Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-06-06 Cooperative journals: 《计算机应用研究》
Abstract： To achieve high efficiency with limitation of energy consumption is a key problem in the system of UAVs. In existing UAV task allocation methods, the relevance between tasks and UAV cruise directions is neglected, which may further influence the energy consumption and delay of task accomplishment. In view of this, this paper proposed an UAV task allocation method based on the relevance analysis between tasks and cruise directions. This method includes two phases: task screening phase and conflict resolution phase based on consensus. In the first phase, this method selects out tasks without turning back for an UAV according to angles between the cruise direction of the UAV and directions of these tasks, then designs an algorithm to further select candidate tasks before interaction from the tasks without turning back according to their energy consumption utility parameters and time urgency parameters. In the second phase, this method solves tasks conflicts between UAVs after exchanging of their candidate tasks according to energy consumption utility parameters and time delay parameters of these tasks in different cruise directions of these UAVs. Simulation results verify that the proposed method can achieve lower average task energy consumption and average task delay.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-05-10 Cooperative journals: 《计算机应用研究》
Abstract： The drawbacks of existing task recommendation in mobile crowd sensing were as follows: on the one hand, not fully considering the influence of spatial-temporal information on worker preference led to low accuracy of recommendation; On the other hand, ignoring the impact of task popularity on recommendation led to poor recommendation coverage. To solve these drawbacks, this paper proposed a novel task recommendation approach based on spatial-temporal information and task popularity analysis in mobile crowd sensing. Firstly, this approach made full use of the relevant information contained in the worker execution record (e. g. , the time and location of worker performing tasks) to accurately predict the preference of worker for performing tasks. Secondly, in order to reduce the impact of popular tasks on recommendation coverage, this paper analyzed task popularity based on worker reputation and task execution record, and designed appropriate task popularity penalty factor. Then, combining worker preference and task popularity penalty factor, this paper provided an appropriate task recommendation list for each worker. Finally, the experimental results show that compared with the existing baseline methods, the proposed method improves the recommendation accuracy by 3.5% and the recommendation coverage by 25%.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-04-07 Cooperative journals: 《计算机应用研究》
Abstract： Due to the overlap of time and space coverage between tasks in mobile crowd sensing, repeated data collection may happen and cause data redundancy problem. In view of this, this paper proposed a task allocation method to reduce data redundancy within and between tasks. Firstly, this method designed a trajectory sequence prediction model based on the long short-term memory (LSTM) neural network, to predict trajectory sequences of task participants within subdivided spatial-temporal units. Then based on the trajectory prediction results, the method proposed an optimization model to minimize data redundancy. Specifically, the optimization model constrained the data redundancy within a single task by minimizing the data redundancy metric in each spatial-temporal unit, and limited the data redundancy between multiple tasks by maximizing the reuse of the sensing data of each task participant in a spatial-temporal unit. Experimental results verify that the proposed task allocation method can effectively reduce the data redundancy within and between tasks.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-04-07 Cooperative journals: 《计算机应用研究》
Abstract： In the Open Source Community, it’s essential to find and recommend suitable participants for newly initiated issues in order to solved the issues and develop the community. This paper proposed to construct a two-layer Graph Attention Network Participant Recommendation Model (GAT-UCG) based on the cooperative relationship records and the historical participated issues records of the developers. The method used to construct the model is obtaining the information of the problem participants and the interaction information of the developers first, and building the developer problem participation graph and the developer collaboration relationship graph respectively, then redistributing the weights to the edges through the attention mechanism, finally, figuring the Top-N recommendation of the problem participants according to the Issue node embedding representation obtained by the output layer. There are 7, 352 issues from popular Github repositories for experiments. The results showed that the GAT-UCG model outperforms the baseline method in three indicators: recommendation accuracy, recall, and F-Score.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2022-04-07 Cooperative journals: 《计算机应用研究》
Abstract： Most existing text classification algorithms based on graph neural network ignore the problem of over-smoothing, and ignore the problem of information loss due to graph topology, resulting in poor classification performance. To solve this problem, this paper proposed a method to measure the smoothness of multiple text graph representations WACD and a regularization term RWACD to suppress over-smoothing. Subsequently, this paper proposed an attention and residual-based network structure ARS to compensate for the loss of textual information due to graph topology differences. Finally, this paper proposed a graph convolutional neural network text classification algorithm RA-GCN. RA-GCN used ARS to fuse text representations in the graph representation learning layer, and used RWACD in the readout layer to suppress over-smoothing. This paper conducted experiments on 6 Chinese and English datasets. The experimental results demonstrate the classification performance of RA-GCN, and the effects of RWACD and ARS are verified through multiple comparative experiments.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2020-09-28 Cooperative journals: 《计算机应用研究》
Abstract： In order to solve the problems of uneven population distribution and different target scales affecting the crowd counting in public places, this paper proposed a novel crowd counting model based on image field division. Firstly, it divided the image scene into two parts: the near and far field of vision area. For the near field of vision area, it used the YOLO based network for pedestrian detection and added scene constraints to avoid repeated counting in the near and far field of vision. For the far field of vision area, it used the improved MobileNets to extract the population density distribution characteristics, and introduced the super-resolution reconstruction module to improve the quality of the population density map. Finally, it obtained the population in the whole image by calculating the sum of the two. This paper tested the proposed model on Shanghai Tech and Mall datasets, and the results show that the model has a significant improvement in accuracy and robustness. Experiments show that the model is feasible.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-05-10 Cooperative journals: 《计算机应用研究》
Abstract： The current truth discovery method cannot solve the case where the object is composed of many single-valued attributes and multi-valued attributes. Separate processing of these attributes will destroy the original association between attributes, resulting in inaccurate results. This paper proposed an associated truth discovery method for multi-attribute data (ATD4MA). It modeled the observation values of the object using the chromosomes in the genetic algorithm. Then it improved the population initialization algorithm and the basic action of the chromosome according to the problem characteristics. By controlling the evolution behavior of chromosomes, it established the optimization model to minimize the weighted sum of difference between the truth-value chromosomes of each object and the observed values provided by each data source. Therefore it solved the problem of truth discovery where the object contains multiple attributes. Experiments on two real data sets show the correctness and effectiveness of the proposed method.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-04-01 Cooperative journals: 《计算机应用研究》
Abstract： This paper proposed an uncalibrated 3D robotic arm grabbing method based on improved YOLOv2 in a multi-object environment. Firstly, in order to reduce the depth learning algorithm YOLOv2 detection multi-object bounding box overlapping rate and 3D distance calculation error. It proposed an improved algorithm for YOLOv2. Using this algorithm to detect and identify the target object in the image, obtain the position information of the target object in the RGB image, and then use the k-means++ clustering algorithm to quickly calculate the distance from the target object to the camera according to the depth image information, and estimate the target object size and pose. Simultaneously, use the improved YOLOv2 to get the bounding box of the gripper and calculate the distance from the robot to the target object. Then the system estimates the distance between the fixture, camera and object in the manipulator coordinate system. Finally, the system uses the PID algorithm to control the gripper to grab the object according to the size and posture of the object and the distance from the object to the gripper. In this paper, the detected boundary boxes of the target object is more accurate based on the improved YOLOv2 than on old one. It also enhances the distance from the fixture to the object and the size of the object as well as the accuracy of the pose estimation. In addition, in order to avoid complicated calibration, this paper proposes a non-calibration method. This learning scheme is different from the traditional uncalibrated estimation method based on Jacobian matrix, because it has good universality. A simulation experiment shows that the proposed method can accurately classify and locate the objects in the image, The Universal Robot 3 robotic arm uses this framework to verify the effectiveness of capturing objects in a cluttered environment.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-03-13 Cooperative journals: 《计算机应用研究》
Abstract：为了解决传统抽象式摘要模型生成的中文摘要难以保存原文本语义信息的问题，提出了一种融合语言特征的抽象式中文摘要模型。模型中添加了拼接层，将词性、命名实体、词汇位置、TF-IDF等特征拼接到词向量上，使输入模型的词向量包含更多的维度的语义信息来确定关键实体。结合指针机制有选择地复制原文中的关键词到摘要中，从而提高生成的摘要的语义相关性。使用LCSTS新闻数据集进行实验，取得了高于基线模型的ROUGE得分。分析表明本模型能够生成语义相关度较高的中文摘要。
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2019-01-03 Cooperative journals: 《计算机应用研究》
Abstract： Answer acquisition to Machine Reading Comprehension focuses on problem selection or abstract interpretation of the content of the article, but the sequence obtained is prone to problems of inaccurate representation and redundant information. A sequence generation model SGN is proposed for answer acquisition in the machine reading comprehension task. First, the SGN obtains the matching expression between problem and article in problem matrix space, and refers to the potential problem information to generate the word vector of the current node. Then, using a selection gate structure to select the current vocabulary from the article or dictionary, and spontaneously learns and generates OOV (Out- Of-Vocabulary) word to solve the problem of inaccurate semantic representation. Finally, use improved Coverage Mechanism to eliminates redundancies in the generated sequence and improve readability. The experiments adopt the artificial data set SQuAD. The results show that the target sequence generated by SGN is more readable than the benchmark model Seq2Seq and is closer to the original semantics.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-12-13 Cooperative journals: 《计算机应用研究》
Abstract： To reduce the training time for deep Q network, the paper researched on two classical control problems, i. e. Cart Pole and Mountain Car on Open AI Gym, by a DQN method combined with prioritized experience replay scheme and the dueling architecture (dueling DQN-PR) . The prioritized experience replay was rank-based and a deep neural network was adopted in the dueling architecture. The simulation results showed that compared with regular DQN, DQN with dueling network and DQN with prioritized experience replay, dueling DQN-PR acquired better learning performance with least training time. Meanwhile, the impacts of parameters on dueling DQN-PR were analyzed in detail, which provides valuable reference for the practical application.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-11-29 Cooperative journals: 《计算机应用研究》
Abstract： This paper proposed FQMAC (fair and QoS guaranteed MAC) protocol used in the scenario of heterogeneous cognitive radio networks coexisting. Time was divided into beacon periods using beacon frame synchronization. During one beacon period, channel sensing, channel negotiation, and data transmission were performed respectively. The channels occupied by secondary users of other heterogeneous networks were treated as useful channels too. Useful channels were classified into different levels according to their quality, and secondary users were classified into different levels according to their service characteristics too. High level users prioritize high level channels. The reasonable duration of the channel negotiation phase was obtained by establishing a Markov chain model. Simulation results show that FQMAC can improve the throughput of the network, guarantee the users' QoS (Quality of Service) , and achieve better fairness.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-11-29 Cooperative journals: 《计算机应用研究》
Abstract： In order to improve the bandwidth utilization, this paper is based on slot multiplexing. It encoded the transmitted signal into a message frame, and transformed the problem into a constrained packing problem. It encapsulated the signal with multiple relationships into a message frame and solved it by BFD algorithm. Based on this, it put forward a method that can maximize the bandwidth and minimize the number of frame ID. Finally, the verification object is the FlexRay chassis synthesis control and safety system. Experimentally, the proposed algorithm increases the bandwidth utilization by 18.7%, reduces the proportion of FIDs in each communication cycle by 90.47%, and increases the static time slot utilization in the cycle by 41.52%.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-09-12 Cooperative journals: 《计算机应用研究》
Abstract： Visual odometry (VO) estimates the pose of a mobile robot by analyzing the image flow captured by the equipped cameras. In order to analyze the development of VO algorithms, this paper reviewed the related technologies of VO and the up-to-date research state combined with some advanced VO systems. Firstly, this paper described the concept and the evolution of VO, and introduced the mathematical description and the classification of VO. Then, it analyzed the key technologies of VO in details, including feature selection, motion estimation and drift reduction. In addition, it also introduced the latest deep learning based VO. At last, it discussed the existing problems and prospects the development trend of VO.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-06-19 Cooperative journals: 《计算机应用研究》
Abstract： This paper presented a dynamic interactive network (DIN) for recognizing textual entailment. Unlike the other interactive models, DIN facilitates the interaction by projecting the embedding vectors into a two-dimensional matrix space, and then uses the output matrices to produce dynamic weights for the GRU encoder that both processes the context information and controls the information flow. It empowers the extraction of logic segments through higher-orders of information interactions and helps the encoder better choose between the context and the interactive information. Experiments on the SNLI corpus show that our model achieves a test accuracy of 88.0%, outperforming the state-of-the-art with only a small amount of the training parameters introduced.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract： The lightweight block ciphers can be widely used in various applications, such as smart cities, internet of things and cloud computation and so on, in order to protect data and information secure. Midori is a lightweight block cipher proposed in ASIACRYPT 2015. Its block size has two scenarios, i. e, 64 bits and 128 bit, denoted by Midori64 and Midori128 respectively. Up to now, there are no results about meet-in-the-middle attacks on Midori128. This paper developed a meet-in-the-middle attack on 10-round Midori128 for the first time. Specifically, studying the basic construction and key schedule of Midori128, this paper constructed a 7-round distinguisher on Midori128 by using the differential enumeration and key-dependent sieve techniques. Through appending one round at its top and two rounds at its bottom, this paper mounted a meet-in-the-middle attack on 10-round Midori128. In the attack, time-memory tradeoff technique and some weak subkeys were considered so as to reduce the time complexity of online phase. Finally, the data, time and memory complexities of our attack are 2125 chosen plaintexts, 2126.5 10-round encryptions and 2105 128-bit blocks, respectively.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》
Abstract： Traditional clustering algorithms often based on a single distance metric, and how to integrate multivariate metrics is a key challenge in clustering algorithms. This paper proposes a multiobjective Evolutionary Multiple Distance Measure Clustering (MOMDC) based on multi-objective evolutionary algorithm. In this paper, using the Euclidean distance and Path distance to design the actual framework. Firstly, The framework uses the two distance measures to preprocess the classes, and then combining the prepolymerization results to reduce the size of the problem. Secondly, using the multi-objective evolutionary algorithm to cluster in two distance spaces in parallel. In the design of multi - objective evolutionary algorithm, chromosomes using real - tag coding, and two fitness functions based on two distance measures are designed to evaluate the chromosomes. Finally, MOMDC will compare to several other classic algorithms in the data set . Experiments show that the framework can achieve good results for different distributed data sets.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-18 Cooperative journals: 《计算机应用研究》
Abstract： For parameter tuning of fractional order #1;controller, this paper proposed the artificial fish swarm algorithm combined with the image method to tune the parameter. The paper used the typical first-order and second-order system to represent the typical speed servo system and used this model to set the parameters of the fractional order #1; controller for the controlled object. Firstly, in the frequency domain, according to the relative stability of the system and the robustness of the gain, the paper derived the equation. After that, using the image method , it can solve the parameters of the fractional order #1; controller. Around the parameter, according to artificial fish swarm algorithm, it can optimize these parameters. Finally, simulation results show that the controller optimized have better dynamic characteristic than controller obtained by the image method and meet the conditions of robust gains.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-24 Cooperative journals: 《计算机应用研究》
Abstract： The model with integral unit is one process object in actual industry especially in chemical engineering and thermal power plant. The open loop experiment could not be done because of systems’ security and economy. Also the close-loop identification method which is mostly commonly used to identify the object is by using relay feedback, due to this method needs several relay feedback tests and only for fixed structure low order models, so it couldn’t meet the actual industrial requirements. According to this circumstance, a new technique of modeling identification for process with integral unit model has been raised. It can find a main frequency band of the object quickly and analyze frequency response. Then the model with integral unit can be identified accurately. The results of simulation show that this method could acquire a great identification effect.
Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-19 Cooperative journals: 《计算机应用研究》
Abstract： Dental hard tissue registration is very challenging due to the complexity of histological structure, the variability of imaging modality and image quality. To benefit from the advantages of the relative works, we proposed a novel dental hard tissue auto-registration method based on multi-modality dental images. It employed the modified ICP algorithm and was especially effective for dental fluorescence imaging and the reflectance imaging. The algorithm firstly pre-processed the dental images to decrease the impact of imbalance local illumination. Secondly, we investigated a robust feature points extraction strategy, i. e. to calculate the boundaries of hard tissue as initial feature point sets, and to further extract the lesion areas based on the prior knowledge of oral histopathology for the refinement of feature points. Finally, our method executed the modified ICP algorithm on the basis of the special designed point-pairs rejection method and the improved registration strategy. The experimental results show that the proposed method is not only robust, but also can converge more quickly and achieve higher accuracy.
Hits
Hits
Downloads
Comment