Yuhei Yamamoto

Road administrators perform traffic censuses for road maintenance and management. In these censuses, surveyors visually count the number of passing vehicles, labor-saving method is necessary because the surveyors must count in many places and it is time-consuming. Mechanical surveys have been introduced as a countermeasure. However, considering the growth of ICT, further development in survey technology is expected. In the field of urban transportation, various technologies are suggested to recognize vehicles from video data using deep learning. For putting it to practical use and spreading some methods widely, it is necessary to survey the practicality of technologies and proper conditions of photographing based on the existing studies. In this research, we developed a technology of recognizing vehicles based on existing methods and took videos under various conditions. By analyzing them using the present technology, we clarified the usefulness of deep learning in the proper conditions of photographing.

In Japan, the Ministry of Land, Infrastructure, Transport and Tourism is promoting a policy called i-Construction for increased productivity, safety, and cost savings. Under this policy, construction machinery with machine control and machine guidance functions based on information and communication technology is used at construction sites. For cranes, technology has been developed in which a camera installed at the jib acquires images of the suspended load and its surrounding features. However, it is not possible to measure the size and height of the suspended load and surrounding features using this technology, because it acquires only an overhead image. This study addresses this problem by developing a visualization system for the suspended load area in order to support crane operator work. The system visualizes the size and height of the suspended load and its surrounding features based on three-dimensional point cloud data using a laser scanner, an inertial measurement unit, and the camera installed at the jib. The system was evaluated in terms of usefulness and reliability through experiments.

It is expected that point cloud data generated by UAV photogrammetry will be utilized in various management scenes such as construction management and disaster. On the other hand, in UAV photogrammetry, there are problems that require enormous time to generate point cloud data and that can not be measured except in bright daytime hours. For these reasons, laser scanner equipped UAV capable of measuring point cloud data at night and generating them in less time than UAV photogrammetry have been developed. However, since the accuracy of generation of point cloud data with the laser scanner equipped UAV varies depending on the conditions of the construction fields and the measurement method, it is necessary to realize the accuracy under various conditions. In this research, the measurement error of the UAV mounted laser scanner by the environment and measurement method are investigated and clarified at the viewpoint of the influence on the generation accuracy of each error factor in the sites of construction management. And the accuracy of generation of point cloud data with measurement method that takes measurement error into account is clarified.

Our country is promoting supporting activities for training players and improving the competitiveness in preparation for international sporting events including Tokyo Olympic Games. Especially recently, one of the supporting activities at present is to utilize high-performance sensing devices for analyzing tactics by measuring the positional data of the players and digitizing their performances. However, sensing devices have not been put to practical use yet because it is hard for the players to wear them during games of various kinds of sports. Therefore, technologies for identifying players by image processing from the video images are the mainstream of the sports information fields. However, they have a problem that the identification accuracy decreases because lots of players are occluded into image data. Thus, in this research, we use multi-cameras from single viewpoint to identify American football players with high accuracy and specify their positions by photographing occlusion points with high zoom magnification, and then applying deep learning of the OpenPose. Finally, we embed the identification results of the OpenPose to the video images taken with low zoom magnification and try to identify players and to analyze their positions.

A large number of highway bridges built during the period of high economic growth are now aging simultaneously, requiring appropriate maintenance. However, the design drawings and completion drawings have been disposed of for some of these highway bridges, or they don't agree with the present conditions. As this makes maintenance difficult, recovery of as-built drawings is required. For this problem, the authors have attempted automatic generation of road alignment drawings by measuring overpasses using MMS. However, the proposed method only verifies the accuracy of the drawing generated using one type of MMS's point cloud data. But, We did not apply the method to others of MMS's point cloud data which have greatly different properties such as the type of laser scanner or the direction of installment. This research verifies the versatility of the proposed method by comparing the result of extracting feature points of the overpass and the accuracy of road alignment generation using two types of MMS's point cloud data having greatly different properties from each other.

In the fields of maintenance of social infrastructure, safety and disaster prevention, there are various studies about space measurement methods using a various equipments to grasp the situation of ground surface and structures. Considering this trend and i-Construction proposed from Ministry of Land, Infrastructure and Transport in 2016, methods is demanded to measurement three dimensional space might quickly be high preciseness. The authors focused on the measurement method by UAV-Based Laser Scanner. The principal aim of this study was to find out the development subject of UAV-Based Laser Scanner by surveying existing researches, techniques and patents of space measurement methods.

Many highway bridges constructed during the high economic growth period have become deteriorated. So their maintenance is required. However, they are facing difficulties in maintenance because their drawings prepared at the time of design or completion were made in paper media, many of which have already been disposed. There is a research that attempts to automatically generate SXF (Scadec data eXchange Format) drawings of a highway bridge to use for its maintenance, using point cloud data obtained by MMS (Mobile Mapping System). In the previous research, the authors use positions of joints on a bridge to divide a point sequence into each span and generate their alignments of the highway bridge. However, when it is unable to extract a joint, or when the start and end points of the alignment are placed at a short distance from the positions of joints on a design drawing, the generated clothoid curve will lose continuity with straight line and circular arc. Thus, this research proposes an analysis method for calculating the start and end points of straight lines, clothoid curves, and circular arcs in a plane and straight lines and quadratic curves in a longitudinal section to allow each alignment to connect with each other naturally. This allows generating the SXF drawings of the superstructure on a highway bridge with high precision without considering the positions of joints.

A large number of posts with useful information to help local residents can be found in CGM (Consumer Generated Media). Such posted messages contain, for example, information about prevention of crimes or emergency responses during disasters for the local communities. Various information about diverse regions is put together and open to the public. However, since anyone can contribute to CGM, it often happens that a mixture of information about multiple regions can be found on the same Web page or that different topics are covered for the same region, making it hard to extract only the information about the particular region or it's contents. This research proposes a method for extracting regional information from CGM by dividing the information circulating in CGM into posted messages, and then grouping them according to the region or it's contents that each posted message treats. It also examines a possibility of creating a regional information map by analyzing the positional information contained in the posted messages classified into groups and mapping them onto GIS.

In the civil engineering construction industry, two or more companies often form a joint venture in alliance with each other in constructing a large-scale structure. However, in order to create a corporate alliance, it is necessary to select appropriate companies by analyzing diverse pieces of information including their alliance records and technologies. Thus studies are made on methods for evaluating candidate partner companies based on the analysis results of the link structure of Web pages as well as of the web contents through natural language processing. However, the existing method has a problem in that the relationship between companies are not taken into consideration and consequently, therefor it is not possible to take the advantages of alliance into consideration. This study evaluates relationship between companies based on the technology division of the company obtained from its project achievement and the collaboration achievement between companies. Then, validity of the classification results is verified by classifying companies by visualizing based on the evaluation results, and comparing them with the past achievement of forming joint ventures.