InterImage 1.30 User Guide - Computer Vision Lab - PUC-Rio
Contents
1. Interlmage 1 30 User Guide www lvc ele puc rio br projects interimage Table of Contents 1 Introduchon ss ii E 1 2 BASIC CONCEDES Ad 2 cae na eer rere ter eee eer rere rere 3 ZZ VOD DOW OP CV AVON S A a tienen aa nutes 3 29 BOON UP OPE IO ando ipa a 4 A BE EE E qa E o no E E E T 5 aero Mterpreta ton CONTO Lanna T S 8 3 SV SUC MA Interact 11 3 4 Main WV AG W ac 11 a A o O ad sd 11 SEZ TOODA ss e a a raia ert 14 3 15 Semanne Net Wihdow assada da sd OS 15 ZLA Layers o sive O 16 So Node Editor Window sanada ada dE 21 SELO TO NV CI in tee Dr a A ec tna ees ceeds 22 3 1 7 Object Information WIAdOW da fed debe bia a 24 322 NEW EGIL Project WiIndoW assinadas Eai dai ee 25 3 21 Supported Resource FOrmMats ii A A 26 9 Liz MAR MADE SIZE A A A a A 26 3 23 Resources in Different Resolutions iii A apa nn 27 3 3 Decision RUE WINdOW stents sd Eai a ee 27 3 5 1 BUIIGING BIOCKS nai a Gl ene eee eee 28 Dee WOO DAR ssa aa Edit ia tiene 29 3 5 0 Upper lower Level QUIS ada 30 Sd Decision Tree dba a ee 30 SD SQUIce Code PAD ias saias cada 6 A a 32 5 5 0 Insert Fait Class WindOM asia sa 32 3 9 7 ANSET Edit Selection WiNdoOW apa cia 33 5 5 0 nsert Fdit Expression WING OW assis ia ora dos 33 3 3 9 Insert Edit Membership WIN Wir 34 3 3 10 Membership Function Window csssssssssscsesesesssssssseseesosararsesesesesssssseseesesoraransesesesess 35 3 3 11 Insert EgitA corecanonWindom aaa 37 34 Analysis Explorer VIG2. Figure 3 29 Insert Membership window This window Figure 3 29 allows you to create sets and expressions of fuzzy logic Type Defines the type of Membership block It has the following options Fuzzy set Inserts a block that returns the membership value of an attribute to the selected fuzzy set Operation Inserts a fuzzy operation Min Max Mean Mul Sum Membership value Inserts a block with a membership value defined by the user Operator Defines the fuzzy operator Attribute Defines the fuzzy set input attribute Fuzzy set Defines the fuzzy set M New Creates a new fuzzy set Page 34 User Guide InterImage 1 30 Edit Edits the selected fuzzy set 9 Delete Removes the selected fuzzy set 14 Import Allows importing a fuzzy set from a fuzz file Not implemented Complement Returns the complement of the computed membership value 3 3 10 Membership Function Window O Membership Function Attribute Fuzzy set name Membership function 78 008299 0 13 Number of points 7 Maximum value l 1 00 Minimum value Function parameters 0 00 0 000000 100 000000 No parameters available eS Right border xOffset 0 000000 U Figure 3 30 Membership Function window This window Figure 3 30 allows you to create and edit the membership function of a fuzzy set Page 35 User Guide InterImage 1 30 Attribute Shows the fuzzy set input attribute Fuzzy set
3. Class Defines the node class Color Defines the node color TopDown Decision Rule Edits the node top down decision rule TopDown Multi Class Defines whether the top down operator decision rule associated to the selected node is multi class TopDown Operator Selects the node top down operator Page 21 User Guide InterImage 1 30 Note You may have noticed that Node Editor is divided into three groups of properties Generic BottomUp and TopDown The scope of this manual is limited to the generic properties of the node For information about operators and their parameters visit See also Decision Rule window page 27 3 1 6 Viewer This window Figure 3 17 allows viewing and interacting with the layers added to the viewer Information Shows the values of the selected image pixel If an object is selected shows its ID Geocoordinates Shows the geographic coordinates while moving the mouse over the viewer SS Fit to Window Centers the image in the viewer Q Zoom Zooms in by clicking the left mouse button Zooms out by clicking the right one Allows focusing on a specific part of the image by selecting it with the left button O Pan Moves the image by dragging the mouse O Information Allows showing information about the selected image or object Layer Information If an image layer is selected shows information about the image Otherwise shows information about the layer objects Page 22
4. G Insert Class Class es All Arealirbana Sombra Vegetacao Merge Neighbours ola Cancel Figure 3 26 Insert Class window This window Figure 3 26 allows you to select the objects of a particular class or classes Class es Select one or more using the Ctrl key classes Merge Neighbors Groups connected objects into larger objects Page 32 User Guide InterImage 1 30 3 3 7 Insert Edit Selection Window G Insert Selection Figure 3 27 Insert Selection window Permite x Permite x Table 3 2 Selection criterion This window Figure 3 27 allows you to select objects that meet a certain criterion This criterion is of the form Expression 1 Defines an attribute or expression Operation Defines a logical operator Expression 2 Defines another attribute or expression 3 3 8 Insert Edit Expression Window Insert Expression Build your expression New attribute Expression Figure 3 28 Insert Expression window Page 33 User Guide InterImage 1 30 This window Figure 3 28 allows you to create a new attribute from another attribute or from the result of a mathematical expression New attribute Defines the new attribute name Expression Defines an attribute or expression 3 3 9 Insert Edit Membership Window O Insert Membership Choose a type Fuzzy set Min Attribute Fuzzy set ees 0000 Complement Cancel
5. Page 19 User Guide InterImage 1 30 Edit Edits the selected layer Remove Removes the selected layer Move Up Move the selected layer upwards Move Down Move the selected layer downwards Save Confirm the changes made to the selected layer 6 Cancel Cancel the changes made to the selected layer Export Exports the selected layer Note For layers of type Shape Selection and Result the Export function allows exporting the objects to a shapefile and calculating attributes See also Decision Rule window page 27 3 1 4 6 Layers List Type Keyname Visible 1 Image default 7 Figure 3 15 Layers list This control Figure 3 15 displays the layers in the order they are arranged in the viewer The Visible option lets you define whether the layer is visible Page 20 User Guide InterImage 1 30 3 1 5 Node Editor Window Y Generic BottomUp Decision E BottomUp Operator Dummy Bottomblp Breakpoint Mone Class Cena Color TopDown Decision LJ TopDown Multi Class TopDown Operator Dummy TopDown BottomUp dad MI Image file Reliability Figure 3 16 Node Editor window This window Figure 3 16 allows you to edit the properties of the semantic net nodes BottomUp Decision Rule Edits the node bottom up decision rule BottomUp Operator Selects the node bottom up operator Breakpoint Defines whether the selected semantic node is an interpretation breakpoint
6. Defines the fuzzy set name Type Defines the shape of the membership function Table 3 3 Number of points Defines the number of points used to draw the function 9 11 13 or 15 yOffset Defines the y axis offset Maximum value Maximum membership value Minimum value Minimum membership value Membership Function Allows you to edit the membership function Just drag the vertices with the mouse Left border Lower limit of the function domain Right border Upper limit of the function domain xOffset Defines the x axis offset Function parameters Some functions allow a fine adjustment of parameters such as slope inflexion point mean e standard deviation Greater than RO ren gt a A aa A C Varnes Mee Page 36 User Guide InterImage 1 30 LN Approximate gaussian IA Y Approximate range 1 Complete range Table 3 3 Standard forms of membership functions 3 3 11 Insert Edit Aggregation Window Insert Aggregation Build your aggregation New attribute Type f Average ha Attribute A For parent E Cancel Figure 3 31 Insert Aggregation window This window Figure 3 31 allows you to create a new attribute from another attribute aggregation The new attribute can also be passed to the upper level of the net New attribute Defines the new attribute name Type Defines the aggregation type Average Standard deviation Maximum Sum Division
7. 1 30 O New Project Ctrl N O Open Project Ctri O Edit Project Ctrl E Save Project Ctrl 5 Save Project As Close Project Ctrl Fa Exit No recent projects Figure 3 2 File menu O Save Project Saves the current project Close Project Closes the current project Exit Quits the program Below the Exit option is offered a list of recent projects making it easy to return to a previous project in which you were working on See also New Edit Project window page 25 3 1 1 2 View Menu Semantic Net Layers Node Editor Object Information Analysis Explorer E Shapefile Editor Samples Editor Figure 3 3 View menu The View menu Figure 3 3 provides the following options Semantic Net Displays the Semantic Net window Page 12 User Guide InterImage 1 30 Layers Displays the Layers window Node Editor Displays the Node Editor window Object Information Displays the Object Information window O Analysis Explorer Tool that helps in building the interpretation model a Shapefile Editor Tool for creating and editing shapefiles Samples Editor Tool for segmentation samples collection and manual classification of polygons See also Analysis Explorer window page 38 Shapefile Editor window page 41 Samples Editor window page 42 3 1 1 3 Actions Menu Start Undo Continue 000 Step Figure 3 4 Actions menu The Actions menu Figure 3 4 prov
8. Count Attribute Defines the attribute to be aggregated For parent Defines if the new attribute will be passed to the upper level of the net Page 37 User Guide InterImage 1 30 3 4 Analysis Explorer Window O AnalysisExplorer Sombra Sm Join Class Selection Membership Aggregation Classify General O Element 0000 09000 Control Panel Y Input Layer UpperJevel Rule basis o O General V Border gt Merge All Input Class E C Merge Connected Somi 9 No Merge Y Output Layer gt 3 Background Image Y Bord Apply default v in J Figure 3 32 Analysis Explorer window This window Figure 3 32 offers tools that help in building the interpretation model Itis a variation of the Decision Rule window Section 3 3 therefore it has an interface that allows creating a decision rule However it adds a Control Panel that has tools that allow us to analyze the characteristics of objects and the result of the rule and help thus building the model A new Attributes tab is also added in the upper right control It allows you to inspect the properties of the selected node in the viewer Page 38 User Guide InterImage 1 30 3 4 1 Control Panel Control Panel Ee Input Layer Views O O O General O 4 Border Input Class a Sombra Output Layer ps 00 Background Image Apply E AO Border default 7 E Figure 3 33 Control Panel V
9. OW sab ai da EO ada 38 dl Control PAN CU asa nas 39 SAZ ANASI TOO S is 40 So Shape nle Editor WIM OW assassino O 41 SO Samples Editor WINdOW a das 42 References iu da 46 User Guide InterImage 1 30 1 Introduction Interlmage is an open source software development initiative that is part of an international scientific cooperation project led by the Computer Vision Laboratory of the Department of Electrical Engineering of the Catholic University of Rio de Janeiro PUC Rio and by the Image Processing and Remote Sensing divisions of the National Institute for Space Research INPE InterImage is a multi platform system for image automatic interpretation written in C and Qt The system provides support for the integration of external image processing Operators that can be coded in any programming language or even be proprietary programs In its basic package Interlmage offers however a set of operators constructed with the functions and classes provided by TerraLib 1 called TerraAIDA http www dpi inpe br terraaida InterImage is based on the GeoAIDA system 2 developed by the Institute of Technology Information of the University of Hannover 3 Germany and inherited from this system its basic functional characteristics besides knowledge structures and control mechanisms A new graphical user interface knowledge representation functionality and image processing operators were later added to the system Chapter 2 of this manual
10. User Guide InterImage 1 30 Pixel 255 217 0 0 X 324449 017003 Y 7387555 937680 Layers Image default 523 x 445 pixels 4 bands Figure 3 17 Viewer Page 23 User Guide InterImage 1 30 3 1 7 Object Information Window fileoEast fileoMNorth fileoSouth fileoWest geoEast geoNorth geoSouth geoWest id object id p status un ury Value Sombra 000000 C fInterimage testes v 324648 000156 7367589 399887 4367322 399878 324334200146 324519 000152 7387456 199883 7387440 599882 324500 400151 3476 211 247 Sombra 1839 1840 0 200000 cl 307 222 Figure 3 18 Object Information window This window Figure 3 18 allows you to view the properties of the selected object in the viewer Page 24 User Guide InterImage 1 30 3 2 New Edit Project Window Resources List O Geographic Coordinates West Connection Settings F Use Remote Server Host localhost Port Number 19209 Figure 3 19 New Project window This window Figure 3 19 allows you to create or edit an interpretation project It offers the following options Name Defines the project name Folder Defines the project folder Key Defines a resource nickname File defines the resource file Default Image Although allowing multiple images Interlmage requires that one of them is selected as default image Page 25 User Guide InterImage 1 30 West North East South If t
11. in larger objects No Merge No object is changed 3 3 4 Decision Tree Tab This tab Figure 3 24 provides a powerful interface to interactively build and edit the decision tree By clicking on a block with the right mouse button a context menu is displayed Figure 3 25 with the following options Edit Edits the selected block Copy Copies the selected block Cut Cuts the selected block Paste Pastes a block in the position of the selected block Comment Uncomment Comments the selected block and its children so it s ignored in the rule execution Insert Inserts a block in the position of the selected block Page 30 User Guide InterImage 1 30 Decision Tee All Figure 3 24 Decision Tree tab Ctrl C Ctrl X Paste Ctrl V Comment Uncomment Insert Insert Child Delete Figure 3 25 Context menu Insert Child Inserts a child block in the selected block Delete Removes the selected block Page 31 User Guide InterImage 1 30 Tip Dragging and dropping block A over block B with the Left Button moves block A to block B position Right Button turns block A son of block B 3 3 5 Source Code Tab This tab was used in earlier versions to allow advanced users to edit the decision rule directly in its original form in Reverse Polish Notation In this version this is no longer possible and probably this tab will be removed in future versions 3 3 6 Insert Edit Class Window
12. last block of the rule Tip To add a block just click on it and it will be inserted at the end of the rule If the block is not inserted it might not be allowed in this position So you need to drag the block and drop it over other block in the rule respecting this convention Page 28 User Guide InterImage 1 30 Left Button moves block A to block B position Right Button turns block A son of block B 3 3 2 Toolbar O 0000 90060 O Figure 3 22 Toolbar New Erases the current rule and initializes a new one 3 Level Up Moves the block one level backwards Level Down Moves the block one level forward A Move Up Moves the selected block upwards within the same tree level Move Down Moves the selected block downwards within the same tree level 1 Cut Cuts the selected block Copy Copies the selected block Paste Pastes a block in the position of the selected block 9 Delete Removes the selected block P Edit Edits the selected block Page 29 User Guide InterImage 1 30 3 3 3 Upper Lower Level Rule Lower level Rule Merge All Merge Connected No Merge Figure 3 23 Lower level rule This control Figure 3 23 defines how objects will be received by the higher lower level of the net Merge All Join all objects of the same class though disjointed into a single larger object Merge Connected Joins connected objects of the same class
13. polygons See also Analysis Explorer window page 38 Shapefile Editor window page 41 Samples Editor window page 42 3 1 3 Semantic Net Window Semantic Net D x Shadow GD Vegetation Figure 3 7 Semantic Net window This window Figure 3 7 allows you to interactively create and edit a semantic network When clicking on a node with the right mouse button a context menu Figure 3 8 is displayed with the following options Copy Copies the semantic node Cut Cuts the semantic node Paste Pastes a node in the position of the selected node Page 15 User Guide InterImage 1 30 Semantic Net 0 x Ctrl C Ctrl X Ctrl V Enable Disable Ctrl Ll Insert Insert Child d Delete Figure 3 8 Context menu Enable Disable Enables disables the semantic node When disabled the semantic node is ignored during the interpretation Insert Inserts a node in the position of the selected node Insert Child Inserts a child node in the selected node Delete Removes the selected node Tip Arrastar e soltar um no A sobre outro no B com o Bot o esquerdo move o n A para a posi o do no B Bot o direito Torna o n A filho do n B 3 1 4 Layers Window This window Figure 3 9 allows you to edit and add layers to the viewer 3 1 4 1 Image Tab Image Selects one of the project images Page 16 User Guide InterImage 1 30 Keyname Defines the layer nickname O Composition Se
14. Thus the top down operator associated to this node will be responsible for identifying objects in the image not only of the respective class but also of sibling nodes classes In this case the operators associated with sibling nodes will not run The operator associated with the multi class node class needs to be able to identify objects from more than one class or a decision rule must be created for this purpose The system provides a default top down operator called Dummy Topdown This operator will output a single region that is equal to the ROI defined in the parent node of the node to which it is associated This operator allows setting the confidence value of its output and the project image that it will be associated to However it doesn t allow using decision rules due to the characteristic of its processing Note For information on other top down operators and their parameters visit http wiki dpi inpe br doku php id interimage operators documentation See also Node Editor window page 21 2 3 Bottom Up Operators Bottom up operators can also be associated to each node of the semantic net The bottom up operator processes the hypotheses of child nodes of the node to which it is associated generated in top down step It can validate hypotheses and discard or resolve spatial conflicts Page 4 User Guide InterImage 1 30 Bottom up operators are also executable programs called by the system core during the interpreta
15. are disconnected don t spatially intersect The aim of the interpretation process is to create an instance net 7 and a corresponding region net Tp applying the knowledge represented by the semantic net T in the interpretation of the region R Initially the hypothesis net T is created and gradually hypothesis nodes H are replaced by instance nodes so that in the end of the process the T net is complete The top down processing is shown in the center of Figure 2 7 It starts in the figure at the point where the hypothesis node H associated to the region R and to the semantic node S has been created From this point on the control passes recursively to the S nodes Hypothesis nodes H are generated through the execution of a top down operator and a decision rule if there is one associated to the semantic node S for the region R The H mi nodes are associated to the R m regions If a Dummy Top down operator is associated to the node S a simple hypothesis node H will be associated to all the region defined by its parent node H R in this case will be equal to R Page 9 User Guide InterImage 1 30 If S m has child nodes the procedure mentioned in the previous paragraph will be repeated for each H until the semantic leaf nodes are reached At this point begins the bottom up processing to the right Figure 2 7 initiating with the leaf nodes parents The bottom up operator and decision rule ass
16. e 1 30 Samples Editor Samples Collector X 324541 470263 Y 7387704 710868 89 Q OO Background Image default Name dla TopDown Operator TA Baatz Segmenter da TopDown a Input Image imagem al Input Bands co 0 1 2 a2 Input Weights 1 1 1 b Compactness W 0 5 c Color Weight 0 5 d Scale Parameter e Use Optimization f Reliability q Eudidean Distan Segmentation V Border IN Import Classification Class attribute Collect Samples Export Class Sombra 0 samples collected out of 0 objects Figure 3 36 Samples Editor window Mask File Allows selecting a binary mask file in PBM format The segmentation will be performed within the area defined by the mask Background Image Allows selecting which image will be used The images available here are those defined as project resources Segmentation Allows configuring the segmentation process Segmenter Allows configuring the parameters for the segmentation process see note below Opacity Sets the polygons opacity Minimum makes objects transparent Border Sets if the polygons border will be displayed or not Allows also selecting the color of the border Page 43 User Guide InterImage 1 30 Segment Executes the segmentation process Import Classification Allows importing the classification information present in the polygons This functionality should be used only after running the Import Samp
17. e output layer The Border field allows you to enable disable viewing the objects border and the third control sets the polygons opacity Minimum value makes objects transparent 3 4 2 Analysis Tools Window O Analysis Tools Attribute X jury bl O Attribute Y llx 443 000000 103 000000 19 000000 Ds COCO ury 0 000000 443 000000 Selection Threshold Preview E Invert Add Selection Figure 3 34 Analysis Tools window This window provides tools for analyzing the statistical distribution of the segments attributes Page 40 User Guide InterImage 1 30 O Histogram Selects the histogram display mode O Scatter Plot Selects the scatter plot display mode Attribute X Selects the attribute which histogram will be displayed In Scatter Plot mode selects the attribute of the x axis which will be combined with Attribute Y Attribute Y Selects the y axis attribute to generate the scatter plot Bins Defines the number of histogram bars Generate Generates the graph Selection Threshold Line Allows setting a point in histogram mode or a line in scatter plot mode that performs a segments selection Those which stay in the red side are excluded and those which stay in the blue side remain Preview Shows in the Analysis Explorer window the result of the selection made with the chosen threshold Invert Inverts the objects selection criterion Add Selection Creates a Selection block in t
18. he decision rule with the generated selection expression 3 5 Shapefile Editor Window This window Figure 3 35 allows creating and editing shapefiles O Load ESRI Shapefile Loads a shapefile Save ESRI Shapefile Saves the polygons to a shapefile Save Mask Image Saves the polygons as a binary mask in PBM format This mask can be used in Samples Editor to define a region of interest O Clear All Removes all polygons 2 Zoom Zooms in by clicking the left mouse button Zooms out by clicking the right one Allows focusing on a specific part of the image by selecting it with the left button Page 41 User Guide InterImage 1 30 CG Shapefile Editor Figure 3 35 Shapefile Editor window Pan Moves the image by dragging the mouse Create Polygon Enables the polygons creation mode To create a polygon click the left mouse button to create the vertices To close the polygon double click or click on the starting point And to undo a vertex click the right button Edit Polygon Enables the polygons editing mode This mode does not allow you to add or remove points just to move them Delete Polygon Enters the polygons removal mode To remove a polygon just click on it 3 6 Samples Editor Window This window Figure 3 36 allows performing a segmentation collecting samples and classifying them manually generating a shapefile at the end of the process Page 42 User Guide InterImag
19. he pixels that compose the segments corresponding to the bands 2 and 3 of the image The membership functions are defined interactively by the user The function FuzzyML2 is shown in Figure 2 6 The last operation of a decision rule is a union operation This operation is responsible for spatial grouping the set of hypotheses selected at the end of the decision rule There are three possibilities Merge All all hypotheses are combined into a single hypothesis which can cover a not contiguous area Merge Connected each group of spatially connected hypotheses are combined into a single hypothesis covering a contiguous region or No Merge hypotheses resulting from the decision rule are not merged Page 7 User Guide InterImage 1 30 Maximum value 1 00 Minimum value 0 00 Figure 2 6 Example of a membership function See also Node Editor window page 21 Decision Rule window page 27 2 5 Interpretation Control The main task of the system core is to control the interpretation process which consists of two complementary steps top down and bottom up In the top down step the control process traverses the nodes of the semantic net from the root to the leaf nodes calling top down operators and decision rules associated to each node The top down processing occurs in parallel with respect to the branches of the semantic net Object hypotheses associated to semantic nodes are created during this process and organized in a net of hy
20. he resource is an image defines its geographical coordinates Resources List Lists the project resources Add Adds a resource to the project Remove Removes the selected resource from the project A Edit Edits the selected resource Save Confirms the changes made to the selected resource Cancel Cancels the change made to the selected resource Use Remote Server Enables the project to communicate with a server Host Defines the host address Port Number Defines the communication port 3 2 1 Supported Resource Formats Table 3 1 presents the resources formats that can be used in a project Note Give preference to image formats that do not have compression tif because they offer a higher quality than jpg which is compressed This way you will get better results Tagged Image File Tagged Image File Geo Tiff ESRI Shape File ESRI Shape File JPEG JFIF jpg Table 3 1 Supported resource formats 3 2 2 Maximum Image Size It is hard to set an upper limit since the answer to this question involves other quantities such as the size of the semantic net or the number of objects produced by the Page 26 User Guide InterImage 1 30 interpretation In our tests however we could successfully interpret images from up to 9 Megapixels 3 000 x 3 000 pixels 3 2 3 Resources in Different Resolutions InterImage is able to work with shapefiles and images of different sizes resolutions a
21. http wiki dpi inpe br doku php id interimage operators documentation Page 45 User Guide InterImage 1 30 References 1 Camara G Souza R C M Pedrosa B M Vinhas L Monteiro A M V Paiva J A Carvalho M T Gatass M 2000 TerraLib Technology in Support of GIS Innovation In H Brazilian Symposium on Geolnformatics GEOINFO 2000 Sao Paulo Brazil Proceedings of GEOINFO 2000 CD ROM 2 Buckner J Pahl M Stahlhut O Liedtke C E 2001 GEOAIDA A knowledge based automatic image data analyzer for remote sensing data In ICSC Congress on Computational Intelligence Methods and Applications 2001 CIMA 2001 Bangor Wales UK Proceedings of the Congress on Computational Intelligence Methods and Applications 2001 CIMA 2001 CD ROM 3 Pahl M 2008 Arquitetura de um sistema baseado em conhecimento para a interpreta o de dados de sensoriamento remoto de m ltiplos sensores PhD Thesis University of Hannover Translation INPE Sao Jos dos Campos INPE 15211 TAE 71 URL http urlib net sid inpe br mtc m17 80 2008 03 07 18 31 26 Sep 2008 Page 46
22. ides the following options 5 Start Executes an interpretation Undo Undoes the previous interpretation step O Continue Continues the interpretation process until find another breakpoint If this does not exist performs the interpretation until the end D Step Goes one step ahead in the interpretation Page 13 User Guide InterImage 1 30 3 1 1 4 Help Menu Help Content Fl Home Page About Figure 3 5 Help menu The Help menu Figure 3 5 provides the following options 6 Help Content Opens the wiki page where part of the program documentation is concentrated Home Page Opens InterImage website About Displays information about the program 3 1 2 Toolbar 00 0060 8006 Figure 3 6 Toolbar The toolbar Figure 3 6 provides the following options o Execute Executes an interpretation e Debug Mode Enables disables debug mode tn Undo Undoes the previous interpretation step gt Continue Continues the interpretation process until find another breakpoint If this does not exist performs the interpretation until the end TS Step Goes one step ahead in the interpretation Ed Analysis Explorer Tool that helps in building the interpretation model Shapefile Editor Tool for creating and editing shapefiles Page 14 User Guide InterImage 1 30 Samples Editor Tool for segmentation samples collection and manual classification of
23. ies segments with the potential to belong to each class The bottom up operators refine the classifications produced in the top down step confirming or rejecting them and solving possible spatial conflicts between them At the end of the interpretation process the hypotheses become validated object instances The output of the interpretation process is a symbolic description of the scene consisting mainly of a net of object instances and labeled images that correspond to regions associated with object classes From the labeled images the system allows the creation of different thematic maps representing the different levels of semantic concepts in the net Page 2 User Guide InterImage 1 30 2 1 Semantic Net A knowledge model in InterImage contains information used by the control process for the interpretation of a scene It is represented by a semantic net Figure 2 2 where the nodes organization is hierarchical and each node can be associated only with one ancestor node parent and one or more child nodes children Semantic Net D x UrbanArea Shadow GD Vegetation Figure 2 2 Semantic net Each node in the semantic net corresponds to an object class expected to be found in the scene Nodes have properties such as top down and bottom up operators as well as generic parameters and other specific operators See also Semantic Net window page 15 2 2 Top Down Operators When building the interpretation model the u
24. iews Switches display modes E Attribute View Spatializes in grayscale the attribute selected in the Attribute field Being the object that has the lowest value of the attribute in black and the highest value in white a y Classification View Displays objects with the colors of the respective classes that have been associated to them G Selection View Shows in red all objects that existed before the rule execution input and in green those that remained after that output R Analyis Tools Opens a window that allows analyzing the statistical distribution of the segments attributes and thus making better decisions about how to build the rule Attribute Selects the attribute to be used in Attribute View mode Apply Applies the selected exhibition mode to the viewer 5 Execute Executes the decision rule a Export Exports the result of the rule to a shapefile Input Class When there is more than one class in the rule input this control allows you to select which class will be displayed in the viewer Background Image Selects the image that will be displayed in the viewer Input Layer Enables disables the display of the input layer The Border field allows you to enable disable viewing the objects border and the third control sets the polygons opacity Minimum value makes objects transparent Page 39 User Guide InterImage 1 30 Output Layer Enables disables the display of th
25. lects the image bands composition for visualization 3 1 4 2 Shape Tab Keyname Defines the layer nickname Color Selects the polygons color Opacity Sets the polygons opacity Minimum makes objects transparent Layers Px Image Image imagem Keyname 000000000 Type Keyname Visible 1 Image default Figure 3 9 Layers window Border Sets if the polygons border will be displayed or not Allows also selecting the color of the border Page 17 User Guide Interlmage 1 30 Image Keyname Figure 3 10 Image tab Color a Opacity E Border g Figure 3 11 Shape tab 3 1 4 3 Selection Tab Class Selects one of the semantic net classes Keyname Defines the layer nickname Color Selects the polygons color Opacity Sets the polygons opacity Minimum makes objects transparent Border Sets if the polygons border will be displayed or not Allows also selecting the color of the border Page 18 User Guide InterImage 1 30 Class Stage Keyname Color po Opacity Border Figure 3 12 Selection tab 3 1 4 4 Result Tab Opacity Border E Figure 3 13 Result tab Opacity Sets the polygons opacity Minimum makes objects transparent Border Sets if the polygons border will be displayed or not Allows also selecting the color of the border 3 1 4 5 Toolbar oOo 0208 000 Figure 3 14 Toolbar VEN Add Adds a layer to the viewer
26. les operator Class Attribute Attribute that contains the class information It is only necessary when the attribute name is other than class otherwise the Import Samples operator automatically recognizes the classes Import Imports the classification information Sampler Offers the samples collection functionality Class Selects the class for which you want to collect the samples The available classes are the classes present in the semantic net Collect Samples Click this button to start collecting the samples for the selected class To select a polygon click on it To deselect click again Export Exports the segmentation to a shapefile and allows calculating attributes Viewer Allows visualizing and interacting with the image and the sample polygons Information Shows the values of the selected image pixel Geocoordinates Shows the geographic coordinates while moving the mouse over the viewer 53 Fit to Window Centers the image in the viewer As Zoom Zooms in by clicking the left mouse button Zooms out by clicking the Q right one Allows focusing on a specific part of the image by selecting it with the left button O Pan Moves the image by dragging the mouse O Selection Selects deselects sample polygons Samples Information Shows information about the samples collection Page 44 User Guide InterImage 1 30 Note For information about the operators and their parameters please visit
27. nd geographic coverage In the case of images those of lower resolution larger pixel size are resampled according to the higher resolution 3 3 Decision Rule Window Decision Rule for Sombra TopDown Decision Tree Join All Class Selection Membership Aggregation Classify General Lowerdevel Rule Merge All 5 Merge Connected No Merge Figure 3 20 Decision Rule window This window Figure 3 20 allows you to create a set of expressions called decision rule These expressions define the structured and explicit knowledge of the user analyst and are used by the system in the interpretation process Page 27 User Guide InterImage 1 30 3 3 1 Building Blocks Join Class Selection Membership Aggregation Classity Figure 3 21 Building blocks The decision rule is constructed through a set of building blocks Figure 3 21 Join Joins several Class blocks Class Select objects of a particular class or classes Selection Selects objects that meet a certain criterion And and Or Allow you to create powerful logical expressions They are used along with the Selection block Expression Allows creating a new attribute from the result of a mathematical expression Membership Works with fuzzy logic Aggregation Allows a level of the semantic net to pass information to the upper level Classify Classifies objects and solve spatial conflicts It is usually the
28. ociated to S will be executed for H as soon as all H nodes have been created the n index identifies the direct descendants of H The operator rule will evaluate the hypothesis nodes H and decide if they will turn into instance nodes or will be removed from the net In addition the bottom up operator rule will turn the regions R associated to I nodes disjoint solving eventual Spatial conflicts The bottom up operator decision rule will group the nodes and will generate new hypothesis nodes H to which the instance nodes will be connected H will then be put in the hypothesis net and associated to node H being the original node H removed The control is then passed to node H for evaluation and grouping of nodes H The bottom up continues until the hypothesis net root node is processed At this point the instance net T will be complete Page 10 User Guide InterImage 1 30 3 System Interface This chapter will guide you through the main elements of the system interface such as menus toolbars dialog boxes and windows 3 1 Main Window Figure 3 1 Main window The elements of the main window Figure 3 1 are 3 1 1 Menus 3 1 1 1 File Menu The File menu Figure 3 2 provides the following options O New Project Creates an interpretation project As O Open Project Opens a project Edit Project Edits the current project Page 11 User Guide InterImage
29. potheses When the processing reaches the leaf nodes it starts the bottom up step From there the control process passes to visit nodes in the opposite direction calling their bottom up operators and decision rules recursively until the root node is reached and that a network of instances have been created In this case the object hypotheses are discarded or turned into object instances A more formal description of the interpretation process is shown in Figure 2 7 N represents a generic node of a net N is a child node of N N is child of N and so on Ty represents a hierarchically structured net i e a tree and N and N represent different nodes in the same hierarchical level of Ty Page 8 User Guide InterImage 1 30 Top down Rules 22 Bottom up Rules O O and Operators and Operators lH R s Ni Figure 2 7 Interpretation process flowchart Let S be a node of the semantic net Ts and let H and be the nodes of the Ty and T nets created by the interpretation process H represents the node associated to an object hypothesis and represents a node associated to an object instance Let R be the representation of a region in the scene to which H and I are associated R is therefore associated to H or l Considering the hypothesis net Ty all R associated to the H nodes are subsets of R considering the instance net 77 all R and R pairs associated to the l mn and 7 nodes
30. ser attaches top down operators to each node of the semantic net The top down operator task is to identify objects in the image under consideration that are likely to belong to the class corresponding to the semantic node to which it is associated Top down operators are executable programs called by the system core during the process of interpretation They can in principle handle not only images but also any type of geo registered data including vector data in a GIS database digital elevation models or other types of raster data Page 3 User Guide InterImage 1 30 When the core calls the top down operator it passes to the operator information about the geographical boundaries of the region to be processed This region of interest ROI is defined by another operator top down associated to an ancestor node Some top down operators may associate confidence values to the hypotheses identified which may later be used to evaluate these hypotheses by a bottom up operator Decision rules Section 2 4 can be explicitly defined by the user to post process the objects identified by the operator Properties of the object hypotheses can be used in this context This processing can mean simply discard some hypotheses calculate new confidence values or even refine the preliminary classification You can check a node of the semantic network as TopDown Multi Class There can be only one node of this type for the child nodes of the same parent node
31. sify object hypotheses generated by top down Operators or to decide between competing hypotheses of objects during the bottom up step The decision rules defined for an arbitrary node of the semantic net are always executed after the execution of the respective top down and bottom up operators associated to that node InterImage has a specific graphical user interface Section 3 3 to support the definition of decision rules Through this interface the user can code simple rules whose basic elements called building blocks are shown in Figure 2 3 Page 5 User Guide InterImage 1 30 A decision rule processes and presents as output a set of objects It can be considered that the basic steps of a decision rule are i select a set of objects ii filter this set of objects discarding objects within the set iii assign a degree of membership to objects within the set and iv resolve spatial conflicts among objects in the set This last step is only meaningful for decision rules associated with the bottom up step The steps listed above can be combined in different ways to create complex rules Aggregation Figure 2 3 Decision rule building blocks The Class building block allows selecting objects of a particular class associated to a semantic node With this block a set of objects is created which can be joined to another set through the Join block Figure 2 4 shows a simple bottom up decision rule for the Vegetation node of
32. tes of object hypotheses These variables are associated with each object hypothesis e g for each different object it may have a distinct value In Figure 2 5 the Expression block is used to store the brightness attribute value of each hypothesis This variable is then used in a block selection The Expression block allows creating complex arithmetic expressions from the attributes calculated by InterImage The Membership block allows the user to define a membership value for objects which can be done through a combination of membership functions as shown in Figure 2 5 The Aggregation block allows the aggregation of attribute values for the set of selected objects Trees bright brightness imagem bright gt 20 ratio imagem layer3 gt 0 35 Min FuzzyML mean imagem_layer FuzzyML3 mean imagem lavyer3 Figure 2 5 Example of a top down decision rule Figure 2 5 shows an example of a simple top down decision rule for the Trees node in the semantic net in Figure 2 2 In this case a segmentation top down operator was associated to the node Trees Initially all segments for which the brightness and the ratio of the band 4 average value are larger than certain thresholds are selected to be regarded as hypotheses of Trees Then each selected hypothesis is given a value equal to the minimum value relevance between the FuzzyML2 and FuzzyML3 membership functions defined respectively on the average values of t
33. the semantic net shown in Figure 2 2 Basically what the rule does is select all the object hypotheses generated in the top down step for Trees and Grass nodes join these hypotheses through Join block and resolve spatial conflicts between the hypotheses of the two classes of objects through Classify block specializing in the rule to the Spatial Resolve block It is interesting to note that if there is a partial spatial conflict between Trees and Grass hypotheses the hypothesis with the lowest membership value will not be completely discarded only the region that intersects the other hypothesis is suppressed e g the region of the hypothesis with lower membership value will shrink In a decision rule Interlmage can calculate a variety of attributes for the hypotheses of selected objects attributes based on spectral values shape texture and topological characteristics of image segments associated with those hypotheses These attributes can be used to select objects within a set through Selection block with a user defined threshold In Figure 2 5 a combination of selection blocks is used to filter the set of objects created in the top down step for class Trees All objects that do not meet the selection criteria will be removed from the set Spatial Resolve Figure 2 4 Example of a bottom up decision rule Page 6 User Guide InterImage 1 30 The Expression block allows you to create variables with user defined names from attribu
34. tion process The input of such operators is a list of regions each region is associated to an object hypothesis belonging to the classes of the child nodes Decision rules Section 2 4 can be explicitly defined by the user to post process the judgment made by the operator Properties of the object hypotheses can be used in this context This processing may mean discard validate hypotheses or resolve spatial conflicts The validated hypotheses will then be considered object instances It is important to note that instances of objects can at a later stage of the interpretation process be discarded This will happen if a hypothesis of a higher level object is discarded The operator also groups the instances of objects assigning to each group a region equivalent to the union of the regions associated to each instance The groups of objects will originate new hypotheses for the semantic node to which the operator is associated replacing the original hypothesis as will be explained in Section 2 5 The system provides a default bottom up operator called Dummy Bottom Up This operator performs no processing leaving it to the decision rule to judge the hypotheses of child nodes objects Note For information on other top down operators and their parameters visit http wiki dpi inpe br doku php id interimage operators documentation See also Node Editor window page 21 2 4 Decision Rules Decision rules can be used both to reclas
35. will present the basics concepts of the system and some theoretical foundations that will help in the understanding of its operation The system interface will be presented in Chapter 3 along with its main features and screens For practical content as examples of interpretation projects tutorials etc visit our wiki http wiki dpi inpe br doku php id interimage Reports about problems requests for additional information and suggestions about new features can be sent to lvc inter ele puc rio br Page 1 User Guide InterImage 1 30 2 Basic Concepts Figure 2 1 describes the components of the interpretation process in Interlmage The system implements a specific interpretation control strategy guided by a structured knowledge model through a semantic net The interpretation control Section 2 5 is executed by the system core which uses as input a set of geo referenced images SIG layers digital elevation data or other geo registered data Through the interpretation of the scene input data are processed with the help of external programs called top down and bottom up operators sa Top down Rules and Operators Input Output Semantic Net Hypotheses Instances Symbolic Description Thematic Maps System Control Figure 2 1 Analysis process components Top down operators are responsible for the partition of the scene into regions considered as object hypotheses This is a preliminary classification which identif
Download Pdf Manuals
Related Search