Forum

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Alexey Pasumansky

Pages: 1 ... 848 849 [850] 851 852 ... 989
12736
General / Re: Generic texture model exported has png file like a puzzle...
« on: September 24, 2014, 11:55:41 PM »
Hello Sky_hie,

Generic mapping mode creates multiple charts to describe the models' texture. In case you need custom parametrization, you can export the model, create custom UVs and bring back the model using Import option for texturing.

12737
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 24, 2014, 08:36:05 PM »
Hello tornado2011,

Thanks, we'll check that. Is there any output in the Console pane?

12738
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 24, 2014, 08:35:27 PM »
Hello Patribus,

Probably the points are closer to the viewpoint than the trackball (center of zoom and rotation)?

12739
Feature Requests / Re: 16Bit
« on: September 24, 2014, 06:00:57 PM »
Could you submit the crash report then?

12740
Feature Requests / Re: 16Bit
« on: September 24, 2014, 05:44:47 PM »
Hello bmc130,

Just repeat the same operation with 16-bit TIFF images. Input, process and export texture.

Pre-release is available for download from the following thread: http://www.agisoft.ru/forum/index.php?topic=2883.0

12741
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 24, 2014, 05:21:59 PM »
Hello hengefjes,

And what is the output in the Console pane related to the tile export? In the export tiles dialog you need to specify export resolution, so the model should be properly referenced to fit these units.
Here's the console output:
Finished processing in 0 sec (exit code 1)
grid: 9x7x3 (5 levels)
Preprocessing images...
preprocessing images... done in 0.697 sec
Finished processing in 0.698 sec (exit code 0)
>>>


And what about the export settings and model referencing?

12742
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 24, 2014, 05:11:07 PM »
Hello ThomasVD,

Tie-point limit defines the upper threshold for the number of projections present on each photo. So actually this parameter effects on the number of points in the sparse cloud. Using zero-value means no limit and the process in such case will work similar to version 1.0. However, if you input zero in Tools Menu -> Thin Tie-Points dialog all the points will be removed.

Note that in some cases the number of projections on some images can be slightly higher than limit.


12743
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 23, 2014, 06:20:41 PM »
Hello hengefjes,

And what is the output in the Console pane related to the tile export? In the export tiles dialog you need to specify export resolution, so the model should be properly referenced to fit these units.

12744
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 23, 2014, 05:50:15 PM »
Hello hengefjes,

OC3 export works only for dense point cloud. Tile export is also based on dense point cloud and would not work if it is missing in the active chunk.

And brief tutorial regarding panorama export is attached.

12745
General / Re: Agisoft PhotoScan 1.1.0 pre-release
« on: September 23, 2014, 05:13:31 PM »
Hello James,

Do you think adding selection while holding Ctrl key should also support multiple selection?

12746
Other Languages / Re: Фильтрация модели
« on: September 23, 2014, 03:06:07 PM »
Добрый день,

Ниже пример такого скрипта. Обратите внимание, что разрешение экспорта указано константами в теле скрипта в градусах, т.е. применимо к системе WGS84. При использовании спроецированных СК потребуется указание разрешения экспорта в метрах.

Code: [Select]
#Batch export of orthophotos based on each photo
#compatibility Agisoft PhotoScan Pro 1.0
#no arguments required

import os
import time
import PhotoScan

doc = PhotoScan.app.document

PhotoScan.app.messageBox("Prossing started.\nPress OK.")  #information message

def intersect(p0, pn, l0, l):
d = ((p0 - l0) * pn) / (l * pn)
return d * l + l0

def surf_height(chunk, photo):
points_h = list()
point_cloud = chunk.point_cloud
num = len(point_cloud.projections[photo])
num_valid = 0

for i in range (0, num):
x = point_cloud.projections[photo][i].index

if (point_cloud.points[x].valid == False):
continue

v = PhotoScan.Vector( (point_cloud.points[x].coord[0], point_cloud.points[x].coord[1], point_cloud.points[x].coord[2], 1) )
vt = chunk.transform * v
vt.size = 3
vt = chunk.projection.project(vt)
points_h.append(vt[2])
num_valid += 1

points_h.sort()
height = points_h[int(num_valid/2)]
return height


chunk = doc.activeChunk

proj = PhotoScan.GeoProjection() 
proj.init("EPSG::4326")

path = doc.path.rsplit("\\", 1)[0]

processed = 0

t0 = time.time()

for i in range (0, len(chunk.photos)):
photo = chunk.photos[i]
photo.enabled = False

PhotoScan.app.update()

for i in range (0, len(chunk.photos)):


photo = chunk.photos[i]

if (photo.transform == None):
continue

x0 = PhotoScan.Vector((0.0,0.0,0.0))
x1 = PhotoScan.Vector((0.0,0.0,0.0))
x2 = PhotoScan.Vector((0.0,0.0,0.0))
x3 = PhotoScan.Vector((0.0,0.0,0.0))

width = photo.width
height = photo.height

# vectors corresponding to photo corners

v0 = PhotoScan.Vector(( -photo.calibration.cx / photo.calibration.fx, -photo.calibration.cy / photo.calibration.fy, 1))
v1 = PhotoScan.Vector(( (width - photo.calibration.cx) / photo.calibration.fx, -photo.calibration.cy / photo.calibration.fy, 1))
v2 = PhotoScan.Vector(( -photo.calibration.cx / photo.calibration.fx, (height - photo.calibration.cy) / photo.calibration.fy, 1))
v3 = PhotoScan.Vector(( (width - photo.calibration.cx) / photo.calibration.fx, (height - photo.calibration.cy) / photo.calibration.fy, 1))
vc = photo.center

v0.size = 4
v1.size = 4
v2.size = 4
v3.size = 4
vc.size = 4
v0[3] = v1[3] = v2[3] = v3[3] = 0
vc[3] = 1

v0_gc = chunk.transform * photo.transform * v0
v1_gc = chunk.transform * photo.transform * v1
v2_gc = chunk.transform * photo.transform * v2
v3_gc = chunk.transform * photo.transform * v3
vc_gc = chunk.transform * vc

v0_gc.size = 3
v1_gc.size = 3
v2_gc.size = 3
v3_gc.size = 3
vc_gc.size = 3

# surface normal

cen_p = photo.center
cen_p.size = 4
cen_p[3] = 1
cen_t = chunk.transform * cen_p
cen_t.size = 3
cen_t = chunk.projection.project(cen_t)

h = surf_height(chunk, photo)

vloc = PhotoScan.Vector((cen_t[0], cen_t[1], h))
vloc_h = PhotoScan.Vector((cen_t[0], cen_t[1], h))
vloc_h[2] += 1

vloc_gc = chunk.projection.unproject(vloc)
vloc_h_gc = chunk.projection.unproject(vloc_h)
surf_n =  vloc_h_gc - vloc_gc

surf_n.normalize()
v0_gc.normalize()
v1_gc.normalize()
v2_gc.normalize()
v3_gc.normalize()

#intersection with the surface

x0 = intersect(vloc_gc, surf_n, vc_gc, v0_gc)
x1 = intersect(vloc_gc, surf_n, vc_gc, v1_gc)
x2 = intersect(vloc_gc, surf_n, vc_gc, v2_gc)
x3 = intersect(vloc_gc, surf_n, vc_gc, v3_gc)

x0 = chunk.projection.project(x0)
x1 = chunk.projection.project(x1)
x2 = chunk.projection.project(x2)
x3 = chunk.projection.project(x3)

x_0 = min(x0[0],  x1[0], x2[0], x3[0])
x_1 = max(x0[0],  x1[0], x2[0], x3[0])
y_0 = min(x0[1],  x1[1], x2[1], x3[1])
y_1 = max(x0[1],  x1[1], x2[1], x3[1])

x_0 -= (x_1 - x_0) / 20.
x_1 += (x_1 - x_0) / 20.
y_0 -= (y_1 - y_0) / 20.
y_1 += (y_1 - y_0) / 20.

reg = (x_0, y_0, x_1, y_1)

photo.enabled = True
PhotoScan.app.update()
p_name = photo.path.rsplit("/", 1)[1].rsplit(".",1)[0]
p_name = "ortho_" + p_name

proj = chunk.projection   ##if chunk projection units are in meters - dx and dy should be specified in meters:

if chunk.exportOrthophoto(path + "\\" + p_name + ".tif", format = "tif", blending = "average", color_correction = False, projection = proj, region = reg, dx = 2.93022e-06, dy = 2.12914e-06, write_world = True):
processed +=1
photo.enabled = False

for i in range (0, len(chunk.photos)):
photo = chunk.photos[i]
photo.enabled = True

PhotoScan.app.update()

t1 = time.time()

t1 -= t0
t1 = int(t1)

PhotoScan.app.messageBox("Processing finished.\nProcessed "+ str(processed) +" images to orthophotos.\nProcessing time: "+ str(t1)  +" seconds.\nPress OK.")  #information message



12747
Feature Requests / Re: Batch Orthophoto Export
« on: September 23, 2014, 02:18:00 PM »
Hello all,

In the version 1.1 the list of batch export operations has been extended (including orthophoto/DEM export).

12748
General / Re: Problem with orthophoto seamlines!!
« on: September 23, 2014, 02:01:44 PM »
Hello Max,

In future - yes, but definitely not in 1.1 version.

Anyone can support this feature: http://www.agisoft.ru/forum/index.php?topic=1777.0 (however, your post is already there, Max). We're paying much attention to all requests, even some of the topics remain unreplied.

12749
Feature Requests / Re: 16Bit
« on: September 23, 2014, 01:54:27 PM »
Hello,

Could you please try version 1.1 pre-release to check if the output texture for 16-bit TIFF input?

12750
Hello Sylvain,

Have you submitted the crash report already? Or probably you can provide the console pane output (log file) related to the problem?

We've received your dataset, but could not manage to reproduce the problem.

Pages: 1 ... 848 849 [850] 851 852 ... 989