Search

The traditional collaborative filtering recommendation systems have many deficiencies, which make them incompetent in the domain of clothing recommendation; we proposed a new ClothNet model based on CNN, RNN, collaborative filtering and the characteristics of the fashion industry. The accuracy and generalization performance of this model are improved compared with traditional systems. The visual information integrated into the ClothNet model enables the recommendation system to alleviate the cold start problem, and new clothes can be added to the recommendation list faster through the visual information. The addition of temporal information enables ClothNet sharply capturing the impact of seasonal and time changes on user preferences. However, because RNN and CNN have the disadvantage of requiring a large amount of data, combining RNN and CNN will make the model more difficult to converge, so we have adopted the LearningToRank training mode and obtained good results.

The purpose of this paper is to study the help of generative adversarial networks (GAN) for face generation, and to explore whether the network can have an effect on complex face generation. Training an image translation neural network model based on a generative adversarial network with the help of a large number of real human face data sets. Using the CV2-based face tagging algorithm and the HED-based face edge extraction algorithm to obtain input information, and then based on the translation neural network model Developing a face generation system through Tensorflow, Torch and other frameworks to realize the function of generating real faces through sketches or “changing faces” through existing faces. Finally, this model provides training configuration and training information.

The fast development of urbanization and development have led to an increase in the difficulties with regards to satisfying the demand of public transport. The quantized evaluation analysis of urban public transport performance is useful to identify its shortcomings and to also enable further optimization of the transport system. This facilitates the improvement of the systems quality of service and provides the means to reduce operating costs. Furthermore, the quantification of the bus lines performance evaluation is crucial in raising the systems operational efficiency and improving the operators level of service. In this study, the data collected from Hangzhou Bus Companys operating income table is evaluated to assess bus performance by Excel and Deap software. The study constructs the conceptual framework that evaluates the performance evaluation of the single bus line, and divides the implementation process of the single bus line service into two stages. Building the Output-BCC (Banker Charnes Cooper) model with DEA method to calculate the operational efficiency and service efficiency, then accesses the Hangzhou bus transport objectively and provides some suggestions.

Vector image is a type of image composed of many geometric primitives. Compared with bitmaps, vector images have the ability to save memory as well as to enlarge without distortion. Meanwhile, it has been commonly adopted in data visualization (image data) because it can be scaled to multiple sizes to fit different scenes. For instance, it can be applied for the illustrations in newspapers and magazines, the logo on the web, the background for poster, the design of text, and traffic signs. However, transforming a bitmap to vector image is still a challenging problem because of the complicated content of a bitmap, which tends to consist of more than just simple geometry. Aiming at this issue, there is a new approach proposed to transform from bitmaps to vector images, which is based on triangle units and consists of three steps. In detail, firstly, there is an initial mesh constructed for one image in pixel level after detecting features. Then, the initial mesh will be simplified by collapsing two vertices as the initial mesh is too dense to represent one image. Specifically, there are two main parts in this step, which are collapse conditions and collapse influences. In the step of collapsing, issues such as overlap and sharp triangles can be conquered by a sort-edge method (which will be illustrated specifically later). The final step is to select one color for each triangle, since it is helpful to save the memory and speed up the process of this method. In addition, one color will represent one triangle; hence, in the final step, the four-triangle sample method will be applied in order to prevent a vector image from generating too large color discontinuity. Once the pretest proceeds without mistake, the method above is able to work for the general bitmaps. Note that our method can be applied to information security and privacy, since one image can be encoded to some triangles and colors.

Cellular Automata is a discrete model studied in computability theory, mathematics, physics, complexity science, theoretical biology and microstructure modeling. This paper proposes a model to balance the flux and safety, to maximize the flux on each road whilst ensuring work well with safety based on the Cellular Automata. In order to analyze the flow of traffic, the traffic of three-lane freeway and four-lane freeway are simulated based on the cellular automata model with the factors of velocity and density in the first step. For safety, a rear-end collision model and overtaking lateral collision model were employed to analyze the relation among safety. Reaction time and distance are introduced in the model, and different vehicles are considered. Safety-related factors are also chosen based on back propagation neural network. At last, this paper establishes a model to examine the performance of the traffic flow under the multi-lane freeway in light and heavy traffic. It simulates the situation in real life based on actual data. The optimal relation and optimal value has worked out between flux and safety through simulating the situation in real life based on actual data. To achieve a better model, further discussion, is made on how to adapt to different situations with the keep-left, the unrestricted, and the intelligent system controlled. The experimental result shows that the number of lanes is not an influential factor under any circumstances. The maximum speed limit plays a significant role in light traffic while it is of no importance in heavy traffic. However, the minimum speed limit plays a significant role in heavy traffic while it is of no importance in light traffic.

The pedestrian detection is a hot research topic in computer recognition. It involves not only the pedestrian location information but also the intrusion detection function, which has wide prospects in the application of vehicle traffic, campus monitoring, and building guard. However, the identification accuracy and recognition speed play an important role in the pedestrian detection, which calls for a fast pedestrian detection approach. The general pedestrian detection implementation, based on the integral channel features method and soft cascade classifier, is the popular technique in the current business application since its better speed and accuracy. Thus, this method uses the feature approximation technique and multiple classifiers to achieve the feature computing, which speeds up the detection without resizing image. To this end, this paper is motived to propose a multi-scale handling method for the fast pedestrian detection, using the tactics detection from sparse to dense. Our pedestrian detection method consists of four parts functions, mainly pedestrian statistics and intrusion detection, pedestrian tracking and pedestrian flow statistics. All these modules are introduced with its details about design and implementation. In Addition, the proposed multi-scale handling method can be applied into most of object detectors to improve their recognition speed. In conclusion, our proposed approach has a good potential application prospect in the video surveillance system.

The issue of searching missing aircraft is valuable after the event of MH370. This paper provides a global optimal model to foster the efficiency of maritime search. Firstly, the limited scope, a circle whose center is the last known position of the aircraft, should be estimated based on the historical data recorded before the disappearance of the aircraft. And Bayes' theorem is applied to calculate the probability that the plane falling in the region can be found. Secondly, the drift of aircraft debris under the influence of wind and current is considered via Finite Volume Community Ocean Model(FVCOM) and Monte Carlo Method(MC), which make the theory more reasonable. Finally, a global optimal model about vessel and aircraft quantitative constraints is established, which fully considers factors including the area of sea region to be searched, the maximum speed, search capabilities, initial distance of the vessels by introducing 0-1 decision variables.

State-owned asset information management focuses on the process monitoring and data mining analysis, due to the state-owned enterprise asset management group information on the existence of multiple levels of management, the number of users and more users and more data reliability, high security requirements of the characteristics; single station server model has failed to meet current demand, and there is a physical device utilization is not high. This article focuses on how to build an LVS load-balancing cluster technology, combined with virtualization and shared storage technology to achieve the three-tier architecture of Web server clusters, in order to achieve low-cost investment in expensive hardware, computing and business storage solutions to achieve the bad environment. Keywords-LVS, cluster technology, state-owned assets management, shared memory, Virtualization

In this paper, by using the improved fingerprint pattern, the fingerprint preprocessing effect of using Gabor filtering have been effectively improved. Then, by combining the advantages of the direction-based binarization algorithm and the local threshold algorithm, this paper can effectively get the fingerprint image binarization and remains the topological structures of the image. Finally, the experiments prove that the fingerprint image enhancement and improved binarization algorithm can improve the accuracy of the fingerprint feature extraction. Keywords-Image enhancement, Binarization, Recursive algorithm