BIIGLE
-
|
I requested an area report for the volume, and I noticed that some images had blank columns (annotation_width_m, annotation_height_m, annotation_area_sqm). The lasers were annotated manually. The images have in common that the lasers are found quite close to one another, but I can't understand why it happens and if there is a way to fix it. Thank you! |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
|
Can you please tell me the report ID? This is the number at the end of the download URL. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks! The report IDs are: 20220, 21036, 21037. They are in the same project, but different volumes. |
Beta Was this translation helpful? Give feedback.
-
|
The laser point detection was primarily developed to be used with images where the camera pointed straight down to the sea floor. It implements a hard 50 m² threshold for the image footprint, as larger values are implausible in this context. If the calculated footprint is larger, the result is ignored. This is the case for the images where you get no size/area results in the report. We are already aware that this threshold does not make sense when you have annotated the laser points manually. This will be fixed as part of biigle/laserpoints#86. I can't say when this change will be implemented though. |
Beta Was this translation helpful? Give feedback.
The laser point detection was primarily developed to be used with images where the camera pointed straight down to the sea floor. It implements a hard 50 m² threshold for the image footprint, as larger values are implausible in this context. If the calculated footprint is larger, the result is ignored. This is the case for the images where you get no size/area results in the report.
We are already aware that this threshold does not make sense when you have annotated the laser points manually. This will be fixed as part of biigle/laserpoints#86. I can't say when this change will be implemented though.