Search

In order to solve the problems of low efficiency and dependability of human experience in shot hole depth recognition by manual interpretation of the massive drilling videos, this paper proposed a repetition counting framework from the perspective of representation object motion, and firstly studied and realized the automatic shot hole depth recognition by adopting the proposed framework by combining multiple deep learning-based models. Firstly, the periodic motion characteristics of the objects in the drilling video were analyzed, and the motion extraction model of the representative object was established based on YOLOv5 and Deep SORT algorithm to obtain the spatiotemporal sequence. Then, the sequence noise and redundant information were filtered according to the spatiotemporal relationship between the drill pipes and workers, to obtain a coherent, smooth, and clean motion sequence. On this basis, a drill pipe motion counting model was established by using the YOLOv5 network to detect the temporal boundary and repetition number of the drill pipe motion. Finally, combined with the pipe motion extraction model and counting model, the comprehensive reliable recognition method of shot hole depth was presented. The shot hole depth recognition trial was carried out on 1029 drilling videos from the Bureau of Geophysical Prospecting INC. The proposed method achieved 90.5% recognition accuracy (Acc) and 0.027 mean absolute error (MAE) with 64 frames per second (FPS) and 0.39 average recognition time ratio (R t). Therefore, the proposed repetition counting framework solved the problem of efficient and intelligent drilling depth recognition of massive drilling videos and thus provided a new idea for repetition counting. • A new repetition counting framework for shot hole depth recognition based on representative object motion was proposed. • The explicit features were added into the optimization of the framework components. • The multi-scale temporal correlation was transformed into process detection to handle the non-stationary period problem. • A sequence denoising method based on spatiotemporal relationship was proposed to handle the noise and redundant information. [ABSTRACT FROM AUTHOR]

Being influenced by the factors of water level difference, tide, shortening closure gap, surrounding terrain and riverbed topography, etc., both the variations of flow velocity and water level in the closure gap are markedly larger than those in the general river. The phenomenon will bring potential safety hazards to the reservoir dam under construction, and thus the measurements of flow velocity and water level in the closure gap become very important in guaranteeing the safety of the reservoir dike project. Because of the safety, the measuring accuracy and the instrument performance, the traditional measuring methods of water level and flow velocity are less than ideal. In order to solve the problem, a buoy-based synthetic measuring system, which includes buoy body, balance weight, GPS (Global Positioning System), MRU (Motion Reference Unit), compass and computer, is developed. Besides, a complete data processing method, which includes the data synchronism, attitude correction and quality control of the measuring data, the self-adaptive extraction of water level and the calculation of flow velocity, is put forward in this paper. The system and the method were used for the project of Qingcaosha reservoir closure, and achieved the water level with a bias of less than 7 cm and the standard deviation of ±3.7 cm, and the flow velocity with a bias of less than 7 cm/s and the standard deviation of ±3.8 cm/s, satisfied the accuracy requirements of monitoring water level and flow velocity, and solved the measurement problem in the project. [ABSTRACT FROM AUTHOR]

To improve the underwater positioning precision of the traditional underwater positioning method that only uses one-way sound propagation, the paper proposes a new underwater positioning method by considering the location difference between transmission and reception of sound waves in circle-sailing positioning. In the method, firstly a fast calculation of the incidence angles was given under the depth constraint. And then, a precise positioning method was put forward by correcting the one-way travel time and considering the process of transmitting and receiving sound waves to double the observations to improve the underwater positioning precision. Two data sets observing the identical transponder were used to test the validity. The experiments verified the proposed method and the results showed that the precision of underwater positioning was improved considerably, especially in the vertical direction. The proposed method achieved the more uniform horizontal locations than the traditional method. • Fast calculation method of incidence angles under depth constraint. • Separated incidence angles and one-way travel times in a round trip. • Double the number of observation equation. • Improve the precision of the seafloor transponder. [ABSTRACT FROM AUTHOR]

The integration of GPS compass, GPS differential technique and ADCP effectively eliminates the impact of magnetic interference and bottom tracking bias due to bedload movement, well improves the accuracy and widens the applications of ADCP measurement. However, an accompanying problem, the heading misalignment angle α between the ADCP's Y axis and external compass will be introduced into ADCP measurement and brings a significant impact on the calculation of flow velocity. Therefore it becomes very necessary to determine and compensate α. Although the conventional method, the Course Method, has been widely used in the determination of α, it still needs to be improved in the operability, efficiency, robustness, self-adapting detection and accuracy. Thus, a robust method, Multi-epoch Method, is proposed in this paper. In this method, with the help of GPS velocity and bottom-tracking velocity, a α can be got at each epoch, then a series of αs can be obtained at a course. Then, the final α can be achieved through filtering and averaging the series of αs. The new method efficiently overcomes the drawbacks of the conventional method, and can conveniently achieve an accurate α. A synthetic experiment has implemented to testify the new method. A α of 6.22° with internal-verification accuracy of ±0.04° and external-verification accuracy of ±0.04° is determined by the new method, which causes an absolute flow velocity error of −0.31~0.29m/s and a relative flow velocity error of nearly 56% in 4 transects, while after the correction of α, both of them decrease to less than 0.02m/s and 3.3% respectively, which are meet with the requirement of ADCP measurement. [ABSTRACT FROM AUTHOR]