EVALUATION OF USER PREFERENCES WHEN CHOOSING CLOUD SOFTWARE PRODUCTS

The article analyzes the assessment of user preferences when choosing cloud software products (CSP). The peculiarity of the choice of OPP is not only the choice of an enterprise - supplier of CSP of the necessary functionality, but also the choice of enterprises-intermediaries of cloud digitalization, which include an enterprise for designing a cloud information system (CIS) based on the selected CSP and a cloud data processing center during the operation of the CIS. A multidimensional characteristic of cloud digitalization enterprises used in the selection process is given. Since the criteria for choosing an CSP are vague, fuzzy numbers, an apparatus of fuzzy sets is proposed for evaluating user preferences.

Three-Dimensional Target Detection Based on RGB-D Data

The current three-dimensional (3D) target detection model has a low accuracy, because the surface information of the target can only be partially represented by its two-dimensional (2D) image detector. To solve the problem, this paper studies the 3D target detection in the RGB-D data of indoor scenes, and modifies the frustum PointNet (F-PointNet), a model superior in point cloud data processing, to detect indoor targets like sofa, chair, and bed. The 2D image detector of F-PointNet was replaced with you only look once (YOLO) v3 and faster region-based convolutional neural network (R-CNN) respectively. Then, the F-PointNet models with the two 2D image detectors were compared on SUN RGB-D dataset. The results show that the model with YOLO v3 did better in target detection, with a clear advantage in mean average precision (>6.27).

Application of Point Cloud Data Processing in River Regulation

Geotextile materials are often used in river regulation projects to cut down sand loss caused by water erosion, to thus ensure a stable and safe river bed. In order to measure the overlap width in the geotextile-laying procedure, we proposed a point processing method for cloud data, which engages point cloud data obtained by 3-D imaging sonar to do automatic measurements. Firstly, random sampling and consensus point cloud segmentation and outer point filtering based on statistical analysis on density were used to extract the upper and lower plane data of the geotextile. Secondly, cluster classification was used to obtain the edge point cloud. Lastly, edge characteristic parameters were extracted by linear fitting, and the overlap width in geotextile laying was calculated. Results show that this measurement scheme is feasible, robust, and accurate enough to meet the requirements in real-life engineering.

FORTLS: An R Package for Processing TLS Data and Estimating Stand Variables in Forest Inventories

Terrestrial Laser Scanning (TLS) enables rapid, automatic, and detailed 3D representation of surfaces with an easily handled scanner device. TLS, therefore, shows great potential for use in Forest Inventories (FIs). However, the lack of well-established algorithms for TLS data processing hampers operational use of the scanner for FI purposes. Here, we present FORTLS, which is an R package specifically developed to automate TLS point cloud data processing for forestry purposes. The FORTLS package enables (i) detection of trees and estimation of their diameter at breast height (dbh), (ii) estimation of some stand variables (e.g., density, basal area, mean, and dominant height), (iii) computation of metrics related to important tree attributes estimated in FIs at stand level, and (iv) optimization of plot design for combining TLS data and field measured data. FORTLS can be used with single-scan TLS data, thus, improving data acquisition and shortening the processing time as well as increasing sample size in a cost-efficient manner. The package also includes several features for correcting occlusion problems in order to produce improved estimates of stand variables. These features of the FORTLS package will enable the operational use of TLS in FIs, in combination with inference techniques derived from model-based and model-assisted approaches.