Patterns on manhole covers prevent the slippage of pedestrians and vehicles that go over them. Owners of manhole covers inspect and repair manhole covers when their patterns wear down due to vehicle traffic. It is necessary to introduce efficient inspection methods and improve wear resistance of the patterns on the large number of manhole covers in Japan. We proposed an efficient method that uses image processing to inspect manhole covers. However, since the patterns are only a few millimeters deep, high-resolution images must be captured by placing the manhole cover with high resolution equipment.
 We propose a novel pattern for manhole covers. The wear of this highly durable pattern is easy to judge in low resolution images. The shape of the pattern changes as wear increases. We experimented to confirm the pattern's high durability and proposed an algorithm that uses Hough transformation processing to detect wear in patterns.

View full abstract

 In this study, the estimation method of fatigue damage accumulation with wireless accelerometer for lighting pole was investigated. The acceleration and stress measurement for the lighting pole vibrated by vortex induced vibration was carried out. The results showed that the fatigue damage seemed to be cumulated at all times. Furthermore, the proposed estimation method of fatigue damage accumulation based on the acceleration measurement was proposed. By applying the proposed method, the damage accumulation was almost same as that obtained by the stress measurement.

View full abstract

 The bedrock that supports civil engineering structures and the natural ground excavated in civil engineering projects are widely distributed in the densely solidified rock conditions to the non-solidified soil conditions. In addition, the banking structures for constructing facilities are generally developed with natural materials, such as rocks, soil, and sand. In the construction work covering these facilities, an in-situ method for evaluating rock quality and soil property rapidly, conveniently, and reliably is required. In this situation, we developed an in-situ method for evaluating deformation characteristics by colliding a metallic sphere attached with an accelerometer with an evaluation object and applying the Hertz theory (elastic collision theory) obtained from response characteristics.

 New ideas and some assumptions are added to the Hertz theory, which is the isotropic elastic theory, and this concept has originality in applying it to the ground material which is elastoplastic in general. Moreover, it can meet the needs of multipoint measurement which has been increasing in recent years. Hence, it can also be centrally managed in the three-dimensional BIM/CIM model in cooperation with GNSS, and then the information data on the quality of the construction is quickly and efficiently implemented as a part of design and construction support.

View full abstract

 This paper proposes real-time monitoring of flood inundation in urban area using LPWA (Low Power Wide Area) to prevent or mitigate flood damage. In order to verify that this concept is feasible, we evaluated reliability of radio transmission, reliability of detecting water level, and energy consumption though eight prototype LPWA water detectors driven by dry cells, six settled on river dike and two in drainage around Yokohama Station. The detector transmits a signal when its float type level sensor detects that water exceeds a specified level. As a result, reliability of transmission was 99.94% for 15105 times trying to send signals for 131 days, and fatal voltage reduction of battery was not founded. Water level at river around Yokohama Station correlates closely with tide. Height of four of six devices on river dike were settled between high and low of tide level so as to evaluate reliability of water detection efficiently. Sensitivity of water detection of these devices varied from 76.6% to 86.6% depending on device. A sensor in a drainage was able to detect raised water level by typhoon. However, it also detected raised water level of which cause was unknown. To improve reliability of water detection is therefore seen as a key future priority.

View full abstract

 In the periodic bridge inspection in Japan, engineers are required to determine the damage level for each structural member by visual inspection in all of seven hundred thousand bridges. Deep learning is expected to realize a future more sustainable and low cost bridge inspection system. In this study, authors constructed convolutional neural network (CNN) that determines the damage level of each bridge member from image data, and discussed its applicability to the bridge inspection system. The image data of bridge members, that were acquired in the inspection work in Yokohama-city, Japan, were used in the verification. The CNNs for bridge main girder, slab, and bearing were constructed with Python and Chainer library. Two classification CNNs with recognition accuracies of 75% in main girder, 70% in slab, and 85% in bearing could then be constructed. However, over-training was observed in those CNNs, and it was also considered that the number of classifications and the design of training data set significantly affected the CNN performance. Questionnaire survey was additionally conducted to verify the acceptance of CNN determination outputs from engineers, who were working on the damage level determination in the bridge inspection. Here, twenty-two engineers were participated. In the results, the acceptance of all of three CNNs were not high; however, it could be considered that the acceptance of CNN determination was also affected by the uncertainty of determination in visual inspection. Moreover, the engineers indicated that the damage level was determined not only by the damage condition itself, but also by its location, bridge type, and surrounded environment. It was shown that there was the potential to improve the performance and applicability of CNNs for damage level determination by converting the knowledge of bridge engineers to the configuration of training data sets.

View full abstract