Maximum memory in Image J

Hello ZooScan users,

We bought a ZooScan (Hydroptic V1) and have problem to set the proper memory 1640 Mo value in ImageJ on our new computer.
The maximum memory we could set to successfully launch ImageJ is 1638 Mo (!).
Here are the features of our computer:
T3400 375W Intel Core 2 Extreme QX6850
3.00GHz, 8MB, 1333MHz
Windows XP PRO 32 bit
4GB DDR2 667 ECC Dual Channel Memory (4x1GB)
Hard disk 160 Go SATA2 (7200 Tpm)
Additional Hard Disk 500 Go (7200 Tpm) SATA
Graphic Card 256MB

We could scan and zip raw image ( narrow frame, 2400 dpi) without any problem, but we experimented problem to convert image from 16 bit to 8 bit. Is this due to the fact that we could not set the maximum memory in ImageJ up to 1640 Mo?
We are going to try to scan an image at 1200 dpi to avoid the convertion step and go through the process directly (to test, if we could process an image).
We are also going to decrease the RAM to 2 GB as recommended in the ZooScan/ZooProcess manual.

We let you know about our progress... if there any
Elvire & Caroline

Hi again,

Well, we have tried to work with only 2 GB Ram (instead of 4 GB)... but it doesn't solve our problem (we still could not set 1640 Mo Memory in ImageJ).

We have also scanned and processed a sample using the Large Frame at 1200 dpi. The scan and the beginning of image process work well (Image crop, substraction of the background), then when the particle analysis starts, the following message appeared:
Unrecognized command: Convex Hull esmer
I've pressed OK and the process stopped with the following message:
Normal END of PROCESS (Converted) Image (process one) process
But no object was detected....

Are all these bugs due to the Memory problem in imageJ?

Elvire

Dear users,
This message appears only if you are using the "new" shape and structure parameters implemented in version 5.06 for the new projects.
If not done, you have to move all the *.java and *.class from the ImageJ/Macros folder to the ImageJ/plugins folder !
Regards
Marc P.

Thanks Marc, I should have read your manual more carefully

We moved the *.java and *.class files as you recommended and re-do the Process (converted) Image (process one).
That time the Particle process started well, but after the analysis of about one third of the image it stopped and restarted the analysis of particles from the beginning to the one third, and restarted again and again..;
We finally aborted the process. Hereby is the copy of the log obtained. Do you think that the 'Analysis method' in 'Particles Process' is correct?

Elvire & Caro

PID
[Image]
Scanning_date= 20080603_1527
Scanning_area= large
Vuescan_version= 8.4.57
[VueScan]
[Input-Perfection4990-Flatbed]
Media=1
Quality=4
BitsPerPixel=3
MakeGrayFrom=2
PreviewResolution=8
ScanResolution=4
Rotation=3
Mirror=1
[Input]
Options=2
[Crop-Perfection4990-Flatbed]
AutoOffset=0
CropSize=0
XSize=30220
YSize=50156
PreviewArea=0
PreviewXOffset=6738
PreviewYOffset=80
PreviewXSize=30223
PreviewYSize=50157
[Crop]
Options=2
[Filter]
Options=2
[Color-Perfection4990-Flatbed]
ColorBalance=0
WhitePoint=0
Brightness=900
[Color]
PixelColors=1
Options=2
[Output-Perfection4990-Flatbed]
PrintedUnits=0
[Output]
JPEGFile=0
DefaultFolder=C:\\Zooscan_test_3jun08\\Zooscan_scan
RawFile=1
RawFileName=egg2_1=.tif
RawFileType=2
RawCompression=0
Copyright=Zooscan_CNRS_2008
LogFile=0
Options=2
[Prefs]
GraphType=2
ExternalViewer=0
ExternalEditor=0
WindowXOffset=0
WindowYOffset=751
WindowXSize=1600
WindowYSize=399
GuidedMode=0
StartupTip=0
TopPanel=0
AbortMode=0

[LUT]
Image_split= NO
Color_Balance= manual
Min= 336
Max= 28105
Gamma= 1
Order= before
Grey_mode_auto= no
OD_range= 1.8
Ratio= 1.15
Median_level= 24439.7

[Info]
Hardware= Hydroptic v1
Software= vuescan8.4.57
ImageType= Zooscan
Author= Caro

[Sample]
Project= test_3jun08
SampleId= egg2
Scanop= Caro
Ship= 999
Scientificprog= Zooscan
StationId= 999
Date= 20000101-1200
Latitude= 99.99
Longitude= -999.99
Depth= 999
CTDref= CTDfilename
Otherref= Other_reference_name
Townb= 999
Towtype= 999
Nettype= 999
Netmesh= 999
Netsurf= 999
Zmax= 999
Zmin= 999
Vol= 999

[Fraction]
Code= A
FracId= FractionId
Min= 999
Max= 999

[Subsample]
SubPart= 999
SubMethod= motoda
CellPart= 1
Replicates= 1
VolIni= 1
VolPrec= 1
Observation= NaN

[Process]
ConfigFile= process_install_both_config.txt
Version= 5.00_2008/02/05_Picheral_CNRS
Multiple_image_process= NO
Date= 20080604_0908

Image_Process
vis1_file= PROCESS
Date= 20080604_0908
egg2_1
ConfigDataImg= +last+no+created_20080603_1424+100+0+4+2400+0.9+0+150+2+0.025+0.25+2.0+20.0+0+neutral+243+174+1+1+1+pixel+100.0+0.3+100+1.0+0.5+3.993+16.0+8+4.0+1+0.05+0.1+10+both
Image_process= Zooscan_1asep
Version= 5.03_Picheral_CNRS
Image_path= C:\\Zooscan_test_3jun08
Image= egg2_1.tif
Frame= _1
Raw_image_depth= 8 bits
Raw_image_size= 94695296
Raw_Image_BackUp_as_ZIP= NO
Final_Image= 8bits
Corrected_width= 12536
Corrected_heigth= 7552
Grey_Level_Adjustement= YES
Resolution= 1200
ODxminref= 11941
ODyminref= 3678
ODxdimref= 30
ODydimref= 300
OD_measured= 54
Max-min_grey_difference_in_OD= 72
STD= 18.5583
OD_correction= NO
Background_correct_using= 20080603_1444_background_large_manual.tif
Image_treatment= neutral
Right_side= Automatic_supress
Right_side_Limit= 11991.5
Transparent_frame= CLEAR
Greytaux= 0.9
Greylimitright= 101
Right_limit= 11931
Greylimitrleft= 195
Left_limit= 157
Greylimittop= 157
Upper_limit= 118
Greylimitbottom= 164
Bottom_limit= 7459
Vis_image_save= YES

Particules_Process
Particule_Process_Date= 20080604_0909
Version= 5.00_2006/12/29_Picheral_CNRS
ConfigDataP= +last+no+created_20080603_1424+100+0+4+2400+0.9+0+150+2+0.025+0.25+2.0+20.0+0+neutral+243+174+1+1+1+pixel+100.0+0.3+100+1.0+0.5+3.993+16.0+8+4.0+1+0.05+0.1+10+both
Resolution= 1200
Calibration= created_20080603_1424
Threshold= 243
PixSize= 0.0212
MinSize= 0.3
MaxSize= 100
Method= default [ 0.3000 - 100 ]
Measures_unit= pixel
Min_Area= 158
Max_Area= 17530163
Separation_Mask= UNUSED
Particle_Analysis= YES
Analysis_Method= from_thumbnail_images_fractalA_skeleton_Slope_Histcum_Mg5_Complexity_Symetry
Particle Analysis ABORTED
Analysed_Objects= 0
TAG= RECOVER
TAG_Incorporated= 0
Results= SAVED
Date= 20080604_0926

Dear Elvire,
We never got this problem... on any computer we have.
Can you remove all files in the "scan" folder and the "work" folder and restart "convert & process in batch" ? Check that you start Zooprocess ONCE ! Click ONLY once on the "Z" in ImageJ.
Regards
Marc P.

Hello Elvire and Marc,

we have been experiencing the same problems as you, Elvire, and are getting desperate here . I solved the problem with the "convex hull error message" by moving the class and java archives. But we are unable to set the memory to 1640, convert or process images on two of our computers, always get a macro error message.

Note that the problem with the Java Virtual Machine and the memory settings is not new. I first observed it on my laptop in March, immediately after an automatic update of Java/Windows, but did not follow it up immediately (continued processing on our main computer instead). Can anyone assist in finding out what could be the origin, for example how the Java and Windows update could be interfering?

Thank you very much.
Sabine