:     ...... not added due to inconsistency of various matching scores.
target coord[38]=[30,    73,    55] around subject [30,    73,    55]:     ...... not added due to inconsistency of various matching scores.
target coord[39]=[12,    147,    38] around subject [12,    147,    38]:     target [38][147][12] -> subject[36][141][8]
target coord[40]=[48,    197,    39] around subject [48,    197,    39]:     target [39][197][48] -> subject[38][178][41]
target coord[41]=[48,    221,    39] around subject [48,    221,    39]:     target [39][221][48] -> subject[42][208][47]
target coord[42]=[12,    214,    39] around subject [12,    214,    39]:     ...... not added due to inconsistency of various matching scores.
target coord[43]=[34,    235,    39] around subject [34,    235,    39]:     ...... not added due to inconsistency of various matching scores.
target coord[44]=[47,    116,    54] around subject [47,    116,    54]:     target [54][116][47] -> subject[58][105][40]
target coord[45]=[42,    221,    54] around subject [42,    221,    54]:     target [54][221][42] -> subject[54][212][39]
target coord[46]=[11,    148,    54] around subject [11,    148,    54]:     target [54][148][11] -> subject[54][142][7]
target coord[47]=[77,    176,    54] around subject [77,    176,    54]:     ...... not added due to inconsistency of various matching scores.
target coord[48]=[27,    205,    54] around subject [27,    205,    54]:     target [54][205][27] -> subject[53][197][22]
target coord[49]=[84,    238,    54] around subject [84,    238,    54]:     ...... not added due to inconsistency of various matching scores.
target coord[50]=[92,    112,    45] around subject [92,    112,    45]:     target [45][112][92] -> subject[53][111][99]
target coord[51]=[137,    184,    45] around subject [137,    184,    45]:     target [45][184][137] -> subject[50][176][128]
target coord[52]=[121,    189,    45] around subject [121,    189,    45]:     ...... not added due to inconsistency of various matching scores.
target coord[53]=[178,    135,    71] around subject [178,    135,    71]:     ...... not added due to inconsistency of various matching scores.
target coord[54]=[178,    183,    69] around subject [178,    183,    69]:     ...... not added due to inconsistency of various matching scores.
target coord[55]=[81,    156,    26] around subject [81,    156,    26]:     ...... not added due to inconsistency of various matching scores.
target coord[56]=[30,    191,    26] around subject [30,    191,    26]:     ...... not added due to inconsistency of various matching scores.
target coord[57]=[33,    234,    26] around subject [33,    234,    26]:     target [26][234][33] -> subject[17][238][13]
target coord[58]=[13,    179,    26] around subject [13,    179,    26]:     target [26][179][13] -> subject[17][161][2]
target coord[59]=[29,    146,    22] around subject [29,    146,    22]:     ...... not added due to inconsistency of various matching scores.
target coord[60]=[18,    226,    18] around subject [18,    226,    18]:     ...... not added due to inconsistency of various matching scores.
target coord[61]=[9,    185,    12] around subject [9,    185,    12]:     target [12][185][9] -> subject[4][169][3]
target coord[62]=[13,    231,    12] around subject [13,    231,    12]:     target [12][231][13] -> subject[4][228][0]
target coord[63]=[23,    107,    29] around subject [23,    107,    29]:     ...... not added due to inconsistency of various matching scores.
target coord[64]=[71,    124,    29] around subject [71,    124,    29]:     target [29][124][71] -> subject[19][138][76]
(1). normalize the point cloud
(2). estimate the initial affine matrix from initial matching using RANSAC

       0,
    (   0). err_invp_min=5.313598

    (   3). err_invp_min=5.036110
       5,
    (   9). err_invp_min=4.519104
    10, 15, 20, 25,
    ( 29). err_invp_min=4.481841
    30, 35, 40,
    ( 43). err_invp_min=4.087120
    45,
    50, 55,
    ( 56). err_invp_min=3.705978
    60, 65, 70, 75, 80, 85,
    ( 88). err_invp_min=3.688620
    90, 95,
    100, 105, 110, 115, 120, 125, 130, 135, 140, 145,
    150, 155, 160, 165, 170, 175, 180, 185, 190, 195,
    200, 205, 210, 215, 220, 225, 230, 235, 240, 245,
    250, 255, 260, 265, 270, 275, 280, 285, 290, 295,
    300, 305, 310, 315, 320, 325, 330, 335, 340, 345,
    350, 355, 360, 365, 370, 375, 380,
    ( 381). err_invp_min=3.660280
    385, 390, 395,
    400, 405, 410, 415, 420, 425, 430, 435, 440, 445,
    450, 455, 460, 465, 470, 475, 480, 485, 490, 495,
    500, 505, 510, 515, 520, 525, 530, 535, 540, 545,
    550, 555, 560, 565, 570, 575, 580, 585, 590, 595,
    600, 605, 610, 615, 620, 625, 630, 635, 640, 645,
    650, 655, 660, 665, 670, 675, 680, 685, 690, 695,
    700, 705, 710, 715, 720, 725, 730, 735, 740, 745,
    750, 755, 760, 765, 770, 775, 780, 785, 790, 795,
    800, 805, 810, 815, 820, 825, 830, 835, 840, 845,
    850, 855, 860, 865, 870, 875, 880, 885, 890, 895,
    900, 905, 910, 915, 920, 925, 930, 935, 940, 945,
    950, 955, 960, 965, 970, 975, 980, 985, 990, 995,
(3). kill the wrong matches by comparing the distance between A and the corresponding B_invp
(4). de-normalization and get arr_2_invp!
    kill 4 pairs, 32 pairs left!
angle12=12.064 [tedge len=23.622 e1=24.434, e2=46.957] - drop this plane
angle13=8.083 [tedge len=6.164 e1=41.400, e2=38.962] - drop this plane
angle13=6.596 [tedge len=6.164 e1=36.235, e2=31.448] - drop this plane
angle13=11.578 [tedge len=6.164 e1=30.594, e2=29.428] - drop this plane
angle13=1.975 [tedge len=1.732 e1=48.990, e2=49.346] - drop this plane
angle12=3.519 [tedge len=1.732 e1=24.000, e2=23.043] - drop this plane
angle12=2.685 [tedge len=1.732 e1=28.931, e2=27.821] - drop this plane
angle12=3.675 [tedge len=1.732 e1=27.019, e2=27.000] - drop this plane
angle12=0.907 [tedge len=1.732 e1=41.449, e2=43.046] - drop this plane
angle12=160.162 [tedge len=39.774 e1=16.062, e2=24.290] - drop this plane
angle13=18.892 [tedge len=16.062 e1=49.417, e2=48.166] - drop this plane
angle23=14.836 [tedge len=15.748 e1=46.141, e2=34.191] - drop this plane
angle23=17.702 [tedge len=15.748 e1=48.929, e2=41.449] - drop this plane
angle23=16.322 [tedge len=15.748 e1=49.880, e2=40.694] - drop this plane
=============================
cc[1]=13.548    target=[87,138,43] subject=[87,122,49]
cc[2]=6.050    target=[68,58,56] subject=[72,58,66]
cc[3]=24.096    target=[121,203,67] subject=[111,198,71]
cc[4]=46.566    target=[151,96,72] subject=[159,103,77]
cc[5]=31.145    target=[49,198,40] subject=[43,179,40]
cc[6]=43.801    target=[29,70,63] subject=[27,60,69]
cc[7]=9.623    target=[86,230,58] subject=[85,219,57]
cc[8]=29.279    target=[6,142,77] subject=[1,136,81]
cc[9]=39.209    target=[166,197,34] subject=[162,188,35]
cc[10]=0.000    target=[78,85,34] subject=[84,84,43]
cc[11]=76.491    target=[198,81,69] subject=[201,79,73]
cc[12]=13.102    target=[89,170,44] subject=[88,152,48]
cc[13]=15.982    target=[69,123,46] subject=[71,108,52]
cc[14]=40.400    target=[74,82,74] subject=[77,81,84]
cc[15]=31.833    target=[147,208,74] subject=[138,204,77]
cc[16]=0.000    target=[60,113,74] subject=[60,114,83]
cc[17]=7.052    target=[66,63,53] subject=[71,62,63]
cc[18]=48.946    target=[197,195,59] subject=[190,188,61]
cc[19]=25.529    target=[137,206,59] subject=[128,199,64]
cc[20]=42.054    target=[172,202,59] subject=[168,197,63]
cc[21]=0.000    target=[76,169,81] subject=[67,164,85]
cc[22]=50.794    target=[12,147,38] subject=[8,141,36]
cc[23]=30.009    target=[48,197,39] subject=[41,178,38]
cc[24]=51.661    target=[48,221,39] subject=[47,208,42]
cc[25]=20.506    target=[47,116,54] subject=[40,105,58]
cc[26]=50.273    target=[42,221,54] subject=[39,212,54]
cc[27]=21.883    target=[11,148,54] subject=[7,142,54]
cc[28]=44.135    target=[27,205,54] subject=[22,197,53]
cc[29]=8.767    target=[92,112,45] subject=[99,111,53]
cc[30]=31.467    target=[137,184,45] subject=[128,176,50]
cc[31]=71.150    target=[13,179,26] subject=[2,161,17]
cc[32]=90.935    target=[9,185,12] subject=[3,169,4]
Overall mean=31.759     std=22.562
-------------------- Finished detecting the best matching points. Now compute displacement field. -------

Prepare save the matching quality record file [t21_256.tif_matching_quality.csv]
method_inconsistency= 0.5077,    model_violation=31.7590
Now save to matching point file of subject to [t21_256.tif_subject.csv]
Now save to matching point file of target to [t21_256.tif_target.csv]

---------------------------------
n=0     x=[87.000 -> 87.000] y=[138.000 -> 122.000] z=[43.000 -> 49.000]
n=1     x=[68.000 -> 72.000] y=[58.000 -> 58.000] z=[56.000 -> 66.000]
n=2     x=[121.000 -> 111.000] y=[203.000 -> 198.000] z=[67.000 -> 71.000]
n=3     x=[151.000 -> 159.000] y=[96.000 -> 103.000] z=[72.000 -> 77.000]
n=4     x=[49.000 -> 43.000] y=[198.000 -> 179.000] z=[40.000 -> 40.000]
n=5     x=[29.000 -> 27.000] y=[70.000 -> 60.000] z=[63.000 -> 69.000]
n=6     x=[86.000 -> 85.000] y=[230.000 -> 219.000] z=[58.000 -> 57.000]
n=7     x=[6.000 -> 1.000] y=[142.000 -> 136.000] z=[77.000 -> 81.000]
n=8     x=[166.000 -> 162.000] y=[197.000 -> 188.000] z=[34.000 -> 35.000]
n=9     x=[78.000 -> 84.000] y=[85.000 -> 84.000] z=[34.000 -> 43.000]
n=10     x=[198.000 -> 201.000] y=[81.000 -> 79.000] z=[69.000 -> 73.000]
n=11     x=[89.000 -> 88.000] y=[170.000 -> 152.000] z=[44.000 -> 48.000]
n=12     x=[69.000 -> 71.000] y=[123.000 -> 108.000] z=[46.000 -> 52.000]
n=13     x=[74.000 -> 77.000] y=[82.000 -> 81.000] z=[74.000 -> 84.000]
n=14     x=[147.000 -> 138.000] y=[208.000 -> 204.000] z=[74.000 -> 77.000]
n=15     x=[60.000 -> 60.000] y=[113.000 -> 114.000] z=[74.000 -> 83.000]
n=16     x=[66.000 -> 71.000] y=[63.000 -> 62.000] z=[53.000 -> 63.000]
n=17     x=[197.000 -> 190.000] y=[195.000 -> 188.000] z=[59.000 -> 61.000]
n=18     x=[137.000 -> 128.000] y=[206.000 -> 199.000] z=[59.000 -> 64.000]
n=19     x=[172.000 -> 168.000] y=[202.000 -> 197.000] z=[59.000 -> 63.000]
n=20     x=[76.000 -> 67.000] y=[169.000 -> 164.000] z=[81.000 -> 85.000]
n=21     x=[12.000 -> 8.000] y=[147.000 -> 141.000] z=[38.000 -> 36.000]
n=22     x=[48.000 -> 41.000] y=[197.000 -> 178.000] z=[39.000 -> 38.000]
n=23     x=[48.000 -> 47.000] y=[221.000 -> 208.000] z=[39.000 -> 42.000]
n=24     x=[47.000 -> 40.000] y=[116.000 -> 105.000] z=[54.000 -> 58.000]
n=25     x=[42.000 -> 39.000] y=[221.000 -> 212.000] z=[54.000 -> 54.000]
n=26     x=[11.000 -> 7.000] y=[148.000 -> 142.000] z=[54.000 -> 54.000]
n=27     x=[27.000 -> 22.000] y=[205.000 -> 197.000] z=[54.000 -> 53.000]
n=28     x=[92.000 -> 99.000] y=[112.000 -> 111.000] z=[45.000 -> 53.000]
n=29     x=[137.000 -> 128.000] y=[184.000 -> 176.000] z=[45.000 -> 50.000]
n=30     x=[13.000 -> 2.000] y=[179.000 -> 161.000] z=[26.000 -> 17.000]
n=31     x=[9.000 -> 3.000] y=[185.000 -> 169.000] z=[12.000 -> 4.000]

#################################
-------------------- Now compute the distances of pixels to the mapping points. -------

z=0 z=1 z=2 z=3 z=4 z=5 z=6 z=7 z=8 z=9 z=10 z=11 z=12 z=13 z=14 z=15 z=16 z=17 z=18 z=19 z=20 z=21 z=22 z=23 z=24 z=25 z=26 z=27 z=28 z=29 z=30 z=31 z=32
-------------------- Now interpolate the DF using linear method. -------

in ApplyDFToChannel(), the subject image data normalizer=255.000
Finish warping the 0th channel.
Done: warping.
Succeed to warp using the DF to the specified channel(s).
The current input file has the surfix [v3draw]
The data is not with a TIF surfix, -- now this program assumes it is RAW format defined by Hanchuan Peng.
size of [V3DLONG]=[8], [V3DLONG]=[8] [int]=[4], [short int]=[2], [double]=[8], [float]=[4]
Succeed to save to file test1_999_DF_sx.v3draw
The current input file has the surfix [v3draw]
The data is not with a TIF surfix, -- now this program assumes it is RAW format defined by Hanchuan Peng.
size of [V3DLONG]=[8], [V3DLONG]=[8] [int]=[4], [short int]=[2], [double]=[8], [float]=[4]
Succeed to save to file test1_999_DF_sy.v3draw
The current input file has the surfix [v3draw]
The data is not with a TIF surfix, -- now this program assumes it is RAW format defined by Hanchuan Peng.
size of [V3DLONG]=[8], [V3DLONG]=[8] [int]=[4], [short int]=[2], [double]=[8], [float]=[4]
Succeed to save to file test1_999_DF_sz.v3draw
Now save the output/warped image [pointer=0x811d250] datatype_warped=1 sz_warped=[256, 256, 108, 1]
The current input file has the surfix [tif]
The warped image has been saved to the file [t21_256.tif].

Originally posted by Hanchuan Peng:

Attachment: DF.png

RE: Overwrite libtiff in Matlab to run IO toolbox

attached is my screen-shot of the three files, which are quite different...(which is what I believe to be correct) I am not sure what happened at your end...
-hanchuan

Originally posted by Thomas Templier:

Hum, it's annoying. I've tested the images on the website with the option -F test1 and I notice the same things. It outputed these files :

Originally posted by Hanchuan Peng:

Attachment: DF_xyz.jpg

RE: Overwrite libtiff in Matlab to run IO toolbox

Ok, yours look like it should look.

On my V3D down in the white window it is written "Voxel Type : UINT8", and on yours you have "Voxel Type: FLOAT32". Should V3D recognize the voxel type automatically ? I have not seen any option to fix this.
I opened each file separately doing
File>Import>Import general image series > 1 file in the same folder, starting image 1, increment 1, file name contains "", pack images in Z dimension

I'm running BA on Fedora 14.
For V3D I have the same on windows 7 too.

My displacement files have a size of 32.0 MB (33,554,475 bytes) and place on disk : 32.0 MB (33,558,528 bytes). (from windows 7)

I try to read them byte per byte in python and all I can get is the "raw_image_stack_by_hpeng" and "L". Then I tried a lot of combinations trying to guess the byte packets sizes but it was unsuccessfull.

So it seems weird but we should be close also I hope.

Thank you for your help

Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

V3D of course can tell what the datatype is (UINT8, UINT16 or FLOAT32).

what version of V3D you are running?

-hanchuan

Originally posted by Thomas Templier:

RE: Overwrite libtiff in Matlab to run IO toolbox

On W7 : v3d_v2.603_win_msvc_64bit
On Fedora : v3d_redhat_fedora_64bit_v2.603

Best
Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

i will put at the website a new version hopefully soon, so you can try the new one.

-hanchuan

Originally posted by Thomas Templier:

On W7 : v3d_v2.603_win_msvc_64bit
On Fedora : v3d_redhat_fedora_64bit_v2.603

Best
Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

Ok.

If I may bother you again, could you give some precisions for the V3D format ?

The problem should come from V3D here normally, not from BrainAligner, so I could still use these displacement fields. I won't use Matlab for it (I did not manage to make it run with the libtiff library as I'm not too experienced).

So the header should have 43 bytes ? 24 + 1 + 2 + 4 + 4 + 4 + 4
Then each number should have how many bytes ? Assuming they have n bytes, something like this should work :

fread(43)
for ...
fread(n)

Am I doing a stupid mistake ?

Otherwise, maybe you know the RAW fileopener from Fiji, and it could be used with the right parameters ?

Image Type : 32-bit Signed, Unsigned, Real, ARGB or ABGR
Width :
Height:
Offset to first image : 43 ?
Number of imges :
Gap between images :

White is zero : yes/no
Little-endian byte order : yes/no

Thank you

Thomas

Originally posted by Hanchuan Peng:

i will put at the website a new version hopefully soon, so you can try the new one.

-hanchuan

Originally posted by Thomas Templier:

On W7 : v3d_v2.603_win_msvc_64bit
On Fedora : v3d_redhat_fedora_64bit_v2.603

Best
Thomas

ImageJ plugin to load V3D's Raw format

43bytes is correct.

But I woulduse imagej if the data is not 8bit. (maybe you can figure a way, but I did that before and the experience was not very smooth...)

So Yinan Wan in my lab indeed developed a nice ImageJ/Fiji plugin to load V3D's Raw format. You can check out the code at

svn co https://svn.janelia.org/penglab/projects...

There is a simple README file there explaining how to run.

-hanchuan

Originally posted by Thomas Templier:

i will put at the website a new version hopefully soon, so you can try the new one.

-hanchuan

Originally posted by Thomas Templier:

On W7 : v3d_v2.603_win_msvc_64bit
On Fedora : v3d_redhat_fedora_64bit_v2.603

Best
Thomas

RE: ImageJ plugin to load V3D's Raw format

Fine !

It seems to work. So the output of BrainAligner is ok for sure and I can deal with my DFs now.

Thank you very much for the help and the high reactivity.

Thomas

RE: ImageJ plugin to load V3D's Raw format

I have also just updated the V3D's website and release a Linux (redhat/fedora) version, V2.687.

Can you test it with the DF files and let me know if they load correctly now?

This new version is a really just a pre-release, as we are developing many new features of the system at this moment, and just do not have time to build all binaries for all platforms for a formal release. But I think this is powerful enough and will like to have your test-drive result, :-)

-hanchuan

Originally posted by Thomas Templier:

Fine !

It seems to work. So the output of BrainAligner is ok for sure and I can deal with my DFs now.

Thank you very much for the help and the high reactivity.

Thomas

RE: ImageJ plugin to load V3D's Raw format

It's working. Drag & drop is working also.
Great.

Best Regards.

Thomas

RE: Overwrite libtiff in Matlab to run IO toolbox

Hi,

I have tried to make the MEX file in Windows Seven with Matlab 64-bit. It doesn't work. I am using the recommended compilers from Mathworks (Microsoft Visual C++ 2010 Express, Microsoft Windows SDK 7.1).
I got various errors concerning:
tiffio.h (fixed by adding the files from libtiff)
strings.h (fixed by replacing it with string.h)
int, long not recognized...

I tried alternative compilers (MINGW64 or CYGWIN) but they are so complicated to set up that it's not worth the hassle, while my primary concern is just to save a 3D stack in a raw file!

Could you just include the compiled MEX32 and MEX64 files in the zip files? That would save trouble to so many people.
Thanks