P-TRAP: Panicle Traits Phenotyping Tool User Manual
Contents
1. P TRAP Panicle Traits Phenotyping Tool User Manual Faroq AL Tam faroq al tam O gmail com Helene Adam adam helene gmail com 3 mathias lorieux ird fr Ant nio dos Anjos antoniodosanjos gmail com Mathias Lorieux Pierre Larmande 2 Pierre Larmande ird fr Alain Ghesqui re alain ghesquiere ird fr Stephane Jouannic stephane jouannic ird fr and Hamid Reza Shahbazkia shahbazkia gmail com l DEEI FCT Universidade do Algarve 8005 139 Faro Portugal 2 IRD UMR DIADE Genome and Development of Rice group 911 avenue agroplois 34000 Montpellier France 3 CIAT Agrobiodiversity and Biotechnology Project Cali Colombia 4 DCT ISMAT Instituto Superior Manuel Teixeira Gomes 8500 508 Portimao Portugal 5 CCMAR CIMAR Laborat rio Associado Universidade do Algarve 8005 139 Faro Portugal contact ptrap ird gmail com Application website http bioinfo mpl ird fr index php option com_content amp view article amp id 67 amp Itemid 78 2013 Contents Contents List of Figures 1 Installation and Main Components oO o JN l OA 44 4 6 256844642842 REA RES RSA RES RS 1 1 System Requirements ooo eee 1 2 Installaon asss ssa eee REDE HRERHERHEEESEE EDS 1 2 1 OU a ee eee ee eee ee 22 Windows 21264466446 ee o ea ee ae eee ER ees 1 3 The Main Window ee 1 3 1 Commands u44266 au eee bee bee e 1 3 2 Project Manager 2 2 ee ee 1 3 3 Workspace 22. The installer window will then appear Figure 1 Once you press Next button the installer will ask you where to put the application and its shortcuts 1 2 2 Windows In Windows operating system locate the file grain windows exe and open it The installer Figure 2 will open click Next button and follow the on screen simple instructions 1 3 The Main Window The main window of P TRAP is shown in Figure 3 This window has three main areas as explained in this figure INTRODUCTION Commands Create projects add source images run tasks customize options and more Welcome to P TRAP Installer Wizard This program will install P TRAP application on your computer Click Next to continue Figure 2 The Windows installation window of P TRAP Navigators part of workspace Navigate explorer and jump between the panels very easily P TRAP 201203012225 Projects x 2 dl 2 e 2 da 2 e 2 Mi 2 e 2 da 2 e 2 a 2 38 da 38 w 38 da 38 38 St 38 e 38 dl 38 e 38 St 38 NNNNNNNNN MN NNNNNNNNN SN WwWWeswnnne Pr PB NNNNN NNN NN SN 1 1 2 2 2 2 3 3 3 3 di 20 2 2 2 3 DSCO9898 ricepr DSC09848_ ricegr DSCO09848_ ricepr DSCO09851_ ricegr DSCO9851 ricepr DSC09854_ ricegr DSC09854_ ricepr DSCO09856_ ricegr DSC09856 ricepr DSC09857_ ricegr DSC09857 ricepr 2 DSC09928 ricegr 2_DSC09928 ricepr 1 DSCO9930 ricegr 1_DSC09930 ricepr 2_DSC09932 ricegr 2 DS
3. Figure 12 Defining the start and end generating circles Figure 13 An error hint generated by the application to allow the user locate the error easily 28 CHAPTER 2 WORKING WITH P TRAP _ _ _ _ Projects e x RicePanicleSet_IRD2010 v images N 2 2 1 2 1 DSCO9841 PG 4 7_2_1_3_2_DSC09921 JPG v results v images 2_2_1_2_1_DSC09841 ricegr da 2 2 1 2 1 DSCO9841 ricepr b G temp Figure 14 A grains file generated from processing a panicle image Once the application finished finding the grains it will notify the user by a message As a result a grain file will be generated as shown in Figure 14 2 8 2 Detecting the Grains in Multiple Images 1 Select the project 2 Click ee button if the grains are clustered Otherwise click ne button A progress bar will appear in the right bottom of the main window to show the percentage of images being processed Once it finished it will show a notification message 2 8 3 Color Segmentaion for grains detection In some cases the main automatic segmentaton method may have problems finding the acurate bounds of the grains For this porpose a manual guided method 1s available In this method the user has to specify the grain and background colors To define these colors e Open the image in the image editor e Right click to open a pop up menu as shown in Figure 8 e select the colors To run in color segmentation mode toggle ON the color segemen
4. cm in real scene how lcm in image Nodes b Min distance 10 O means all white and yellow circles are nodes except the execluded one Execlude None if Calcualation Method Build and Fix old detects the junctions and branches based on the circles neighborhood Increamental new As the user defines the start and end circles it incremenatly simulates the panicle Method Increamental new v Export Import OK Cancel Hely Figure 17 The panicle options dialog 1 Select the project 2 Click aes button 31 When the application finished collecting the results of all images it will notify the user by a message The reports files are shown in Figure 10 2 10 Options P TRAP offers a set of options for the graphical interface and the core process as well This sections describes the processing options The panicle options can be found in Tools gt Options menu then choose Panicle Options Figure 17 These options are explained in Table NA 32 CHAPTER 2 WORKING WITH P TRAP Table 2 2 P TRAP panicle options Option Description Structure Finding Spikelet Length Controls the minimum length of the branch any branch that has length less than this value will be mark as a noise and removed Icm This option defines the ratio between the cm in the image and in th real panicle Nodes Controls the minimum distance between two nodes in the main axis in the pa
5. enabled when One of the editors opened pp Open existing project Create new project Figure 4 The toolbar of the main window 1 3 1 Commands In the commands area the tool bar has several buttons with different functionalities explained in Figure 4 1 3 2 Project Manager In this area Figure 5 all files and folders in the project can be dealt with very easily A context menu right clicked menu Figure 5b can be used to manage the files 1 3 3 Workspace In Workspace the user can view and edit images structure and grains results In this area specialized editors for the images structure and grains are available These editors share the same commands for zooming and moving the scene The ImageEditor Figure 6 allows the user to select the region of interest and crop the image Figure 6 hn the editor the scene is the entire area where the user can work 1 INTRODUCTION 11 Project name p gt The folder that contains the source images p The results folder The intermediary results for the structure and the grains Can be used by the user to store additional files a Project components Open Rice Image Find Structure Find Grains Find Grains Traits Cut Copy Delete Rename Tools Properties b Project components menu Figure 5 P TRAP project folders 2_2_1_3_2_DSC09846_ all gt r Figure 6 The images editor 12 CHAPTER I INSTALL
6. has to define the start and end generating points yellow colored circles This is described in details in Figure 12 During post processing the results the application will show hint if some circles need to be connected Figure 13 2 TASKS 25 O Primary Generating X Delete Figure 12 Circles context menu Once you finished the editing and specifying the start and end generating circles click z to save the changes you have made 2 8 Working with the Grains There are two methods to detect the grains The first one deals with the clustered grains and is able to detect the grains on the branches The other method deals with separated grains This section deals with both cases In the beginning we explain how to find the grains of a single image then the post processing is explained 2 8 1 Detecting the Grains in a Single Image To process a single image in order to find the grains do the following 1 Select the image file and right click to open the pop up menu Chapter 1 Figure 5b 2 Choose Find Grains if the image has clustered grains otherwise choose Find Grains Traits 26 CHAPTER 2 WORKING WITH P TRAP R 2 2 1 2 1 DSCO9841 X a Select the start generating circle right click and then choose Generating from the menu Wi 2 21 2 1 DSCO9841 X c Connect the new circle 2 TASKS 21 d Right click on this new connected circle and choose Generating from the menu
7. of the main window osoo a ee P TRAP project folders ooa aa a The images editor ana cas Rw RA ee ewe ee ee we ee ew The panicle structure editor n o a aa a a The grains editor s sss seme waa REDE DAS DEE Ma The grains editor can be also used for editing separated grains The files that P TRAP uses produces 0 a Working with P TRAP l O ANY Nn Aa A WO WN The New Project dialog Select the project type 040 The New Project dialog Name and location of the project A new project named RicePanicleSet_IRD2010 is created 0202 The Open dialog Add images to the project 2 20000 New images are imported to the project 2 2 2 ee ee View edit image in the Image Editor 20202000 Zoom in out in the scene in the Image Editor 0 Image editormenu 2 iiidid ideri 5 O Oo wo N 10 11 11 12 13 14 14 20 ZZ 10 11 12 12 13 14 15 16 17 18 List of Figures Scale images in a selected project o aoa a a a 22 Structure detection notifications 2 0 0 eee ee ee 23 A structure file generated after the image processed 24 Circles context men s ee soso maa a a a a a 23 Defining the start and end generating circles o o ao a a 21 An error hint generated by the application to allow the user locate the error easily 27 A grains f
8. or 2 Right click on the image a context pop up menu appears choose Open Rice Image Figure 5b 3 The Image Editor will then open the image and let you tweak it Figure 6 4 Hold the to zoom in out the scene Figure 7 Please note that this key mouse combination can be used in all editors in P TRAP to perform the zoom in out task Besides this editor has a popup menu to select the color of the grain and background This is be used in grain detection task later on this chapter 2 TASKS ES FG N 2_2_1_2_1_DSc09841 X all gt z 3 Figure 6 View edit image in the Image Editor en Er 221 21 DSCO9841 aJbJ vJ o Figure 7 Zoom in out in the scene in the Image Editor 21 22 CHAPTER 2 WORKING WITH P TRAP Pick up grain color Pick up background color Figure 8 Image editor menu File Edit Image View Navigate Run Toate Window Help Scale image factor Scale 0 5 e Scale images pasam X Projects Add to Favorites Templates Ens add Option OK fess P images im a oo a Scale images command b Scale images factor dialog Figure 9 Scale images in a selected project 2 5 Crop an Image 1 Select the cropping area on the image by using the movable and resizable frame see Figure 7 2 Click 2 6 Scaling Images This step allows you to scale your images if they have huge or very small sizes otherwise it is not necess
9. 2 Working with P TRAP Introduction 0 0 ee ee kk ek ee Getting Started 2 a Adding Images to P TRAP 0 2 0 00048 10 10 10 13 13 15 4 CONTENTS 2 3 Working with Images 2 2 ee ee 20 2 4 View the Source Images o oo aa 00 0002 2 eee 20 2 5 Crop an Image aoaaa 22 2 6 Scaling IMAGES s e sa ew Ee ee eee 2 22 Zi Working with the Panicle Structure 23 2 74 The Structure of a Single Image 23 2 4 2 The Structure of Multiple Images 23 2 7 3 View and Edit the Structure 24 2 8 Working with the Grains 0 000022 eee eee 25 2 8 1 Detecting the Grains in a Single Image 25 2 8 2 Detecting the Grains in Multiple Images 28 2 8 3 Color Segmentaion for grains detection 28 2 8 4 View Edit the Grains Files 28 2 9 Collecting the Reports 0 2 0 0 0000 eee eee 29 PA OPO serah re Rre ERa DENSA NDA DADA EGE EBS 31 2 10 1 Defining the Scale ooa aa a 32 2 10 2 Image s Background o oo a 32 3 Troubleshooting ooo oaa a 33 A XML Files 35 1 Structure XML File 0 2020 2 00 a 35 List of Figures Installation and Main Components l O WO N A a A Q N pd The Linux installation window of P TRAP oaaao aa a The Windows installation window of P TRAP aoaaa a The main window of P TRAP aoaaa eee ee The toolbar
10. ATION AND MAIN COMPONENTS a Structure editor 00 Fy O End Tertiary Fit to view Q 6 Seconday Save as image O O Primar Measure ey Remove flying circles e Generating Delete all X Delete b Structure editor menu c Structure compo nents menu Figure 7 The panicle structure editor In order to crop the image first select the area where the panicle 1s then click save 5j button By using the handles of the cropping frame the user can easily choose the exact area to be crop out This editor is used for both editing the grains detected on the branches of the panicle or the grains used for detecting grains traits Figure 9 2 FILES 13 amp 2 213 2 DSCO9846 a Grain editor Add grains Remove all zero particels Correct all Fit to view ne a Save as image O O Primary Delete all x Delete b Grain editor menu c Grains menu Figure 8 The grains editor 2 Files In this section the files that the application works with and generates are described These files are described and shown in Figure 10 2 1 Images The application works on most of the well known image formates It supports both the colored and grayscale images For better performance it is better to scale the images as we will see in 14 CHAPTER I INSTALLATION AND MAIN COMPONENTS goleja Figure 9 The grains editor can be also used for editing separated grains Projects x RicePa
11. C09932 ricepr 2_DSC09934 ricegr 2_DSC09934 ricepr 3_DSC09935 ricegr 3_DSCO09935 ricepr 7_2_1_1_2_DSc09916 ricegr 7_2_1_1_2_DSC09916 ricepr 7 2 2 1 1 DSC09906 ricegr Sel 7 1 1 ANSCNQAQNA ricanr i Projects Manager Manage projects files Workspace Edit and proofread Status Panel and view the results the system s viewers the results Every component is editable Shows the current can be used if applicable to satisfy the users intuitions ongoing tasks Figure 3 The main window of P TRAP 10 CHAPTER I INSTALLATION AND MAIN COMPONENTS yi e N Dark background use this option only if the background of the images is draker than the panicle or the grains Ei Br View panicle s anchors Show the major and minor axes of the grains and their anchor points Collect results Generate reports for panicle s structure and grains Detect grain s traits Detect grain s traits L Express Run Detect the grains and the structure in the selected project Detect structure Find the panicle s structure in the selected project Detect grains Find grains in all images in the selected project Import images This button allows the user to import the images to the current project Copies of the original images are added and the original ones are kept intact Save These buttons used to save the modification the user makes during post processing the results They become
12. ary To scale images do the following 1 Select the project name 2 Go to the Jools menu and select Scale images Figure 9a 3 In the scaling dialog choose the scale factor depending on the scale you want to work with and click OK Figure 9b Usually 1024 x 1024 images are enough for good processing However huge sized images have to be scaled down for better processing especially for the grain detection task Therefore it is strongly recomended to scaledown the images if the grain detection detects only partial parts of the grains 2 TASKS 23 Structure Finding 2 2 1 2 1 DSC09841 la a Processing progress Message Structure Finding Process finished EF b Structure processing notification message Figure 10 Structure detection notifications 2 7 Working with the Panicle Structure This section describes how to find the structure both for a single image and for all images in the project 2 7 1 The Structure of a Single Image To process a single image in order to find the structure do the following 1 Select the image file and right click to open the pop up menu Figure 5b 2 Choose Find Structure During processing the panicle image a progress bar in the status panel of the main window shows the current situation Figure 10a When the processing finished the system will notify you by a message as shown in Figure 10b When the processing finished a structure file will be added to the result
13. ertices gt contains a set of vertices these are converted to cycles as you can see them in the application They describ the junctions and the ends of the panicle structure e lt vertex id java awt Point x 994 y 1282 x 994 y 1282 type End fixed false gt A single vertex Its attributes are id an identifier of the vertex 35 36 APPENDIX A XML FILES x and y the xy coordinates of the vertices type the type of the vertex isfixed true if the vertex has been edited by the user e lt edges gt The edges that connects the graph vertices e lt edge vertexl java awt Point x 1047 y 1053 vertex2 java awt Point x 1149 y 1015 gt an edge that connects two vertices given their identifiers vertex 1 and vertex2
14. f Type All Fies OK Cancel Figure 4 The Open dialog Add images to the project 5 Click Finish A new project will be then created as shown in Figure 3 2 2 Adding Images to P TRAP In this step one or more images will be imported to the project To do that follow these steps 1 Click on the Import Images button in the toolbar 2 In the Open dialog Figure 4 locate your images files and click OK You can choose single or multiple files to add to the project Hold or keys during clicking on the files to select deselect multiple file To select all files press key combination 3 Click OK the image will be added to the images folder in the project Figure 5 4 Click finish 20 CHAPTER 2 WORKING WITH P TRAP Projects x m RicePanicleSet_IRD2010 v images N 2 2 1 2 1 DSCO9841 PG 4 7_2_1_3_2_DSC09921 JPG v results images b temp Figure 5 New images are imported to the project It is important to note at this point P TRAP does not uses the original images Instead it imports a copy of the original ones So the original images are kept intact 2 3 Working with Images This step has some optional tasks if your images do not have huge size we consider more than 1024 x 1024 as a big image and do not have labels or additional hand made marks 2 4 View the Source Images To view an image you can do one of the following 1 Double click on the image file in the project
15. ile generated from processing a panicle image 28 The anchors of the grains 2 ee 29 Grains number correction sooo o 30 The panicle options dialog ooa aaa a 31 Defining the scale in image n a aaa a 33 Chapter 1 Installation and Main Components 1 Introduction P TRAP is an application for analyzing plant panicle images Itis a cross platform application developed in Java and designed on top of Netbeans Platform 7 1 It allows the user to process the images in a project based approach The images to be processed are added to a project and when analyzed the intermediary results are saved in separate files The final results can be collected in master and details files Additionally The panicle structure and the grains are stored in XML files which can be exported to be used by other applications Besides this manual several video tutorials can be found in the application website 1 1 System Requirements 1 System Platform independent 2 Processor Core 2 Due or higher 3 Memory 2 GB or more 8 CHAPTER I INSTALLATION AND MAIN COMPONENTS P TRAP Installer Welcome to P TRAP Installer Wizard This program will install P TRAP application on your computer Click Next to continue Next gt Cancel Figure 1 The Linux installation window of P TRAP 1 2 Installation 1 2 1 Linux Open the terminal go to the folder where the grain linux sh is located and type sh grain linux sh
16. jimages folder Figure 11 2 7 2 The Structure of Multiple Images 1 Select the project 2 Click on XT button 24 CHAPTER 2 WORKING WITH P TRAP Projects X E HE Hae RicePanicleSet IRD2010 G images N 2 2 1 2 1 DSCO984LJPG w 7_2_1_3_2_D5C09971 JPG Ww results vO images dd 2_2_1_2_1_DSC09841 ricepr b temp Figure 11 A structure file generated after the image processed 2 7 3 View and Edit the Structure This 1s a very important task because it allows the user to manipulate the structure To view the structure of a given image do the following 1 Double click on the structure file in the project or 2 Right click on the image a context pop up menu appears choose View edit results file menu item Use the key mouse ctrl mouse wheel combination to zoom in out the scene To move the entire scene hold the down and move the mouse To edit the structure Table 2 1 shows the required key mouse combinations Table 2 1 Structure editing key mouse combinations Functionality Command Add a circle Double click Delete a circle Right click then choose Delete Figure 12 Connect two circles Hold key down then while pointing on one circle drag the mouse Remove a link Click on the link then move one of its ends to an empty place To change the type of a circle Right click then choose the type you want see Figure 12 Due to the variation from one panicle to another the user
17. nicle If the distance between two nodes is less than this value they are counted as a single node Which nodes has to be excluded from counting Calculation Method Min distance Exclude Method Which method that applications has to use in order to de tect the panicle structure Usually the incremental method works better 2 10 1 Defining the Scale The scale label is widely used in many biological work P TRAP allows the user to define the measuring scale in the image The following steps explain how to define the measuring scale 1 Open the image that contains the scale by double clicking on the image 2 Drag from the one end of the scale Figure 18a and hold until you reach the other end then right click Figure 18b 3 The scale dialog appear Figure 18c define how many unite in the scale and click OK The application will then use the ratio between the real panicle size and the panicle size in the image to make accurate calculations 2 10 2 Image s Background In some cases the images backgrounds are darker than the objects Therefore P TRAP offers an option such that the user defines the type of the image s background The default option is white you can change it to dark by toggling the Dark Background button ON 3 TROUBLESHOOTING 33 a Drage on one end of the scale b Righ click once the other end is reached Select scale Pixels 61 Unit Centimeter cm Amount 1 Centimete
18. nicleSet IRD2010 E images po pet ES RO La ne a da o Seis SRS spas espessa 02 2 1 3 1 DSC09845 JPG Ea lt gt Source images added by the user Nou 2221532 D5 CO9846 5PG v G results p Farticles v images Se E CD EEEE Sse sess ae rs amp 22 131 DSCO9845 ricegr The grains files resulted from finding MM 2_2_1_3_1_DSC09845_ ricepr_ the grains of the source images 2 2 1 3 2 DSCO9846 ricegr y The structure files resulted from finding ft 22132 DSCO9846 ricepr the structure of the source images E alttraits csv gt The detailed report file contains all results in details Lj GrainsTraits csv Pp The grains traits report MainTraits csv The panicle main traits report 2 FILES L3 Section 2 6 2 2 Structure and Grains Files The structure and the grains of the panicle are stored in XML files These files can be used by other applications If you are interested to use them in your application please see Appendix A In fact the P TRAP s editors described earlier uses these files to visualize the results and allow the user to visually post process them These files are shown in Figure 10 Chapter 2 Working with P TRAP 1 Introduction This chapter provides a step by step illustration for using P TRAP It is divided to a set of examples In each one a specific task is addressed and explained The work with P TRAP is very simple The idea 1s to create a project tha
19. r cm Cancel OK c Define how many unites in the scale Figure 18 Defining the scale in image 3 Troubleshooting e Not very good detection 1 Try to scale down the images if they are very large or crop the unwanted back ground or artifacts at the borders 2 If the background is dark try to use the dark background option e Unable to save the detected structure Try to remove all flying circles black colors you can use the right click menu to remove them all Also make sure you have determined the start and end generating points e The report is empty Make sure you have defined the start and end generating points in 34 the structure CHAPTER 2 WORKING WITH P TRAP Appendix A XML Files 1 Structure XML File The structure XML file 1s compactly describing the structure of the panicle as a mathematical graph composed of vertices and edges What 1s needed to use this file 1s any XML parser plugged in your application to convert this file into your graph data structure by tackling only the lt graph gt entity and its components The entire structure of the file is described as the following e lt result signature STRUCTUE imagepath 18 2 2 1 2 DSC09873 PG gt The is the main tag which means this is a structure result file identified by the signature STRUCTURE e lt graph gt Describes a graph of vertices and edges this graph represents the panicle struc ture e lt v
20. t contains several images with the same scale and background intensity i e dark or light background Then process the project with options suitable for its images 2 Tasks 2 1 Getting Started In this section the creation of a new project is explained To create a project do the following 1 Open the application 2 Click on the New Project button in the tool bar 3 In the New Project wizard Figure 1 choose Rice Processing Project and click Next but ton 4 Specify a name and where the project will be stored Figure 2 17 18 CHAPTER 2 WORKING WITH P TRAP New Project Steps Choose Project 1 Choose Project Categories Projects Description Sample Rice Processing Project project lt Back Einish Cancel Help Figure 1 The New Project dialog Select the project type New Project Steps Name and Location 1 Choose Project 2 Name and Location Project Name RicePanicleSet_IRD2010 Project Location home faroq Projects Browse Project Folder faroq Projects RicePanicleSet IRD2010 Figure 2 The New Project dialog Name and location of the project 2 TASKS 19 r RicePanicleSet IRD2010 gt images b E results b E temp Figure 3 A new project named RicePanicleSet IRD2010 is created Look In A test images x ft HB 0 2 2 1 2 1 DSCO9841 O 7_2_1_3_2_DSC09921 JPG File Name Files o
21. tion o 2 8 4 View Edit the Grains Files To view a grain file do one of the following 2 TASKS 29 2_2_1_2_1_DSC09841 x Figure 15 The anchors of the grains 1 Double click on the grain file in the project or 2 Right click on the image a context pop up menu appears choose View edit grains file To view the the grains major and minor axes Figure 15 toggle the View Anchors e button ON Editing the grains file includes adding removing and correcting the number of grains To add a grain right click in the place you want to add a grain and then choose add grain To remove a grain or a grains cluster right click on the grain and then choose remove grains For modifying the number of grains double click on the grain then write the correct number of grains Figure 16 Once you finish post processing the grains click the save button to save the changes you have made 2 9 Collecting the Reports As mentioned before P TRAP generates three reports in Comma Separated Values CSV for mat Once the processing and post processing tasks are performed the final reports can be 30 CHAPTER 2 WORKING WITH P TRAP b Edit the number of grains c Corrected number of grains Figure 16 Grains number correction 2 TASKS generated by Options a General Appearance Files Panicle Options Structure Finding Spikelet Length 40 The Length a spike to be removed Measuring
Download Pdf Manuals
Related Search