Knowledge Base
Recent Changes

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
dev:desktop_ext:image_pixel_objects [2021/04/23 06:03]
paulien [Concepts] 		dev:desktop_ext:image_pixel_objects [2021/04/26 11:58]
paulien removed
Line 1: Line 1:
-====== Image Annotation to Object ====== +====== Image Annotations to Objects ======
-FIXME +
-This page describes how to use the extension "Image Annotation to Object"\\+
  
-{{orbit_desktop:how_to_find.png?25&nolink  |}} [[dev:desktop:workspace:main_toolbar|Main Toolbar]] > Extensions > Image Annotation to Object+This page describes how to use the extension "Image Annotations to Objects"\\
  
-The "Image Annotation to Object" extension makes it possible to convert image annotations defined with pixel coordinates or pixel bounds into 3D coordinates or 3D boxes using the pointcloud resource.\\+{{orbit_desktop:how_to_find.png?25&nolink  |}} [[dev:desktop:workspace:main_toolbar|Main Toolbar]] > Extensions > Image Annotations to Objects 
 + 
 +The "Image Annotations to Objects" extension makes it possible to convert image annotations defined with pixel coordinates or pixel bounds into 3D coordinates or 3D boxes using the pointcloud resource.\\
  
  
 ===== Concepts ===== ===== Concepts =====
  
-Image annotations for mapping run imagery can be defined in pixel coordinates or pixel bounds defined by two pixel coordinates. The image pixels are converted into 3D coordinates or 3D boxes using the mapping run pointcloud. Both original and optimizes images can be used. +Image annotations for mapping run imagery can be defined with two pixel coordinates or pixel bounds defined by four pixel coordinates. The image pixels are converted into 3D coordinates or 3D boxes using the mapping run pointcloud and original/optimized images. 
  
-In the image_pixel.ini file describes the tags in the annotation file used for the conversion.  +The image_pixel.ini file describes the tags in the annotation file used for the conversion.  
-Every annotation can have a tag to overlay on the opened image.+Every annotation (and optionally the tag) can be overlayed on the opened image.
  
  
 +===== Pixel Definitions =====
  
 +Select how the annotations are defined.
  
-===== Source =====+==== Coordinates ====
  
-==== Pixel Definitions ====+Every annotation is defined by 1 coordinate having two pixel coordinates: X and Y. 
  
-Select how the annotations are defined:\\ +==== Bounds ====
-  * Coordinates: every annotation is defined by 1 coordinate +
-  * Bounds: every annotation is defined by 2 pixel coordinates that form the bounds of the rectangle+
  
-==== Reference Files ====+Every annotation is defined by 4 pixel coordinates that form the bounds of the rectangle: MinX, MaxX, MinY and MaxY. 
 + 
 +The center of the bounds can be used to convert to 3D. The same algorithm as for pixel coordinates will be used.  
 + 
 +==== Display Annotations ==== 
 + 
 +The original image annotations and tags can be displayed on the imagery via the [[dev:desktop:preferences:mapping|]]. 
 +===== Reference Files =====
  
 Choose how the reference files are structured: Choose how the reference files are structured:
-  * One file for every image+==== One file for every image ==== 
 +There is one csv or xml file in the original or processed folder that is linked to an image via equal filenames.  
 + 
 +Example of csv file. 
 + 
 +<code> 
 +Filename PixelMinX PixelMinY PixelMaxX PixelMaxY PhotoName Tags 
 +DSC02497 6366 4079 6973 4621 DSC02497.jpg corrosion,structural_steel,light 
 +</code> 
 + 
 +Example of xml file. 
 + 
 +<code> 
 +<?xml version="1.0" ?> 
 +<annotation> 
 + <folder></folder> 
 + <filename>DSC02490.JPG</filename> 
 + <object> 
 + <tags>corrosion,structural_steel,light</tags> 
 + <comment></comment> 
 + <bound> 
 + <xmin>1541</xmin> 
 + <xmax>2008</xmax> 
 + <ymin>4257</ymin> 
 + <ymax>4555</ymax> 
 + </bound> 
 + </object> 
 +</annotation> 
 + 
 + 
 +</code> 
 + 
 +Example of the image_pixel.ini file to describe the xml tags.
          
-  * One file for all images+<code> 
 +Element=object 
 +Filename=filename 
 +PixelX= 
 +PixelY= 
 +PixelMinX=bound/xmin 
 +PixelMaxX=bound/xmax 
 +PixelMinY=bound/ymin 
 +PixelMaxY=bound/ymax 
 +Tags=tags 
 +Attribute0=folder 
 +Attribute1=comment 
 +...     
 +</code> 
 + 
 +==== One file for all images ==== 
 + 
 +There is one csv or xml file to be selected in which all annotations are described. The same structure as for one file for one image can be used. 
  
-Supported formats are .csv and .xml.  
 ===== Target ===== ===== Target =====
  
 After selecting the target file location and CRS, choose to create 3D point objects or boxes.  After selecting the target file location and CRS, choose to create 3D point objects or boxes. 
  
-The result is an ovf with the +===== Algorithms =====
  
-The pixels_not_converted.txt contains the list of objects that couldn't be converted. +For every situation, slices are created along the vector line between the camera position and the pixel point. 
  
 +==== 3D Coordinates from Pixel Coordinates ====
  
 +Within a radius around the vector, points are detected and clustered to create the 3D coordinate. 
  
 +==== 3D Coordinates fom Pixel Bounds ====
  
 +Within the pixel coordinate bounds around the vector, voxels are created and clustered to create the 3D coordinate. 
  
-===== Advanced Parameters =====+==== 3D Boxes from Pixel Coordinates ====
  
-Depending on which source and target options have been set, the used parameters will be enabled.+With a radius around the vector, points are detected and clustered to create the 3D box. 
  
-==== 3D Coordinates from Pixel Coordinates ====+If objects on ground, use ground margin.
  
-Slices are created along the vector line between the camera position and the pixel point. +==== 3D Boxes from Pixel Bounds ====
  
-==== 3D Boxes from Pixel Coordinates ====+Withing the pixel coordinate bounds around the vector, voxels are created and clustered to create the 3D box.
  
-Slices are created along the vector line between the camera position and the pixel point. +If objects on ground, use ground margin. 
  
-==== 3D Coordinates fom Pixel Bounds ==== 
  
  
  
-==== 3D Boxes from Pixel Bounds ==== 
  
  
 +===== Advanced Parameters =====
 +
 +Depending on the chosen source and target options, the used parameters will be enabled.
  
 ==== Search Slice Thickness ==== ==== Search Slice Thickness ====
  
 The thickness of the slices created from camera position to pixel point to search for points.  \\ The thickness of the slices created from camera position to pixel point to search for points.  \\
-  *For creating 3D Coordinates: If a slice has an object, don't search in the following slices.\\ +  *For creating 3D Coordinates: If an object is detected in a slice, don't search in the following slices.\\ 
-  *For creating 3D Boxes: Do search in the following slices up until the Search Dist Max is reached.+  *For creating 3D Boxes: If an object is detected in a slice, do search in the following slices up until the Search Dist Max is reached.
  
 ==== Search Dist Max ==== ==== Search Dist Max ====
  
-The maximum distance from the camera to the detected object. +The maximum distance from the camera to create slices and detect points. 
  
 ==== Search Radius and Cluster Distance ==== ==== Search Radius and Cluster Distance ====
  
-The radius around the vector between camera position and pixel point to search for cluster points.  +The radius around the vector between the camera position and pixel point to search for cluster points within a slice.  
- +This value counts also as the cluster distance which is the minimum cluster gap size. 
-Tooltip: The cluster size to identify coordinates in the direction from camera position to image position. +
  
 ==== Cluster Pts Min ==== ==== Cluster Pts Min ====
Line 93: Line 152:
  
 The ground margin used to identify a 3D box reached the ground to be able to end the clustering.  The ground margin used to identify a 3D box reached the ground to be able to end the clustering. 
 +To be used if the object to identify is located on the ground. If you don't remove the ground, the object will become bigger. 
 +
  
 ==== Voxel Size ==== ==== Voxel Size ====
  
-The box is divided into voxels of this size. +The box in the slice is divided into voxels of this size. 
  
 ==== Voxel Pts Min ==== ==== Voxel Pts Min ====
Line 104: Line 165:
 ===== Results ===== ===== Results =====
  
- +The result is an ovf where every object corresponds to one annotation, having the attributes from the image_pixel.ini file or csv file. The following attributes are extra: 
-The result is an ovf with the +  * FD_ImageAnnotationSequence: Used to indicate the sequence of annotation on the same image.  
 +  * FD_PixelDistanceFromAnnotation: The distance to the pixel coordinate or the center of the pixel bounds, expressed in pixels.  
 +  * FD_InsideAnnotationBounds: The value can be zero/false or one/true. True means that the display of the created object is still within the bounds of the original annotation. False means that the display of the created object is outside the bounds of the original annotation. For pixel coordinate definition always false/zero. 
  
 The pixels_not_converted.txt contains the list of objects that couldn't be converted.  The pixels_not_converted.txt contains the list of objects that couldn't be converted. 
 +