One of my main research interest is to find analytical and numerical solutions to the non-linearity of the photogrammetric orientation algorithms, which I do believe is one of the main reasons that slows the acceptance of photogrammetry as a generic 3D measurement technology. Now the question is: Does the non-linearity of the photogrammetric orientation algorithms is solved? Or there are some remaining issues?
The equation has been linearized but the solution needs needs an initialization of the parameters, otherwise several approaches based on bundle adjustem has been proposed.
Here you can find simple matlab codehttp://www.mathworks.com/matlabcentral/fileexchange/40467-space-photo-resection
Frederico describes a common 2-step approach in which you find a direct approximate solution and then switch to an optimized non-linear solution using the approx. solution as a starting value.
Some of the following references relating to direct solutions might be useful. It's possible they are out of date as I haven't looked at this issue for quite some time.
"Direct linear transformation ..." by Abdel-Aziz and Karara, 1971
"Die gegenseitige Orientierung von zwei Strahlenbuendeln .." by Hofmann-Wellenhof, 1979 (in German). Requires 6 common points in the images.
"A computer algorithm for reconstructing a scene from two projections", Longuet-Higgins, 1981. Needs 8 common points in the images.
"Closed-form solution of absolute orientation using unit quaternions", B. K. Horn, 1987.
Dear Stephen, thank you very much for your feedback. In fact, during the last 10 years I worked on some aspects of the non-linearity problem, in particular, the ones that are related to the exterior orientation parameters. The essence of my question is what else can be done to push the photogrammetric applications into a new horizon and this is by addressing the bottle-neck issues such as the non-linearity problem.
Autonomous initialisation of exterior orientation parameters using a collinearity search‐based solutionGH Seedahmed
The Photogrammetric Record 23 (121), 90-108
Hi Gamal
Thanks for the further clarification of what you want to achieve. It's a good objective. Since I'm a member of RSPSoc I get Photogrammetric Record and have just looked at your reference. I had already marked it with a sticker. You're probably ahead of me on this one but I hope I may have mentioned a couple of papers not on your radar and which you might add to your background research.
Progress here is very useful for close-range photogrammetry and industrial applications, which is where my interests lie. I'm hoping to get a knowledge-base website set up on 3D techniques for industry so maybe stay in touch. Feel free to contact me via [email protected].
Hello Gamal: You might be interested in having a look at the following paper:
Mazaheri, M., and Habib, A., 2015. Quaternion-Based Solutions for the Single Photo Resection Problem. Photogrammetric Engineering and Remote Sensing Journal, Vol. 81, No. 3, pp. 209 – 2017.
This paper presents a quaternion-based closed form and iterative solutions for solving the SPR problem without the need for approximate values or partial derivatives.
Best regards, Ayman.
Dear Ayman: First of all very nice to hear from you. Indeed, I will look to your paper. In fact, I marked your paper in my to do list for reading.
Dear Stephen: Once again, thank you very much for your elaboration. In fact, during the last 7 years most of my practice in photogrammetry is concentrated on close-range applications such as beam deflection, bridge motion monitoring, cracks measurement, industrial parts measurement. During this practice I became deeply aware of several issues the requires more research work such as the non-linearity. Indeed, I will stay in touch. My school email is: [email protected]
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography