errors on maps/gt - check it out

[trigal]

there are some errors/discrepancy bewteen the OSM ground truth and the BEVS.

After a review process performed using check_osm_with_gt.py and the associated output stored in
sftp://[email protected]/media/augusto/500GBDISK/nn.based.intersection.classficator.data/check_osm

with the output that is like an image + a txt file; if the first image differs from the last image, there is an error;

• first: anchor,
• second: image that should be like the anchor, a positive,
• third: a negative
• fourth: the category of the anchor in the GT

['/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_09_30_drive_0018_sync/OSM/0000000003.png'] is type: 6
['/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_09_30_drive_0027_sync/OSM/0000000643.png'] is type: 6
['/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_09_30_drive_0020_sync/OSM/0000000549.png'] is type: 4

Last IMG is the GT of the ANCHOR
anchor lat: 49.049534450773
anchor lon: 8.3965921942334
positive lat: 48.986718086668
positive lon: 8.3938004084313
negative lat: 49.05333930648
negative lon: 8.3969599629762

Alvaro ended up with the following issues

Secuencia 28
	frame 3378 
	frame 3379 
	frame 4028
	frame 4029
	
Secuencia 33
	frame 758 - 772
	frame 991 - 997
	
Secuencia 27 (10)
	frame 1055 - 1072
	frame 1220 - 1244
	frame 3850

[trigal]

Frames 3378 and 3379 should be labelled as type 3 rather then as type 0 (see the map); but there is a car and probably in the GT we missed the 1 at the end;
tendrian que estar como type 3, pero el la GT lo tenemos a 0 porque ya no se ve mas nada; asi que esta bien la GT

so, change GT from

...
0000003378;2;0;0   -->   0000003378;2;3;1
0000003379;2;0;0   -->   0000003379;2;3;1
0000003380;1;0;0   -->   0000003380;1;3;1
0000003381;1;0;0   -->   0000003381;1;3;1
0000003382;1;0;0   -->   0000003382;1;3;1
0000003383;0;0;0   -->   0000003383;0;3;1
...

saved in both

	/media/ballardini/4tb/ALVARO/Secuencias/2011_09_30_drive_0028_sync 
	/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_09_30_drive_0028_sync

[trigal]

0000004028.png and 0000004029.png the type 4 (up+right) is correct; the issue is with the frame generated by the software (first image)

car coming from above

this is due to the number of points in the map

so, map was simplified as:

---

map changed in:

/media/ballardini/4tb/ALVARO/ira-software/maps
sftp://149.132.178.247/home/ballardini/alvaro_ws/src/ira_open_street_map/maps

BEV images were updated accordingly from

to

the process includes

1. change the map with JOSM
2. delete the files from the tmp folder sftp://149.132.178.247/tmp/GT which structure is like 
ballardini@johnny-i5:/tmp/GT$ tree -d
.
├── 2011_09_30_drive_0018_sync
├── 2011_09_30_drive_0020_sync
├── 2011_09_30_drive_0027_sync
├── 2011_09_30_drive_0028_sync
├── 2011_09_30_drive_0033_sync
├── 2011_09_30_drive_0034_sync
├── 2011_10_03_drive_0027_sync
├── 2011_10_03_drive_0034_sync
├── in the root the script will put the BEVS
3. move the images to the corresponding folder and then change the filenames

/tmp/GT/2011_09_30_drive_0028_sync
ballardini@johnny-i5:/tmp/GT/2011_09_30_drive_0028_sync$ mv ../GT*.png .
ballardini@johnny-i5:/tmp/GT/2011_09_30_drive_0028_sync$ for i in $( ls *.png ); do alvaroFilename=`echo $i | cut -d '_' -f 2 ` ; mv $i 00000$alvaroFilename; done
ballardini@johnny-i5:/tmp/GT/2011_09_30_drive_0028_sync$ 

4. these images then need to be moved to the "backup" on my 4tb disk and to the ivan-pc to be loaded in the NN.

[trigal]

Secuencia 33, frames de-a 758-772:

there were errors in the map.

regarding 991 to 997 i think there is a road missing in OSM.

in the following map, the GREEN path are stairs, is not the road

neither google maps does not help, and no street view is available

mmmm

no!!!! the fact is that this is not a road!

then, again, change in map 2011_09_30_drive_0033_sync.osm

saved in

/media/ballardini/4tb/ALVARO/ira-software/maps
sftp://149.132.178.247/home/ballardini/alvaro_ws/src/ira_open_street_map/maps

also the GT needs an update for

0000000962;5.83095;1;0 	-->> 0000000962;5.83095;3;0
0000000963;5.83095;1;0 	-->> 0000000963;5.83095;3;0
0000000964;5;1;0 		-->> 0000000964;5;1;0
0000000965;3.60555;3;0 	-->> 0000000965;3.60555;3;0
0000000966;2.82843;3;0 	-->> 0000000966;2.82843;3;0

This is a screenshot of the operations...

files update in

/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_09_30_drive_0033_sync/OSM
/media/ballardini/4tb/ALVARO/OSM/2011_09_30_drive_0033_sync

[trigal]

2011_10_03_drive_0027_sync

frame 3850

missing "in crossing"

0000003848;2.23607;3;1
0000003849;1.41421;3;1
0000003850;1;0;0  <<<< here, last item is 1 not 0 -> 0000003850;1;0;1
0000003851;79.6304;0;0
0000003852;78.2368;0;0

adding "1" to the GT and copying/sync this file to

/media/ballardini/4tb/ALVARO/Secuencias/2011_10_03_drive_0027_sync
sftp://[email protected]/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_10_03_drive_0027_sync

---

frames 1055-1072 have this kind of error

probably error in map

---

frames 1220-1244

issue here is the "density" of the OSM map

tried to simplify the road and split the ways

but there was no way to create the type 2 intersection; modifying the GT accordingly.

0000001237;5.09902;2;0  -->> 0000001237;5.09902;4;0
0000001238;14.8661;2;0  -->> 0000001238;14.8661;4;0
0000001239;14.1421;2;0  -->> 0000001239;14.1421;4;0
0000001240;14.1421;2;0  -->> 0000001240;14.1421;4;0
0000001241;14.1421;2;0  -->> 0000001241;14.1421;4;0
0000001242;12.7279;2;0  -->> 0000001242;12.7279;4;0
0000001243;12.7279;2;0  -->> 0000001243;12.7279;4;0
0000001244;12.0416;2;0  -->> 0000001244;12.0416;4;0
0000001245;11.3137;4;0  -->> 0000001245;11.3137;4;0
0000001246;10.6301;4;0  -->> 0000001246;10.6301;4;0

values before (1220-1236) are not considered and do not appear in the PNGs, skipping from 1217 to 1237

gt file stored in

/media/ballardini/4tb/ALVARO/Secuencias/2011_10_03_drive_0027_sync
sftp://[email protected]/home/malvaro/Documentos/DualBiSeNet/data_raw/2011_10_03_drive_0027_sync

[trigal]

@Pharaun85 tendria que estar terminado, pero yo creo que podemos nuevamente generar los "check_osm" y verificar todo de nuevo... tarde horas a hacer estas cosas :-(

los nuevos png para controlar todo estan aca
/media/augusto/500GBDISK/nn.based.intersection.classficator.data/check_osm_again

[trigal]

Secuencia 33, frame frame 964

0000000960;8.06226;1;0
0000000961;7.2111;1;0
0000000962;5.83095;3;0
0000000963;5.83095;3;0
0000000964;5;1;0     ->>> this is 3 non 1! 
0000000965;3.60555;3;0
0000000966;2.82843;3;0
0000000967;22.3607;3;0
0000000968;21.3776;3;0

change in GT;
no need to "generate" new real-OSM png(s)