What are the three development stages of photogrammetry and their respective characteristics?

Photogrammetry has experienced three development stages: analog method, analytical method and digitalization.

1. Simulating aerial photogrammetry refers to simulating the photography process by optical or mechanical methods, so that the position, attitude and relationship of two projectors can be restored during photography, forming a geometric model smaller than the actual situation, that is, the so-called geometric inversion of the photography process. The measurement on this model is equivalent to the field measurement, and the measurement results are directly drawn on the drawing table by mechanical or gear transmission, such as topographic maps or various thematic maps.

2. With the development of computer and computing technology, people began to use computers to complete complex geometric calculations and a large number of numerical calculations in photogrammetry. This led to the emergence of analytical aerial triangulation, analytical mapping and numerical control orthophoto projectors, and began in the late 1950s, opening up a new era of analytical photogrammetry.

1957 Dr. Chejlava put forward the idea of analytical drawing by computer. Due to the development level of computers at that time, analytical and drawing instruments have experienced nearly 20 years of development and trial use. In the mid-1970s, with the development of computer technology, analytical drawing instruments entered the commercial stage.

3. The further development of analytical photogrammetry is digital photogrammetry. Broadly speaking, digital photogrammetry refers to collecting digital graphics or digital images from the data obtained from photogrammetry and remote sensing, and processing various values, graphics and images in a computer to study the geometric and physical characteristics of the target, so as to obtain various forms of digital products and visual crystal products.

The digital products here include digital map, digital elevation model, digital orthophoto, survey database, geographic information system and land information system. The visualization products here include topographic map, thematic map, vertical and horizontal section map, perspective map, orthophoto map, electronic map, trend map and so on.

The method of fully automatic digital processing of digital/digitized images in computer is called full digital photogrammetry, which includes two parts: automatic image matching and positioning and automatic image interpretation. Automatic image matching and positioning is the analysis, processing, feature extraction and image matching of digital images, and then the spatial geometric positioning is carried out, the digital elevation model is established, and the digital orthophoto image is obtained. The visualization products are contour map and orthophoto map.

Automatic image matching and location is a computer vision method because it can replace the stereoscopic observation process of human eyes. Automatic image interpretation is to solve the qualitative description of digital images, which is called digital image classification. The low-level classification methods of digital images are based on gray scale, features and textures, and statistical classification methods are often used. The advanced classification of digital images is based on knowledge and constitutes a classification expert system.

principle

Although there are various classifications of photogrammetry, their basic theoretical basis is the same, that is, the mathematical model of photographic images. For a single photo, this mathematical model is based on the fact that the object point, the lens center and the image point are on the same straight line during photography, and the equation thus established is called collinear condition equation or imaging equation.

For a stereo image pair (two photos with certain overlapping images taken by different photography stations), a mathematical model that can show the geometric relationship between inside and outside can be derived. In practical work, these mathematical models constitute the theoretical basis of single-image photogrammetry and double-image photogrammetry.