Analitik ve digital fotogrametrik nirengi yöntemlerinin kıyaslanması
Başlık çevirisi mevcut değil.
- Tez No: 75592
- Danışmanlar: DOÇ. DR. GÖNÜL TOZ
- Tez Türü: Yüksek Lisans
- Konular: Jeodezi ve Fotogrametri, Geodesy and Photogrammetry
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 1998
- Dil: Türkçe
- Üniversite: İstanbul Teknik Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Jeodezi ve Fotogrametri Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: 99
Özet
ÖZET Bu çalışmada, fotogrametrik nirengi konusunda son birkaç yıl içerisinde kaydedilen gelişmeler ve yeni yöntemler ele alınarak, günümüzde bu kapsamda dünyada gerçekleştirilen araştırma çalışmalarının gerisinde kalmamak, uygulamaya girmekte olan bu konuların ülkenizde de yaygınlaştırılmasına öncülük yapmak amaçlanmıştır. Bu çalışma yedi bölümden oluşmaktadır. Birinci bölümde konunun önemi vurgulanmış, ülkemizde ve yurtdışında yapılan çalışmalara değinilmiştir. 1960'lı yıllardan beri uygulanan ve bir kısmı geçerliliğini yitirmekte olan, klasik fotogrametrik nirengi ölçüm ve dengeleme yöntemleri, ikinci bölümde ana hatlarıyla verilmiştir. Üçüncü bölümde, harita üretim çalışmalarının en önemli maliyet unsurlarından birisini oluşturan yer kontrol noktalan sayısında, çarpıcı bir tasarruf sağlayan kinematik GPS destekli fotogrametrik nirengi yöntemi açıklanmıştır. Sayısal görüntüler ve bu görüntüleri işleyebilecek yetenekteki sayısal fotogrametrik çalışma istasyonlarının geliştirilmesiyle gündeme gelen, otomatik fotogrametrik nirengi yöntemi, dördüncü bölümde incelenmiştir. Sayısal fotogrametrik nirengi ve otomatik SAM (sayısal arazi modeli) oluşturmanın esası görüntü eşleme yöntemlerine dayanır. Bu yöntemler ve aralarındaki farklar dördüncü bölümde vurgulanarak, sistemin genel iş akışı açıklanmıştır. Beşinci bölümde, tez çalışması kapsamında ele alman konularla ilgili olarak yapılan uygulamalarda kullanılan yazılım ve donanımlar, altıncı bölümde de; çeşitli ölçeklerde ve farklı boyutlarda seçilen bloklar ile gerçekleştirilen ölçüm ve dengeleme sonuçlan verilmiştir. Yedinci ve son bölümde ise, elde edilen sonuçlar doğrultusunda, her iki yöntemin de uygulamaya sokulmasıyla sağlanacak yararlar açıklanmıştır. viii
Özet (Çeviri)
SUMMARY THE COMPARISION OF ANALYTICAL AND DIGITAL AERIAL TRIANGULATION METHODS The main purpose of this study is to examine the last developments and new methods which are related to aerial triangulation and, to try to be a pioneer for widespread applications in Turkey. This study consists of seven chapter. Following an introduction, the significance of subjects is pressed in the first chapter. The conventional aerial triangulation methods, such as strip, polynomial, independent models and bundle adjustment are overviewed in chapter two. One of the main topic aimed in this study“Kinematic GPS supported aerial triangulation”is given in the third chapter. As it well known, the main problem in stereo compilation of aerial photographs is to determine the parameters of exterior orientation. The coordinates of camera projection center (Xo, Yo, Zo) are determined by means of sufficient numbers of control points. On the other hand, the coordinates of projection centers are determined by using relative kinematic GPS positioning, combination with minimum number of control points, in combined block adjustment. Generally diferential phase observations are used in GPS supported aerial triangulation for accuracy reasons. The precision of positioning mainly depends on the geometry of satellite cluster. Nowadays, distribution of GPS satellites provide a PDOP (Position Dilution Of Precision) value equal or lower than 6, almost every hour of the day and every where of the world. At least two GPS receivers are used in relative kinematic GPS positioning. One of them is located on the ground and may be 400-500 km around the project area, and the other is located inside of the airplane. Both receivers collect carrier wave phase observations (Lj, L2) with C/A and/or P code pseudorange in the data rate of 1 second or lower, at the same time. Relative kinematic positioning eliminates almost all systematic positioning errors and also reduces the effects of cycle slips. Three main problems can be occured during GPS positioning :. GPS receiver and camera operate independently at the exposuring time (time offset).. GPS antenna and aerial camera may be located in different spatial position in the aircraft ( antenna eccentricity).. Phase ambiguity has significance during data processing. The general approach to these problems are also given in the third chapter. On the other hand, the most significant factor to the accuracy of kinematic GPS positioning IXis the deviation of coordinates depending on the time, that means drift errors. These errors are corrected in combined block adjustment. The mathematical model of the combined block adjustment and the possible block configurations are also given in detail, in chapter three. The fourth chapter deals with digital aerial triangulation. The main concepts such as digital image, digitization, image matching and the methods of image matching are given in that chapter. The most important difference among various matching algorithms is the use of different main primitives in matching. The primitives fell into two broad categories; either window composed of gray values or features extracted in each image a priori are used in the actual matching step. The resulting algorithms are usually called area based matching (ABM), and feature based matching (FBM), respectively. In both cases there is a choice between local and global support for the primitives. The terms local and global are not sharply defined. Local refers to an area seldom larger than about 50*50 pixels in image space, global means a larger area and can comprise the whole image. Some software are based on feature extraction and feature matching, some others work with area matching methods, like cross-correlation or least squares matching. Each method has its own conditions of application and system consequences. A basic difference between feature matching and least squares matching is the final precision. With feature matching the standart deviation of image coordinates is 0.3- 0.4 pixel, whilst least squares matching can reach 0.1 pixel. The typical follow chart comprising of block preperation, block measurement and block adjustment of the automatic aerial triangulation is also given in detail in that chapter. Hardware and software used in that study are given in chapter five. SKIP(1.2) (Inpho GmbH) is used for GPS data processing. The blocks which were established by means of both analytical and digital measurements were adjusted by PATB-GPS program. Aerial photograhps of the selected blocks were digitized by using SCAI roll film scanner (C. Zeiss). Both 14 urn and 28 um pixel resolutions were used. Digital image measurements were carried out by ST 10 digital photogrammetric workstation, using Phodis AT automatic aerial triangulation program package. The features of these hardware and software are given in that chapter. The sixth chapter deals with practical applications which were carried out for that study. Applications can be seperated in two groups; for kinematic GPS supported aerial triangulation and for automatic aerial triangulation. One of the blocks (Imrahor) for first group consist of 157 aerial photograhps, which were taken by RMK-TOP (30cm) camera in standart overlaps. There are 1 1 east-west and 3 cross- strips. There are 9 control points distinguished almost to the corners of the blocks. GPS data were collected by two Trimble GPS receiver in 0.5 seconds data rate. Mean PDOP value was 3.0 and mean approximate coordinate accuracy was ± 15 m. around.Photogrammetric image measurements were carried out by using Planicomp CI 00 Analytical Instrument. The block was adjusted by PC based PATB-GPS software. Obtained results are given at below table. It is clear that the results given above table are sufficient for large scale mapping. These results equal or even better than the results obtained by independent models method in standart (i=2b, 4b) control point distribution. Required control points to be established and measured on the ground are reduced almost 90 %. That is very significant for countries where requirements at various scales mapping increase day by day. XIOn the other hand, the disadvantages of the SKIP(1.2) software which was used in that study such as using only L, frequency, the necessity of the same type receiver on both stations, using of the same same data rate are being eliminated by SKIP(2.2) version, which will be commercially available in near future. Also instead of approximate solution, absolute phase ambiguity will be provided by that version. Nowadays, kinematic GPS supported aerial triangulation method is being used only at General Command of Mapping. This method should be prolonged to the all mapping organisations in Turkey. Also, to compare the resulting accuracy and time consumption of both analytical and digital aerial triangulation methods, some block measurements were carried out.One of these blocks consists of 51 photographs at the scale of 1 : 16 000. There are five east-west strips in the block. The diapositives were scanned in 28 pm pixel resolution. Artificial tie points pugged with PM-1 point transfer device were used for analitical measurements. On the other hand, more than 200 tie points for each image were extracted by Phodis AT Software. Also, the interior orientations of the images were accomplished in an automatic way with 0.1 - 0.2 pixel accuracy. The overall results of the block adjustment is given below table. The conclusion and remarks are given in chapter seven. Here, the attractive result is time consumption for above mentioned block. 60 hours consumed in analytical method for whole process. On the other hand, 20 "hours with digital method were sufficient. XIIThe scanning time for roll films or diapositives during the block preperation step may be eveluated as a disadvantage comparing with analytical method. But, using digital images become an advantage when the overall system performance in digital production steps such as orthophoto, mosaics are considered. Kinematic GPS supported aerial triangulation also provides an advantage to the automatic aerial triangulation by means of derived projection center coordinates, to be used in block topology. As it well known only two images can be used at the same time for measurement in analytical method. The homologous points can be measured in six or more images at the same time in most of the digital methods. Normally human operator is not required for automatic tie point extraction. However, manual tie point measurement is required when the results of the block adjustment are transferred to the analog or semi-analytical instruments. High number of the extracted tie points provide a stable block geometry. The rate of miss-matching is also very low (1-3 %). Some batch processes such as computation of image pyramid and automatic tie point extraction can be carried out after working time. That means, an extra capacity may be provided. It can be seen that the resulting accuacy (cr0) of digital method is better than the analytical method. However, the r.m.s. values of Z coordinates for block points are higher than the analytical method. As a result, we may assest that the automatic aerial triangulation has reached accuracy level of analytical method. On the other side, the present prices of digital systems are lower than the prices of analytical instruments, and it may be expected that in the near future they will be cheaper than today. Operators who are used to working in analytical systems can work with digital systems in most of the production steps. Blocks consisting of 250-300 models can easilly be manuplated by analytical instruments. But digital systems have a constraint that the large volume of digital data needs high disk capacity. It is not easy to process more than 100 images at the same time without dividing subblocks. On the other side, every subblock needs at least four control points at the corners. That is acctually a block when the kinematic GPS supported aerial triangulation is considered. xiii
Benzer Tezler
- Büyük ölçekli sayısal harita üretiminde fotogrametrik nirengi işlemleri
Aerial triangulation procedures in the production of large scale digital maps
OSMAN AKKAYA
Yüksek Lisans
Türkçe
2000
Jeodezi ve FotogrametriYıldız Teknik ÜniversitesiJeodezi ve Fotogrametri Ana Bilim Dalı
PROF. DR. ABDULLAH PEKTEKİN
- Sayısal fotogrametrik sistemlerde görüntü ayarlarının otomasyonu
Automation of image orientation of digital photogrammetric systems
CUMHUR ALTAY
Yüksek Lisans
Türkçe
1998
Jeodezi ve FotogrametriYıldız Teknik ÜniversitesiJeodezi ve Fotogrametri Mühendisliği Ana Bilim Dalı
PROF. DR. ABDULLAH PEKTEKİN
- Ülkemizde Kadastral Amaçlı Sayısal Fotogrametrik Harita Yapımı
Digital Photogrammetric Map Production For Cadastral Purposes In Our Country
ORHAN ERCAN
Yüksek Lisans
Türkçe
1988
Jeodezi ve FotogrametriYıldız Teknik ÜniversitesiHarita Mühendisliği Ana Bilim Dalı
DOÇ. DR. AYHAN ALKIŞ
- Digital topografik haritalar ile digital ortofoto haritaların doğruluk, maliyet ve üretim zamanı açısından karşılaştırılması
Comparison of digital orthophoto maps with digital topographic maps
OKTAY EKER
- Yazılım fotogrametrisi ile orman alanlarına yönelik coğrafi verilerin elde edilmesi
The acquisition of geographic data through softcopy photogrammetry in forestry
HÜSEYİN YURTSEVEN
Yüksek Lisans
Türkçe
2008
Jeodezi ve Fotogrametriİstanbul ÜniversitesiOrman Mühendisliği Ana Bilim Dalı
DOÇ. DR. HAKAN YENER