Search

The measurement and control gate, as a new type of measurement and control equipment, has been widely used for water quantity control in irrigation areas. However, there is a lack of methods for calibrating the flow inside the measurement box at present. This paper establishes a flow prediction model based on a back propagation (BP) neural network and its optimization algorithm by using 450 sets of sample data obtained from the indoor gate overflow test and verified the effectiveness and accuracy of the prediction model by using another 205 sets of sample data. The results show that the gate flow prediction model based on a BP neural network and its optimization algorithm has self-adaptability to different flow patterns, and its prediction accuracy is significantly higher than that of the traditional water measurement method. Compared to the unoptimized BP model, the BP model optimized by the genetic algorithm (GA) or particle swarm optimization (PSO) has higher prediction accuracy and better error distribution. Both GA and PSO algorithms can be used to optimize the initial weights and thresholds of the BP flow prediction model. However, by comprehensively analyzing the prediction accuracy, error distribution, and running time, the PSO algorithm has better optimization performance compared to the GA algorithm. The prediction model can provide a reference for flow rate calibration and the anomaly rejection of measurement and control gates in the irrigation area.

The emerging computer vision and deep learning technologies are being applied to the intelligent analysis of sports training videos. In this paper, a deep learning based fine-grained action recognition (FGAR) method is proposed to analyze soccer training videos. The proposed method was applied to indoor training equipment for evaluating whether a player has stopped a soccer ball successfully or not. First, the problem of FGAR is modeled as human-object (player-ball) interactions. The object-level trajectories are proposed as a new descriptor to identify fine-grained sports videos. The proposed descriptor can take advantage of high-level semantic and human-object interaction motions. Second, a cascaded scheme of deep networks based on the object-level trajectories is proposed to realize FGAR. The cascaded network is constructed by concatenating a detector network with a classifier network (a long-short-term-memory (LSTM)-based network). The cascaded scheme takes the advantage of the high efficiency of the detector on object detection and the outstanding performance of the LSTM-based network on processing time series. The experimental results show that the proposed method can achieve an accuracy of 93.24%.

Reliable and accurate carrier phase ambiguity resolution is the key to high-precision Global Navigation Satellite System (GNSS) positioning and application. With the fast development of modern GNSS, the increased number of satellites and ambiguities makes it hard to fix all ambiguities completely and correctly. The partial ambiguity fixing technique, which selects a suitable subset of high-dimensional ambiguities to fix, is beneficial for improving the fixed success rate and reliability of ambiguity resolution. In this contribution, the bootstrapping success rate, bounded fixed-failure ratio test, and the new defined baseline precision defect are used for the selection of the ambiguity subset. Then a model and data dual-driven partial ambiguity resolution method is proposed with the above three checks imposed on it, which is named the Triple Checked Partial Ambiguity Resolution (TC-PAR). The comprehensive performance of TC-PAR compared to the full-fixed LAMBDA method is also analyzed based on several criteria including the fixed rate, the fixed success rate and correct fixed rate of ambiguity as well as the precision defect and RMS of the baseline solution. The results show that TC-PAR could significantly improve the fixed success rate of ambiguity, and it has a comparable baseline precision to the LAMBDA method, both of which are at centimeter level after ambiguities are fixed.

With the rapid development of the satellite navigation industry, low-cost and high-precision Global Navigation Satellite System (GNSS) positioning has recently become a research hotspot. The traditional application of GNSS may be further extended thanks to the low cost of measuring instruments, but effective methods are also desperately needed due to the low quality of the data obtained using these instruments. Thus, in this paper, we propose the analysis and evaluation of the ambiguity fixed-rate and positioning accuracy of single-frequency Global Positioning System (GPS) and BeiDou Navigation Satellite System (BDS) data, collected from a low-cost u-blox receiver, based on the Constrained LAMBDA (CLAMBDA) method with a baseline length constraint, instead of the classical LAMBDA method. Three sets of experiments in different observation environments, including two sets of static short-baseline experiments and a set of dynamic vehicle experiments, are adopted in this paper. The experiment results show that, compared to classical LAMBDA method, the CLAMBDA method can significantly improve the success rate of the GNSS ambiguity resolution. When the ambiguity is fixed correctly, the baseline solution accuracy reaches 0.5 and 1 cm in a static scenario, and 1 and 2 cm on a dynamic platform.