Monday, 25 February 2013

Photogrammetry in Highway Engineering


Photogrammetry in Highway Engineering:

            Photogrammetry, often called “remote sensing,” is defined as the science or art of obtaining measurements by means of photography. Quite commonly it is construed more broadly to encompass procedures for photo interpretation and for converting single photographs into composite ones (called mosaics) and info maps. Photogrammetry in this broader sense, and particularly that based on aerial photographs, is today a basic working tool of the highway engineer. Applications appear not only in highway location, but also in planning, geometric design, rights of way, traffic studies, drainage, soil classification and identification, earth-work measurement, materials location, and pavement condition surveys.

Mapping by Photogrammetric Methods:

            In highway practice, aerial photography and the preparation of mosaics and maps may be done in part or in whole by a highway agency or by contract between private companies and the individual highway agencies. If done by contract, the agreement generally stipulates the “results to be obtained” and leaves the “manner and method” of specific photographic and photogrammetric equipment and procedure to the company. To illustrate, a common specification for accuracy of topographic maps states that 90% of the elevations be correct within one-half of a contour interval and remainder within one contour interval. Another requirement is that 90% of the planimetric features be positioned on the map within one-fortieth of an inch of correct location- the rest within one-twentieth of an inch. Incidentally, these specifications clearly demonstrate the high degree of accuracy that can obtained.  
            Vertical aerial photographs taken with the camera pointed straight down are the most useful for highway mapping purposes. The country to be covered is photographed in parallel runs with the individual pictures lapped both in the direction of flight (end lap) and between successive runs (side lap). For stereoscopic uses, end lap must be greater than half the picture width (possibly specified as not less than 55% nor more than 65%) in order that the center (principal point) of one photograph is included in both adjacent photographs. Side lap should average 25%, with percentage less than 15 or more than 55 unacceptable bases to height and width to height ratios. Selection of the height from which photographs are to be taken depends on the uses which they, or the maps to be made from them are to have for map-making purposes, the variables include the focal length of the aerial camera, the desired combination of map scale to photograph scale. The latter is, in turn, a function of the stereoscopic projector used for map making.   
            Several instruments of varying complexity are available for converting data from the aerial photographs into maps. These include the kelsh stereoscopic plotter, the Nistri photograph, the wild autograph, and the Zeiss stereoplanigraph. All utilize the concept that when the area common to a pair of matched photographs is viewed through a stereoscope, the topography is seen in relief. It is possible with any of these instruments to produce an accurate map showing all natural and artificial features. Also, contours may be drawn or spot elevations determined. Only those features that cannot be identified on the photographs must be located by ground measurement. Difficulties will of course be encountered in attempting to map heavily wooded areas where the ground is not visible in the photographs. Even in this instance, mapping is some times possible by taking photographs when the trees are bare or by setting ground elevations by using estimated tree heights.
            Commonly, map making form aerial photographs is done by a skilled operator. However, devices have been developed that determine the locations and elevations of points electronically for eventual punching into computer cards.
            Earlier procedures required ground-control survey points in every photograph to set its scale and elevation. Today, however, procedures have been developed to greatly reduce ground-control requirements by aero triangulation. This is done by successively trying together a chain of geometric figures defined by marked points on the ground. Other points to be incorporated into ground surveys also are marked for easy identification on the photographs.
            Accuracies of ground-control triangulation networks or traverses must be in keeping with the scale of the aerial photographs. It has been generalized that third-order triangulation is satisfactory in rural areas and second-order fro urban locations. For these, maximum errors in distance are 1:5000 and 1:10,000, respectively. Sufficiently accurate angular measurements are easily obtained with modern theodolites. Two new concepts of distance measurement have improved speed and precision. One class of devices, which includes the tellurometer and electro tape, employs high-frequency radio waves. A “master” unit is placed at one end of the line and a “slave” unit at the other. The second group, typified by the geodimeter, employs the phase relationships between outgoing and reflected modulated light waves. Signals from the instrument placed at the one end of the line are returned by a reflector placed at the other. Both kinds of equipment are rugged and portable.
            Increasingly, highway surveys are being tied into the state plain coordinate systems developed by the U.S Coast and Geodetic Survey as an adjunct to its nationwide triangulation network. These master coordinate systems provide a firm base and a check not only for ground-control surveys are only for ground control and for filling in detail that cannot be gained from the photographs. Scale for the finished maps is, commonly, for rural areas 1:62,500 or approximately 1 mi to 1 in.; for urban areas the scale is 1:24,000 ft to 1 in. state by-state lists showing the areas that have been mapped are available on request from the Geological survey.

No comments:

Post a Comment