Yeast Colony Counter User Manual
Contents
1. 2 Installation of the tool The tool comes in the form of a self extracting archive which contains the tool self ColonyCounter exe and the Matlab Compiler Runtime Installer MCRInstaller exe which is necessary to run the tool When running the ColonyCounter_pkg archive the files are extracted into the directory of archive and the MCRInstaller is automatically launched It asks to choose the setup language choose the default En glish and for permission to install the VCREDIST X86 Visual C Redistributable choose Install The installation continues automatically until the MCR InstallShield Wizard asks for several installation options fill in the User information or just click on Next select installation directory and then launch the installation The InstallShield may ask about installation of the NET Framework just click OK to continue When the installation process is finished the ColonyCounter exe program is unpacked to the directory This is the tool self other files can be safely removed 3 Launching the tool To launch the tool simply run the ColonyCounter executable e g with Start Run ColonyCounter The start of the program takes some 15 s on our system 4 Home screen After launching the tool the home screen appears as shown in Figure 1 It is used to select the following Colony Counter Colony Counter Image directory Result file Mode Supervised Batch2. pk research matlabfns Loy and Zelinsky 2003 Loy G and Zelinsky A 2003 Fast radial symmetry for detecting points of interest IEEE Transactions on Pattern Analysis and Machine Intelligence 8 25 959 973 Schier and Kov 2011a Schier J and Kov B 2011a Automated counting of yeast colonies using the fast radial transform algorithm In Pellegrini M Fred A Filipe J and Gamboa H editors BIOIN FORMATICS 2011 Proceedings of the International Conference on Bioinformatics Models Methods and Algorithms pages 22 27 Rome Italy INSTICC SciTePress Schier and Kov 2011b Schier J and Kov B 2011b A semi automatic software tool for batch processing of yeast colony images In Zhang J J editor Proceedings of the IASTED International Conference Signal Processing Pattern Recognition and Applications SPPRA 2011 volume 721SP PRA pages 206 212 Innsbruck Austria IASTED ACTA Press
3. Build 22 02 2011 About start processing Figure 1 Home screen of the tool e Image directory directory with the images to process directory tree is possible as well e Result directory location and file name for the file containing the results of processing e Mode either Supervised with graphical editor for correction of the colony detection results or Batch fully automatic e Start processing is used to start the processing after making the necessary selections 4 1 First start of the tool On the first start of the tool the Result directory and Start processing items are disabled The user is forced to make the selection of the image directory after which the Result directory item will be enabled Finally after having selected the result file the Start processing item will also get enabled Note the values of the image directory and result file selections are stored in the Co CounterConfig txt file between the runs of the tool 4 2 Subsequent run of the tool On the subsequent runs of the tool the values of the image directory and result file from the last run are re loaded The Start processing item is now enabled on the tool launch If the image directory is changes warning about the change is issued reminding the user to check the location of the result file see Figure 2 and the Start processing button gets disabled Image directory changed Image directory changed check the location of the r
4. Yeast Colony Counter User Manual Jan Schier Bohumil Kov schierAutia cas cz 420 2 6605 2470 Department of Image Processing http zol utia cz Contents 1 Introduction ti Amage characteristics 244 48 624624412 v s ey ae ab Blu de ol id bo de te yd be 2 Installation of the tool 3 Launching the tool 4 Home screen At cRIFSEStALbOriING TOO 3 4 30 WG 4 a oer Ao ed ee ee Be eo ice Moe i 4 2 Subsequent run of the tool 2 767 8 we ew a a oe we LD ae Bea ee ee ewe A 4 3 Selection of the result file 2h X senna ik Se Gh 68 l Be A ek Be es ele Se Boy 5 Processing the dishes in the supervised mode 6 Result file filename and format 7 Processing flow 8 Image checking 9 A note on the tool performance 10 Acknowledgements 2010 2011 UTIA AV CR v v i All disclosure and or reproduction rights reserved Revize Revize Datum Autor Popis zm n v dokumentu 1 10 201 770 1 1102010 Vytvo en dokumentu 23 02 2011 Aktualizace pro SW verzi 22 02 2011 1 Introduction This document describes the use of the yeast colony counter developed in UTIA AV CR Prague It can be used to process series of images of Petri dishes containing yeast colonies It is aimed for a research laboratory using standard imaging eguipment darkroom general purpose camera on a stand and hand manipulation of the dishes rather than specialized single purpose camera system It is assumed that dark background is used in t
5. e background samples of background from the image corners are used To determine threshold for the blobs extraction Otsu method is used Next step is resolution of the clustered blobs the Fast Radial Transform Loy and Zelinsky 2003 is used to construct symmetry map of the objects in the image Then the nonmaxsuppts function Kovesi 2005 is used to find local maxima of this map which represent the centers of the colonies Finally the position of the centers if checked to filter out cases which would be obviously wrong e g colony center falling in the background Input Set of images Preprocessing Thresholding Dish locating Image segmentation Resolution of clustered colonies Blob center location map Blob center determination Figure 6 Processing flow of the colony counter 8 Image checking In the first step image is also checked for correctness to ensure sufficient robustness of the processing This check involves mainly the position and size of the dish it should be roughly centered and the inner part that is the part with the grow medium must be completely present in the image so that the coverage is evaluated correctly The image of the dish must be also sufficiently large to provide reasonable accuracy of detection Currently the setting is that the dish must occupy at least 50 of the image height The background distortions such as that shown in Figure 7 are eliminated in the process of dish loca
6. esult file Figure 2 Warning message after directory change 4 3 Selection of the result file The default mode which is used if the user does not make any selection on a subsequent run of the tool neither image directory nor result file is to continue the processing of the given image directory and to append to the result file The tool goes through the result file checks the results against the files in the directory tree deletes the data for the images that are not present any more and continues with processing at the point where is has been interrupted on the last run lf the user decides to open the Result file selection window and selects the same file he gets a series of warning dialogues e First warning about overwriting the file comes from MSWindows and could not be eliminated Nev ertheless the user will have chance to choose to append or overwrite the results file in the following dialogue windows so it is safe to choose to overwrite at this point e The warning that is issued next is shown in Figure 3 Here user is offered the possibility either to Append default or to Overwrite the result file e If he selects to overwrite the file he is warned once more as shown in Figure 4 with the default selection set to No that is do not overwrite the results Selecting Yes in this dialogue will result in overwriting the results 5 Processing the dishes in the supervised mode In the supervised mode the results of
7. he images The tool evaluates two figures for the images the relative coverage of the dish and the number of colonies in the dish It provides two modes of operation automatic or semi automatic In the automatic mode it processes the batch of images without any user intervention In the semi automatic mode it first performs automated thresholding dish localization and a counting attempt then it provides the user with an editor where he can interactively correct the result of counting Two image formats are accepted at this moment jpg and tiff Other formats may be added on request Note The tool is citeware If you find the application useful for your purpose we will be very thankful if you can cite our paper Schier J Kovar B Automated counting of yeast colonies using the Fast Ra dial Transform algorithm BIOSTEC 2011 4th International Joint Conference on Biomedical Engineering Systems and Technologie Rome Italy 2011 see also http www utia cz node 631 0355696 1 1 Image characteristics The images under consideration are characterized by the following properties typical resolution of an image is 1024x 768 pixels dark background is used to increase image contrast illumination by two linear lights along the short edges colonies are roughly round shaped variations of dish position varying diameter of the dish in image varying intensity of the dish background colonies often form clusters
8. little sense 6 Result file filename and format The file with results is stored in the CSV Comma Separated Values format which can be loaded into MS Excel OpenOffice Calc and probably other spreadsheet programs as well The values are separated by semicolon The default name of this file as suggested by the tool is the name of the root directory of the image tree with the results suffix e g images_results csv However any filename can be selected by user A line in the file has the following structure 1 path to the image from the root of the image directory tree e g images dense 1 jpg Note this is the same value as displayed in the title bar of the image editor Section 5 2 relative coverage of the dish in percents 3 number of colonies in the dish The result file is updated immediately after processing each dish so it is save to interrupt the process ing 7 Processing flow Processing flow of the tool is outlined in Figure 6 First step of the processing includes thresholding of the image and localization of the dish let us recall that the dishes are hand manipulated Sec 1 hence the position if the dish in each image is Slightly different In this step the image is also checked for correctness The thresholding is used to separate the dish from the background and to perform image segmenta tion that is to extract the clusters of blobs To determine the threshold for separation of the dish from th
9. processing the detected colonies are presented in an editor window as shown in Figure 5 This window has several function and information areas Going to overite the result file Are You sure Figure 4 Overwrite conformation dialogue a a o x Enen PDP I o CO Relative area 23 92 Number of colonies 56 Stop Save image Figure 5 Dish editor window Window title path and filename of the image Top area results relative area and number of colonies Central area image of the dish with colonies with marks on detected colonies Bottom part Buttons Delete marks Add marks and Zoom used to correct the counting result and the button Next dish The meaning of Delete and Add is obvious Zoom button switches to zoom mode which can be used to zoom in an area of interest in the dish Both click hold drag and scroll wheel can be used Next dish is used to finish processing of the given dish and to proceed to the next one Right column Stop button can be used to interrupt processing Save image can be used to save image of the dish with detection signs for archival purposes Note the filename of the saved images is by default given an appendix saved The user can choose any filename and location however the images with this suffix if present in the image directory are ignored by the tool when it seeks for the images to process obviously processing screenshots would give
10. tion and background masking as long as the distortion does not overlap with the dish in vertical projection Figure 7 Example of background error 9 A note on the tool performance The performance of the tool has been tested using 245 images of Petri dishes containing mostly less than 60 colonies per dish The performance is summarized in Table 1 More details are given in missed false 0 17 21 1 47 0 18 20 19 2 99 0 21 23 25 3 31 0 24 26 29 4 00 0 2 29 2 3 28 0 30 40 35 4 17 0 41 49 30 4 00 0 50 60 29 3 87 2 00 gt 60 30 5 36 3 22 Table 1 Dependence of the counting errors on the number of colonies in the dish The system average counting error is under 4 ISchier and Kov 2011a and Schier and Kov 2011bl 10 Acknowledgements The images included in the package have been prepared in the Department of Genetics and Microbiology Faculty of Natural Sciences Charles University The software has been developed in cooperation with this laboratory The development of the tool has been supported from the Ministry of Education Youth and Sports project 1M0567 and from the project TA01010931 of the Technology Agency of the Czech Republic References Kovesi 2005 Kovesi P D 2005 MATLAB and Octave functions for computer vision and image pro cessing School of Computer Science amp Software Engineering The University of Western Australia Available from http www csse uwa edu au
Download Pdf Manuals
Related Search