Loading...

Table of Content

    25 June 2024, Volume 52 Issue 6
    2024, 52(6):  0. 
    Asbtract ( 27 )   PDF (46KB) ( 116 )  
    Related Articles | Metrics
    Green & Intelligent Transportation
    HU Xizhi, CUI Bofei, WANG Qin, et al
    2024, 52(6):  1-11.  doi:10.12141/j.issn.1000-565X.230262
    Asbtract ( 172 )   HTML ( 3)   PDF (5307KB) ( 46 )  
    Figures and Tables | References | Related Articles | Metrics

    With the development of autonomous driving technology, visual simultaneous localization and mapping (SLAM) technology has attracted more and more attention. In the memory parking scene, it is necessary to establish a prior map of the parking lot scene. Thus, when the car enters the same parking lot again, visual SLAM can help to construct and locate the scene. In order to improve the robustness, accuracy and efficiency of the map built by SLAM, first, a lightweight deep learning algorithm is used to improve the poor robustness of the traditional feature extraction algorithms in different scenarios, and the deep separable convolution is adopted to replace the previous common convolution structure, which greatly improves the time efficiency of feature extraction. Next, the Patch-NetVLAD algorithm is improved based on ResNet network, and the improved residual network as well as the original VGG network is retrained on MSLS data set. Then, image retrieval is used for rough positioning, candidate image frames are selected, and camera pose is solved by fine positioning to complete global initialization relocation. On this basis, the improved bag of words algorithm is used to retrain the images in different parking lot scenes, and all the algorithms are transplanted into the OpenVSLAM architecture to complete the mapping and positioning of the actual scene. The experimental results show that the proposed visual SLAM system can complete the construction of many scenes such as aboveground, underground and semienclosed parking lots, with an average longitudinal positioning error of 8.42 cm and an average horizontal positioning error of 8.30 cm, both of which meet the engineering requirements.

    SUN Jian, WU Jiyan, LI Zheng, et al
    2024, 52(6):  12-23.  doi:10.12141/j.issn.1000-565X.230210
    Asbtract ( 147 )   HTML ( 4)   PDF (2942KB) ( 25 )  
    Figures and Tables | References | Related Articles | Metrics

    At present, the market share of ridesharing in China is relatively low, and there is still huge potential to be fully tapped in alleviating traffic congestion and reducing energy consumption and emissions. Incentive-based traffic demand management strategies can promote people’s willingness of ridesharing, but the design of incentive schemes is highly correlated with the market penetration of ridesharing. An unreasonable incentive scheme may easily lead to increased cost or even project failure. In order to further stimulate the potential ridesharing demand and make reasonable use of transportation resources, a road segment-based incentive optimization model for ridesharing is proposed, which uses ridesharing as the fulcrum and rewards as the lever to reduce the total social travel cost. The upper model of the proposed model aims to find the optimal incentive scheme to minimize the total social travel cost, and the lower model is a user equilibrium allocation model of ridesharing vehicles and single-driver vehicles under the incentive scheme. The iterative algorithm combining the genetic algorithm and the Frank-Wolfe algorithm is used to solve the upper and lower models, respectively, and the feasibility and effectiveness of the model are verified by using the Sioux Falls and Nguyen Dupuis transportation networks as numerical examples. The results show that budget investments that do not meet market penetrations may result in a significant increase in total social travel costs; and that, under the optimal incentive scheme, the total social travel cost is reduced by about 24.53%, the congestion of 50% of congested links is alleviated, and the fairness issue in traffic demand management is alleviated to a certain extent. Thus, there comes to the conclusion that the proposed model can provide theoretical support for managers to set up scientific and reasonable incentive-based ridesharing schemes.

    LONG Xueqin, WANG Han, WANG Ruixuan
    2024, 52(6):  24-33.  doi:10.12141/j.issn.1000-565X.230375
    Asbtract ( 195 )   HTML ( 4)   PDF (4721KB) ( 29 )  
    Figures and Tables | References | Related Articles | Metrics

    In order to further clarify the differences in routing behaviors of different taxis, this paper adopted the method of frequent sequence mining to extract the frequent path between the same OD pairs, construct path sets, and analyze the similar characteristics of path sets from static and dynamic perspectives. By taking the trajectory data of taxis in Xi’an City as the research object, the path set between OD pairs is obtained through grid division and road network matching. Then, the frequent path is redefined, the PrefixSpan evolution algorithm is adopted, and the dynamic threshold and frequency index based on the obtained frequent subsequences are introduced to mine frequent paths. Furthermore, in order to complete the construction of three kinds of effective path sets, the shortest path and other paths are extracted, and the general properties of the constructed path sets are analyzed. Finally, the similarity between two-dimension time series (tracks) on the path is represented as dynamic similarity, and the similarity between one-dimension directed sequences (sections) is represented as static similarity, and the similarity analysis of three types of paths is carried out based on the improved longest common subsequence and dynamic time regularity algorithm. The results show that: (1) the similarity between the frequent path and the shortest path is rather high, meaning that most taxis still choose the road with the lowest travel time but not the shortest path; (2) time and distance are still the main considerations for travelers when choosing a path, but travelers do not completely pursue the shortest time or distance; (3) the calculated dynamic similarity is significantly higher than the static similarity, which means that the two-dimension sequential similarity on the path is higher than the one-dimension shape similarity; and (4) the two proposed methods both possess the highest similarity between the frequent path and the shortest path and the lowest similarity between the shortest path and other paths The consistency of the comparison results indicates that the similarity of the static path can be roughly measured by the that of the dynamic trajectory. The proposed frequent path mining algorithm is of certain reliability. It can provide supports for urban traffic managers with recommend routes and planed roads.

    YUAN Renteng, WANG Chenzhu, XIANG Qiaojun, et al
    2024, 52(6):  34-44.  doi:10.12141/j.issn.1000-565X.230258
    Asbtract ( 304 )   HTML ( 7)   PDF (2570KB) ( 46 )  
    Figures and Tables | References | Related Articles | Metrics

    In order to accurately recognize and estimate the lane-changing intentions of vehicles, a vehicle lane change intention recognition model based on TCN-LSTM network is proposed, which combines the temporal processing capability of TCN (Temporal Convolutional Network) with the gate memory mechanism of LSTM (Long Short Term Memory Network). In the investigation, firstly, the driving intentions of the target vehicle are divided into three types, namely going straight, changing lanes to the left, and changing lanes to the right. The running state indicators of the target vehicle and its surrounding neighboring vehicles (including the adjacent front and rear vehicles in the same lane, left lane and right lane) are extracted from the Citysim vehicle trajectory dataset using the median filtering algorithm. Secondly, to overcome the low recognition accuracy, long training time and slow parameter updating existing in statistical theories and traditional machine learning methods, the dilated convolution technique is used to extract the temporal features of time series, and the gate memory units are used to capture the long-term dependency relationships of temporal features. With 54 indicators, including the speed, acceleration, heading angle, heading angle change rate, and relative position information of the target vehicle and surrounding neighboring vehicles, as input parameters, and with the lane change intention of the vehicle as the output indicator, a vehicle lane -change intention recognition model based on the TCN-LSTM network is constructed. Finally, the recognition accuracy of TCN, SVM (Support Vector Machines), LSTM, and TCN-LSTM models under different input time steps are comparatively analyzed. The results show that, when the input time series length is 150 frames, the recognition accuracy of the TCN-LSTM model reaches a maximum of 96.67%; and that, in terms of overall classification accuracy, as compared with LSTM, TCN and SVM models, the TCN-LSTM model improves the classification correctness of lane change intention by 1.34, 0.84 and 2.46 percentage points, respectively, which demonstrates better classification performance.

    WANG Yongxing, BI Jun, XIE Dongfan, et al
    2024, 52(6):  45-55.  doi:10.12141/j.issn.1000-565X.230359
    Asbtract ( 201 )   HTML ( 3)   PDF (1741KB) ( 185 )  
    Figures and Tables | References | Related Articles | Metrics

    As the existing layout schemes of static wireless charging (SWC) facilities often neglect battery degradation costs, this paper proposes a layout optimization method of SWC facilities for electric buses by considering battery degradation characteristics. Firstly, by considering the operation characteristics of electric buses under opportunity charging mode, a layout optimization method of SWC facilities is developed with simultaneous consideration of charger deployment costs and battery degradation costs, with the function mechanism of battery state of charge (SOC) variety ranges on battery degradation rate being integrated into the model, and with the accumulated energy consumption constraints being introduced in the model to ensure that the SWC layout scheme can satisfy the bus route operation demands. Then, an improved TS (Tabu Search) algorithm is presented to solve the model by overcoming its computational complexity, and the initial solution and neighborhood structure of the algorithm are constructed according to the model characteristics. Finally, a numerical example is designed to verify the model and algorithm. The results indicate that the layout of SWC facilities has significant effects on battery degradation; that the proposed model can reduce 3.8% of the total annualized cost, as compared with the conventional model that neglects the battery degradation characteristics; that the battery degradation cost accounts for up to 72.3% of the total annualized cost under current battery technology and cost conditions; and that the improved TS algorithm is better than the original one because it significantly improves the solution efficiency. Moreover, a sensitive analysis is conducted to explore the impacts of multiple critical factors on optimal results, finding that both the upper bound of battery SOC and the SWC facility charging power have significant negative correlation with the total annualized cost, while the battery’s unit capacity cost, the SWC facility layout cost and the vehicle energy consumption rate all have positive correlation with the total annualized cost in various degrees.

    LEI Cailin, ZHAO Cong, LOU Ren, JI Yuxiong, DU Yuchuan
    2024, 52(6):  56-72.  doi:10.12141/j.issn.1000-565X.230314
    Asbtract ( 121 )   HTML ( 4)   PDF (5581KB) ( 128 )  
    Figures and Tables | References | Related Articles | Metrics

    Roadside sensors have been widely installed on highways to collect full-sample real-time vehicle trajectory data, which supports full time and space control of traffic flow, microscopic driving behavior analysis, etc. However, rapid evaluating data quality is a challenge for management departments. Data quality evaluation methods in previous studies are limited to complicated operation and single dimension, which cannot meet the quality evaluation requirements of real-time vehicle trajectory data in dynamic traffic flow. In order to rapidly evaluate the quality of vehicle trajectory data from roadside millimeter-wave radar, a data quality evaluation method is proposed through mining the information of data. First, based on the typical errors of the measured trajectory data, 9 secondary evaluation metrics are established from four perspectives, including trajectory completeness, consistency, accuracy, and validity. Then, the comprehensive metric is calculated based on the CRITIC weighting method. Finally, an empirical analysis is conducted based on the vehicle trajectory data (3 549 in total) obtained by millimeter-wave radars in four different scenarios. The results show that the installation type and model of the millimeter-wave radar obviously influence the quality of vehicle trajectory data., and that the proposed evaluation method can distinguish the quality differences of vehicle trajectory data effectively. This study provides a support for the short-term performance decay monitoring and the type selecting of roadside sensors. Also, it gives a reference for improving the quality of vehicle trajectory data.

    Mechanical Engineering
    LI Xiao, LI Yapeng
    2024, 52(6):  73-80.  doi:10.12141/j.issn.1000-565X.230398
    Asbtract ( 85 )   HTML ( 3)   PDF (2894KB) ( 14 )  
    Figures and Tables | References | Related Articles | Metrics

    In view of the lack of effective treatment for neurogenic and myogenic bladder, a new solution, namely an artificial bladder detrusor system, was proposed from an engineering perspective. Based on the shape memory effect of shape memory alloy (SMA) springs, a system structure consisting of a wireless power transfer module, a control module, a feedback module and an executive module was designed to realize the assisted urination in accordance with human urodynamics. A finite element model of human bladder was established, and the storage process as well as the assisted urination process of the human bladder was simulated and analyzed. Based on the simulation results and the mathematical model of SMA springs, the structural parameters of SMA spring actuator were optimized. Then, the temperature-free height equation of SMA springs was derived according to experimental data. Furthermore, by combining with the thermodynamic formula and the spring mathematical model, the system control equation was derived, and an open-loop control strategy of the system was proposed on this basis. Finally, based on the feedback module of the system, a proportional integral differential (PID) closed-loop control strategy was designed, and a simulation experiment platform was constructed to study the urine flow rate characteristics of the system. The results indicate that the principle of the system is feasible, and that the assisted urination process is continuously controllable. Under different urine volumes, the two control strategies can both achieve assisted urination in accordance with human urodynamic principles. This research can provide guidance for the design of clinically applicable artificial bladder detrusor system and also offer a reference for the design of other implantable devices using SMA springs as actuators.

    MO Haijun, LIANG Daoming, LIN Chenbin, LIU Xiang
    2024, 52(6):  81-88.  doi:10.12141/j.issn.1000-565X.230360
    Asbtract ( 112 )   HTML ( 6)   PDF (2063KB) ( 32 )  
    Figures and Tables | References | Related Articles | Metrics

    Rapid mold manufacturing can be used to print wax molds for investment casting, shorten the production cycle and improve the production efficiency. However, in the actual forming process, due to the uneven temperature distribution in different positions of the workpiece, the internal stress will be different, which may result in warping deformation, and then have a significant impact on the forming quality of the workpiece. Moreover, due to the constraints of forming parameters such as 3D printing speed and deposition layers’ number, it is difficult to reduce the degree of warping deformation of the workpiece and improve the forming efficiency simultaneously. To solve this problem, this paper establishes a mathematical model for the warping deformation of formed parts, and combines experimental design and mathematical calculation methods to explore the influence mechanism of printing speed on the degree of warping deformation and printing efficiency of casting wax direct writing. Experimental results show that, at a certain printing speed, the warpage value decreases with the increase of the number of deposition layers, while gradually increases with the continuous increase of printing speed; and that the higher the printing speed, the closer the printing time of different samples is to a certain stable value, which means that the impact of printing speed on the forming efficiency decreases with the increase of printing speed. In addition, by assigning weight coefficients respectively to the forming warping deformation and the printing efficiency, a continuous function model for the optimal printing speed of surface contours is established, and the effectiveness of the model is verified. The results show that the continuous function model of optimal printing speed based on the warping deformation of casting wax can simultaneously reduce the warping deformation and improve the printing efficiency.

    LIU Chenyu, WANG Anlin, LI Xiaotian, LIU Jiaming, LI Xiaoxiang
    2024, 52(6):  89-98.  doi:10.12141/j.issn.1000-565X.230457
    Asbtract ( 77 )   HTML ( 3)   PDF (3339KB) ( 33 )  
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the durability problem of electromagnets for proportional valves of construction machinery, in order to improve the resistance of electromagnets to thermal failure under random load conditions, a parametric redesign model of proportional electromagnets was proposed based on multi-physical field coupling theory and robust optimization theory. By taking a proportional electromagnet with basin-type suction structure as the research object, the effectiveness of the proposed parametric model was verified through steady-state electromagnetic test and temperature distribution test. Under the premise of ensuring the accuracy of electromagnetic calculation, the parameters such as magnetic conductivity and heat transfer with fuzzy magnitude in the system were calibrated. With the main structural parameters of electromagnet and coil as control factors, and with the random error of wire diameter of coil enameled co-pper wire caused by the uncertainty of production process as noise factor, orthogonal tests were designed based on Taguchi method, and the evaluation function of the thermal robustness redesign of multi-factor weighted proportional electromagnet was defined. Then, with the thermal load of the proportional electromagnet obtained in the field test of the excavator as the response calculation heat source, the redesign of the key structural parameters with the minimum system response variation under noise disturbance was carried out, under the constraint of allowable temperature rise that does not cause the coil insulation failure. The results show that the coil length and the number of turns are the main factors affecting the thermal robustness of the electromagnet, and that the coil window shape determined by the winding process determines the magnetic permeability and heat transfer capability of the system. The thermal robustness redesign method of proportional electromagnet proposed in this paper is of engineering reference value for the custo-mized design of electromechanical products under magneto-thermal coupling.

    YU Ruiming, WU Yefei, WU Bing, SHEN Yuzhong
    2024, 52(6):  99-109.  doi:10.12141/j.issn.1000-565X.230464
    Asbtract ( 75 )   HTML ( 3)   PDF (5261KB) ( 25 )  
    Figures and Tables | References | Related Articles | Metrics

    Aiming at the problem that the unbalanced force and sealing force under high-temperature and high-pressure conditions are difficult to reach the required leakage level, a double spool pilot cage type control valve was designed, in which a large spool and a small spool work together in sections and the medium pressure in front of the valve is fully utilized to increase the sealing force of the valve seat. Then, the calculation equation of unbalanced force, the flow equation and the flow characteristics of the valve were analyzed, the fluid flow was simulated, and the pressure loss after the fluid flows through the control valve was analyzed. Moreover, the fluid flow in the inner cavity of the large valve spool was simulated, and the pressure before the valve, the temperature in the inner cavity as well as the sealing force was analyzed. The motion simulation of large and small spools was also carried out, and the driving force of the actuator, the axial resultant force of the valve spindle and the speed of the large spool were all discussed. The results of valve body’s pressure-tight test and valve seat’s leakage test of normal-temperature medium show that there is no visible leakage and deformation in the valve body; and that the measured leakages of three tests are respectively 1, 1 and 2 drops per minute when the set leakage is 10 drops per minute. In addition, the results of valve seat’s leakage test of high-temperature medium show that, when the set leakage is 1.275 mL/min, the measured leakages of three tests for 250 ℃ high-temperature steam are respectively 0.11, 0.11 and 0.13 mL/min, while those for 300 ℃ high-temperature steam are respectively 0.19, 0.19 and 0.22 mL/min, finding that all the measured leakages are less than the set ones. The double valve spool pilot cage structure proposed in this paper enhances the sealing force of the regulating valve, and, the pressure strength of the valve body and the sealing performance of the valve seat both meet the use requirements.

    Computer Science & Technology
    HU Yongjian, ZHUO Sichao, LIU Beibei, WANG Yufei, LI Jicheng
    2024, 52(6):  110-119.  doi:10.12141/j.issn.1000-565X.230105
    Asbtract ( 97 )   HTML ( 3)   PDF (2582KB) ( 36 )  
    Figures and Tables | References | Related Articles | Metrics

    Most existing Deepfake face forgery detection algorithms suffer from the insufficient generalization performance despite that their intra-dataset detection performance is fairly good. This is because these methods mainly rely on local features that are prone to overfitting, which leads to unsatisfactory cross-dataset detection performance. In order to solve this problem, a face forgery detection method based on multi-scale spatiotemporal features and tampering probability is proposed, which helps to maintain good performance for cross-dataset testing, cross-forgery testing as well as video compression by detecting the inevitable temporal inconsistency between continuous frames in deepfake videos. The proposed detection method consists of three modules: a multi-scale spatiotemporal feature extraction module is employed to reveal the discontinuous traces of fake videos in the temporal domain, a three-dimension dual-attention module is designed to adaptively compute the correlation between multi-scale spatiotemporal features, and an auxiliary supervision module is used to predict the tampering probabilities of randomly selected pixels to form a supervision mask. Then, the proposed algorithm is compared with the baseline algorithm and the latest relevant works on large-scale public standard databases such as FF++, DFD, DFDC and CDF. Experimental results have show that the proposed algorithm has the best overall performance for cross-dataset testing and video compression, and has the above-average performance for cross-forgery testing. Meanwhile, it maintains good average performance for all intra-dataset testing. All the experiments demonstrate the effectiveness of the proposed algorithm.

    LI Song, WANG He, ZHANG Liping
    2024, 52(6):  120-127.  doi:10.12141/j.issn.1000-565X.220728
    Asbtract ( 129 )   HTML ( 4)   PDF (1627KB) ( 20 )  
    Figures and Tables | References | Related Articles | Metrics

    The skyline query in road networks has important application value in the fields such as intelligent transportation, point of interest discovery, and location services. In order to solve the problem of low efficiency of skyline queries in road network environment and the lack of privacy of query results, a differential privacy-based skyline query method in road network environment is proposed. In this method, first, aiming at the characteristics of large data amount and complex data in the initial dataset of road network environment, the dataset is preprocessed, and three pruning rules are proposed based on the properties of the skyline layer divided by distance attributes and the Voronoi diagram of the road network. Next, based on the pruning rules, a dataset pruning algorithm in road network environment is proposed, which can effectively filter out a large amount of redundant data. Then, for the filtered dataset, a storage method of grid index is utilized to save the storage space. Furthermore, a skyline extension tree based on grid index is designed, and an algorithm for querying global candidate skyline point sets is proposed based on the extension tree and the corresponding pruning rules. Finally, for the query result set, a differential privacy budget allocation model is employed to allocate privacy budgets, and a result set publishing algorithm based on information divergence is proposed, thus effectively improving the privacy of data information. Experimental results show that the proposed query method achieves a query accuracy of more than 99%. It improves the query efficiency by more than 10%, as compared with the traditional skyline query methods in larger datasets. When the total differential privacy budget is 0.01, 0.10, 0.50 and 1.00, the relative error of the proposed privacy budget allocation method is lower than that of the equal difference and equal ratio allocation methods.

    ZHANG Wendong, WU Ziwei, SONG Guochang, HUO Qingao, WANG Bo
    2024, 52(6):  128-137.  doi:10.12141/j.issn.1000-565X.230143
    Asbtract ( 114 )   HTML ( 3)   PDF (1115KB) ( 20 )  
    Figures and Tables | References | Related Articles | Metrics

    The named entity recognition of traditional Chinese medicine (TCM) classics is the basis for constructing TCM knowledge graph, and is of great significance for the extraction and intelligent presentation of TCM knowledge. However, the knowledge system of TCM has a huge structure, and the publicly available corpus is scarce and semantically complex. Most of the current researches focus on the expression of character vectors, and do not fully consider the rich semantic features in the structural characteristics of special Chinese characters. Moreover, due to the rich semantic meaning of Chinese characters, there are still problems of insufficient expression of the potential features and polysemy of one word. In this paper, a named entity recognition method based on SiKuBERT and multivariate data embedding is proposed by combining the corpus features of ancient Chinese medicine books with the structural information of ancient Chinese characters. In this method, the word feature information is created by SiKuBERT, and on this basis, word features and radical features are embedded to capture the semantic information of Chinese characters, so that characters with similar radical sequences can be close to each other in the spatial vector. Then, the method is used to identify the names of people, herbal medicines, diseases, pathologies, and meridians in the Materia Medica dataset. The experimental results show that the proposed method is able to effectively extract five types of entities in the text, with an F1 score of 86.66%, a precision rate of 86.95%, and a recall rate of 86.37%. As compared with the SiKuBERT-CRF model based on word features, the proposed method integrates the word information with the structural information of traditional Chinese characters, which enhances the entity recognition effect, and the overall F1 score is improved by 2.83 percentage points. Moreover, the proposed method is most effective in the recognition of Chinese herbal medicine names and disease names with significant radicals, with the corresponding F1 scores respectively being improved by 3.48 and 0.97 percentage points, as compared with the SiKuBERT-CRF model based on word features. In general, the performance index of the proposed method is higher than other mainstream deep learning models and possesses good generalization ability.

    SUN Digang, ZHANG Ping
    2024, 52(6):  138-147.  doi:10.12141/j.issn.1000-565X.230420
    Asbtract ( 102 )   HTML ( 4)   PDF (2590KB) ( 23 )  
    Figures and Tables | References | Related Articles | Metrics

    Due to the hand self-occlusion and the lack of depth information, the estimation of 3D hand pose based on monocular RGB images is not accurate enough in estimating relative depth of joints, and the generated hand pose violates the biomechanical constraints of the hand. To solve this problem, by combining the prior knowledge contained in the hand structure and the hand grid information, a deep neural network based on prior knowledge and mesh supervision is proposed. The articulated structure of the hand skeleton implies that there exists a specific relationship between the projections of the 3D hand pose in the 2D image plane and the depth direction, but the differences in hand structure between individuals make it difficult to describe this relationship intuitively and formally. Therefore, this paper proposes to fit it through learning. Specific relationships also exist between joint positions and bone lengths of the same finger, bending directions of different segments of the same finger, and bending directions of different fingers, which are designed as loss functions to supervise network training. The proposed neural network generates hand meshes in parallel with hand poses, supervises the network training through mesh annotation, and optimizes the pose estimation without increasing the network complexity. Furthermore, the neural network is trained using a mixed dataset to further improve its generalization capability. Experimental results show that the proposed method outperforms other methods in terms of internal cross-validation accuracy in multiple datasets, cross-dataset validation accuracy, and time and space complexity of the model. As a result, the prior knowledge of hand skeleton and the mesh supervision improve the accuracy of pose estimation while keeping the neural network compact.

    LIU Hao, YUAN Hui, CHEN Chen, GAO Wei
    2024, 52(6):  148-156.  doi:10.12141/j.issn.1000-565X.230188
    Asbtract ( 109 )   HTML ( 4)   PDF (3565KB) ( 32 )  
    Figures and Tables | References | Related Articles | Metrics

    With the development of science and technology, the accuracy of 3D point cloud acquisition equipment has been continuously improved, and the acquisition of massive 3D point cloud data has become a reality. However, the irregular distribution and huge number of data points of 3D point cloud bring great challenges to data storage and transmission. Therefore, 3D point cloud coding is imperative. From the perspective of data sampling, this paper transforms the 3D point cloud coding problem into a 3D point cloud sampling-reconstruction problem, and proposes a sampling-based 3D point cloud geometry coding framework. In this framework, firstly, the down-sampling method is used to sample the original 3D point cloud to the sparse 3D point cloud with a specified number of points. Then, the sparse 3D point cloud is encoded using any existing coding methods (the number of encoding points is significantly reduced, which can effectively reduce the encoding rate). Finally, by using the proposed upsampling method, the decoded sparse 3D point cloud is interpolated as a high-quality dense 3D point cloud similar to the shape of the original input point cloud. Experimental results show that, as compared with the latest G-PCC provided by MPEG, the proposed 3D point cloud geometry coding framework improves the objective quality of the reconstructed 3D point cloud by 5.49 dB on average, and presents better subjective visual effect.

News
 
Featured Article
Most Read
Most Download
Most Cited