Sławomir Mikrut

Integration of Image and Laser Scanning Data Based on Selected Example

Abstract The paper aims at presentation of results of research on integration of image and laser data based on selected example. Since a few years the authors have been conducting research on processing image data, and those obtained from laser scanning in the form of the so-called point cloud. In experiments data from terrestrial and mobile laser scanning gained for two different objects were compared: a parish house from Goźlice located in the open-air ethnographic museum at the village of Tokarnia, Poland, and part of the Cracow-Warsaw railway line. The results of those experiments proved that data in the form of point cloud were not always sufficient for a precise 3D model reconstruction. Supplementing point clouds with photogrammetric images seems to be the best solution.

Detection and Recognition of Selected Class Railway Signs

Abstract The paper aims at presentation of results of research on detection and recognition of selected class railway signs (W11p). When conducting the research, the authors have proposed their own algorithm, which achieved about 90% effectiveness at detecting W11p signs and 98% effectiveness at classifying them. The processes of localisation, segmentation and recognition of W11p signs were considerably simplified thanks to the application of backpropagation neural network. The authors believe that two non-standard methods related to the use of the network deserve attention: the application of an interactive method of generating the training set, owing to which also pixels highly diversified in terms of their colours could be included, and the use of a full spectrum of neural network responses, which made it possible to accomplish a feedback. It consisted in an automatic adjusting of the network responses’ threshold to the results of segmentation and recognition.

Attempts at Automatic Detection of Railway Head Edges from Images and Laser Data

Abstract The more and more high resolution of aerial and ground images, as well as high density of laser data cause that they are more and more widely applied in many engineering projects. Given the current technical parameters, it is also possible to map railway infrastructure not only from the ground level but also from airborne locations (photogrammetry, laser scanning). Testing the usefulness of those data in obtaining information about railway infrastructure, and in particular, in detecting rail heads has been a subject of research of this paper authors. The paper presents results of experiments, consisting in verification of existing solutions and testing own algorithms for an automatic extraction of railway rail heads. The tested algorithms of object detecting and locating produced preliminary, satisfying results. The authors believe it to be reasonable to continue their research work.

Integration of Digital Images and Laser Scanning Data for the Survey of Architectural Objects

Abstract The goal of integration is to utilize a spatial model of an object obtained from the cloud of points and, possibly, to support it with a photogrammetric model in those places, where scanner measurements may be hard to perform, or even impossible. Since the image offers a possibility of texturing objects obtained most often from digital metric images, integration makes it possible to create a model of a view resembling the actual one. Another advantage of a photogrammetric model is the possibility of obtaining object edges with high accuracy (due to, among other things, the application of sub-pixel algorithms). It is just that aspect which, in relation to scanning, is described in the paper. The correctness of locating objects, which have been extracted from points, is tested by placing them on a photogrammetric model. In the proposed method, the authors tested extraction of objects, which was based on methods using the first and the second derivatives of a digital image. Those algorithms had been developed and tested earlier in many engineering projects that were based on a multilevel analysis of digital images. Research works were performed on several selected test facilities, located in the open-air ethnographic museum at Tokarnia, Poland.