D.А. Roschin - Research Scientist, FSBI «3 CNII» of RussiaDefenseMinistry (Moscow) E-mail: whwhwh@mail.ru

The photography is just the projection of objects on the image plane and we can judge about properties and characteristics of depicted objects only indirectly based on our knowledge about them. Besides the quality of the resulting image depends on many factors such as the characteristics of the camera, the shooting angle, light and the speed of movement of the objects on the stage relative to the survey point. But despite all the diversity of external factors affecting the quality of the image, the picture remains virtually the only visual data carrier that can contain the total information about the shape of the object, its size, color, and spatial location relative to other objects. To improve the accuracy and precision of information on photos about the object, using the special precision camera and developed methods of scanning to produce digital images of objects for which high precision is possible to determine their geometrical parameters. The article discusses various methods for the detection of object contour, having the shape of a ball, in a digital image. Comparative analysis of methods to determine the most accurate and resistant to the effects of the threshold multiplier used as detector. The proposed method definition of gradient by calculating the maximum values of the Laplace operator, which allows to significantly reduce the error detection circuits and to avoid gross errors in the further estimation of the geometric parameters of the object. The photogrammetric system based on single camera is described. The system is used as a means of measuring distances to objects. Estimation of technical and metrological features is realized. A ball-shaped target of known radius is used as an ancillary object, which is attached to the object which distance we need to determine. The size of the target-s radius on photo is determined by using digital image processing. To restore the contour of the target is used circumference approximation by the method of least squares. By matching the physical size of the radius of the target and its size in the image, the distance to the desired object is calculated. At the end of the study were made conclusions characterizing the capabilities of modern cameras that are able to detect an object with a radius of a few centimeters at considerable distances on the order of kilometers but to measure the distance to the object with acceptable accuracy up to a few millimeters will be possible only at a small distance from him, amounting to several meters.

 

1. Kishik A.N. Cifrovaja fotografija. Prakticheskoe rukovodstvo po semke i obrabotke izobrazhenijj v PhotoshopCS. M.: OOO «DiaSoftJUP». 2005. 352 s.
2. Ajjsman Kehtrin, Duggan SHon, Grejj Tim. Cifrovaja fotografija. Iskusstvo fotosemki i obrabotki izobrazhenijj. Izd. 2-e. SPb.: OOO «DiaSoftJUP». 2005. 640 s.
3. JAne B. Cifrovaja obrabotka izobrazhenijj. M.: Tekhnosfera. 2007. 584 s.
4. Canny J. A Note on the Gradient of a Multi-Image // Computer Vision, Graphics, and Image Processing. 1986. V. 33. R. 116-125.
5. Di Zenzo S. A Note on the Gradient of a Multi-Image // Computer Vision, Graphics, and Image Processing. 1986. V. 33. R. 116-125.
6. Roshhin D.A. Povyshenie ehffektivnosti opticheskikh sistem raspoznavanija obrazov na osnove cvetovojj gistogrammy // ZHurnal «Prikladnaja informatika». 2011. № 4(34). S. 120-124.
7. Gonzalez R., Woods R. Digital Image Processing. New Jersey. Printice-Hall. Inc. 2002. 1072 p.
8. Roshhin D.A. Povyshenie tochnosti obnaruzhenija kontura na cifrovykh izobrazhenijakh dlja  obektov, imejushhikh formu shara // ZHurnal «Prikladnaja informatika». 2016. № 10. S. 83-86.
9. Karetin A.N. MNK dlja approksimacii dannykh okruzhnostju. URL: http://mykaralw.narod.ru/articles/mnk_circle.pdf
10. Roshhin D.A. Nesmeshhennaja ocenka raskhozhdenija shkal vremeni territorialno raznesennykh ehtalonov vremeni i chastoty po signalam globalnykh navigacionnykh sputnikovykh sistem // Nauchnyjj zhurnal «Aprobacija». 2015. № 12 (39). S. 35-39.