2009/3/30 Thijs Marinussen <[email protected]>: > >>> I've managed to compile the project >>> >> >> on what system? >> > I'm using windows with msvc 2008 express to compile. >>> I read the other thread with some ideas for patches, and implemented >>> one of them, now where should I sent it? >>> >> >> email it here as an attachment, please. >> > Attached a fix for bug 2166837 "nona opens unused files for no apparent > reason". Since this is very small I'll make another one to demonstrate > my skills better. > > Thijs > > > > > > Index: hugin_base/nona/Stitcher.cpp > =================================================================== > --- hugin_base/nona/Stitcher.cpp (revision 3767) > +++ hugin_base/nona/Stitcher.cpp (working copy) > @@ -156,7 +156,8 @@ > //bool color = (bands == 3 || bands == 4 && extraBands == 1); > > // check if all other images have the same type > - for (imgNr = 1 ; imgNr < pano.getNrOfImages(); imgNr++) { > + for (UIntSet::const_iterator it = usedImgs.begin(); it != > usedImgs.end(); ++it) { > + imgNr = *it; > vigra::ImageImportInfo > info2(pano.getImage(imgNr).getFilename().c_str()); > if ( pixelType != info2.getPixelType() ) { > UTILS_THROW(std::runtime_error, "image " << > pano.getImage(imgNr).getFilename() > >
Hi, First I'd like to thank you for the patch. But I've some notes about it(as I've been looking at this bug some time ago). Unfortunately there is one culprit – removing this loop also removes check if the images are the same pixel type (nona seems not capable of remapping images of different pixel type, at least it didn't worked for me correctly (and even if it worked, there could probably be problem with enblend (I'll test this later))). The cleaner way how to fix this problem would be to run this loop only when the other images are present. And the cleanest way how to fix it would be fixing nona for better support of images with different pixeltypes if it really is necessary (I mean it could be only my problem that images were not remapped as I' expected). 2009/3/30 Tim Nugent <[email protected]>: > Hi everyone, > > Some of you may know that I worked on Hugin/Panotools for last year's GSoC > and developed the cloud detection code library 'Celeste': > > http://ultrawide.wordpress.com/2008/11/12/using-celeste/ > > Some background: > > -I code on Ubuntu and Centos Linux (and run XP and Fedora under Vmware). I > also have access to a large Linux cluster here at University College London > where I'm in the 3rd year of my PhD in Bioinformatics. > -I'm a keen photographer: below water I shoot with a Canon G7 in Ikelite > housing with Inon wide angle and macros lenses; above water I use a Nikon > D90 with Tokina 10-17mm fisheye. > -As stated above, I worked on Hugin last year; I also write Perl code for > the BioPerl project. > > I have checked with Yuv and Leslie Hawthorwn and have been given the OK to > reapply this year. The two projects I'm considering are outlined below. > Please let me have your comments and suggestions. > > Best wishes, > > Tim > > > Straight-line detection for automated lens calibration > ====================================================== > > -Either create a new tab or modify CP editor tab. > -Semi-automatic: Draw box around line using wxWidgets. > -Fully-automatic: use whole image rather than providing selection. > -Use x,y coordinates of selection to take subsection of image using VIRGA. > -Smooth subsection using Gaussian convulation to highlight regions of the > image with high first spatial derivatives [1]. > -Run Canny edge detector [2] on resulting subsection. > -The Canny detector is regarded as state-of-the-art [3] and is also a VIGRA > function [4] therefore no external dependencies. > -For fine tuning, scale and threhold paramaters can be user adjusted. > -Threshold resulting image until only 1 edge remains. > -This is potentially difficult and will depend on good (semi-automatic) > selection of the line. > -Mark line on image in GUI. > -User decides whether detection is correct. If it is not, go back and adjust > selection or paramaters. > -Move across or down resulting edge and gather x,y coordinates. > -Add these coordinates as straight line control points. > -Run optimiser and save lens calibration data. > > [1] http://homepages.inf.ed.ac.uk/rbf/HIPR2/canny.htm > [2] http://en.wikipedia.org/wiki/Canny_edge_detector > [3] http://en.wikipedia.org/wiki/Edge_detection > [4] > http://hci.iwr.uni-heidelberg.de/vigra/doc/vigra/group__EdgeDetection.html > > > Bracketing Panorama Model > ========================= > > -I have already written a patch [1] which deals with bracketed images, but > there is a lot of room to extend this. > -Automatic detection of bracket images based on EXIF data. > -Tab for manual selection/grouping of bracketed images. > -Modify panorama model so that controls points are only generated once, then > these CPs are applied to corresponding bracketed images. > -Only first set of images are aligned, then alignemnt is applied to > corresponding bracketed sets. > -Options to stitch each bracketed set separately (submit to PTBatcher) or > merge to HDR. > -Possible integration of pfstools tone map opperators, though this depends > on library availability - development currently in progress [2]. > > [1] http://ultrawide.wordpress.com/2008/11/16/hacking-hugin-part-1/ > [2] > http://groups.google.com/group/pfstools/browse_thread/thread/4794cd064f5fd18e > > Hi Tim, I like the Bracketing Panorama Model most. Anyway, good luck with applications to everyone! Regards Lukáš "stativ" Jirkovský --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "hugin and other free panoramic software" group. A list of frequently asked questions is available at: http://wiki.panotools.org/Hugin_FAQ To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/hugin-ptx -~----------~----~----~----~------~----~------~--~---
