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 - Trimatrik92

Pages: [1]
General / Re: Scaling a model in Z direction
« on: April 22, 2020, 08:13:28 AM »
Hi Alexey

I observed the model shrinking along z-axis. As you said, I performed camera alignment optimization operation having all the parameters from the General section enabled. However, there is still a shrink of factor 2 along z-axis.

General / Scaling a model in Z direction
« on: April 20, 2020, 08:42:56 PM »
After scaling a object by importing camera coordinates in the reference pan, I observed the 3D point cloud to be down scaled by a factor of 2 in z-direction. The scaling was perfect in x, and y-direction. Since Z can be affected by the focal length or the photographing height, both were checked and seemed okay. In the camera calibration box, the focal length was appeared as it is and the photographing height was measured from the test object to the front part of the camera lens.  How can I correct the scaling in Z direction?

I am doing a project to measure the roughness parameter of a plastic surface( 35 mm X 35 mm) coated with a matte white spray paint. I am using Nikon 3400D with macro lens (60 mm focal length) and the camera is fixed with a stepper motor that can move over the surface in x and y-direction. No reference objects(scale bar/chess board) were used and the camera is set upside down to take the surface images. Camera positions and movements are known and I imported this known coordinates(CSV files) in Agisoft. Every time when exporting the coordinates from the point cloud, I am getting the result with a large deviation when comparing with the standard microscopic values. I really don't know how to figure out this problem. Can anyone help me out here? Thanks.

General / Dense Point cloud generation : too many outliers
« on: June 28, 2019, 07:05:04 PM »
Hi all

I have created a 3-D model of a small object (35 mm x 35 mm) to have some statistical measurements. My camera model is Nikon 3400D and I used a stepper motor to move the camera over my surface and have taken 30 photos of the surface. I placed the surface on a white table and used LED to illuminate the object. Camera settings are same for all 30 pictures. From the camera movement file, I imported the camera positions (coordinates) in photoscan instead of using any scale bar. In point cloud model I observed a lot of outliers which is causing error when comparing the data with the microscopic values (standard).

Does anyone have any suggestions? I have attached the photos in the link below.

Thanks !

General / Camera calibration : When to use ?
« on: June 21, 2019, 07:36:38 PM »
I'm using Photoscan to make a 3D point cloud for a small object (35 mm * 35 mm) and trying to measure the roughness parameter of the surface. Should I use Agisoft lens for calibration? I am not sure when do I need to use camera calibration for making a 3D model. Is there any other way to calibrate camera while using Agisoft photoscan?
Thanks !

General / Measurement of Surface Roughness
« on: January 26, 2019, 01:18:26 AM »
I have created 3D models of some Frost surfaces using Agisoft photoscan and tried to evaluate the roughness parameters (RMS value and peak-to-valley height).  Camera position was known in sort of coordinates and microscopic photos were taken. However, The error for one surface was much higher while evaluating the RMS height percentage difference and peak-to-valley height. Would you please recommend me how to minimize this error using Agisoft or any other photogrammetry software?  Thanks.

** Frost surface and the result RMS error are attached below.

General / Need Suggestion: Minimuzing sourface roughness error
« on: January 22, 2019, 11:48:02 AM »
I have created a 3D model of a frost surface using Agisoft photoscan and tried to evaluate the roughness parameters (RMS value and peak-to-valley height).  Which photogrammetry software other than Agiaoft may work best for this purpose?  Thanks.

Pages: [1]