Search

The explosion of short videos has dramatically reshaped the manners people socialize, yielding a new trend for daily sharing and access to the latest information. These rich video resources, on the one hand, benefited from the popularization of portable devices with cameras, but on the other, they can not be independent of the valuable editing work contributed by numerous video creators. In this paper, we investigate a novel and practical problem, namely audio beat matching (ABM), which aims to recommend the proper transition time stamps based on the background music. This technique helps to ease the labor-intensive work during video editing, saving energy for creators so that they can focus more on the creativity of video content. We formally define the ABM problem and its evaluation protocol. Meanwhile, a large-scale audio dataset, i.e., the AutoMatch with over 87k finely annotated background music, is presented to facilitate this newly opened research direction. To further lay solid foundations for the following study, we also propose a novel model termed BeatX to tackle this challenging task. Alongside, we creatively present the concept of label scope, which eliminates the data imbalance issues and assigns adaptive weights for the ground truth during the training procedure in one stop. Though plentiful short video platforms have flourished for a long time, the relevant research concerning this scenario is not sufficient, and to the best of our knowledge, AutoMatch is the first large-scale dataset to tackle the audio beat matching problem. We hope the released dataset and our competitive baseline can encourage more attention to this line of research. The dataset and codes will be made publicly available., Comment: 11 pages, 5 figures

In the industrial production process, temperature control is often very critical. However, in industrial production temperature control systems, there are often characteristics such as multiple disturbances, large inertia, large delay, and nonlinearity, which will make the system temperature difficult to control quickly and accurately. Based on this, this topic takes the air source heat pump as the research object to study the temperature control of the air source heat pump insulation water tank. According to the established mathematical model, design its generalized predictive controller, fuzzy controller, and digital PID controller, which is based on Matlab. The simulation is performed to compare the results of the three control methods. The simulation results show that the generalized predictive controller has high temperature control accuracy and good real-time performance, which is more suitable for the control of the system.

The mathematical model of the temperature control system of air source heat pump is established by collecting experimental data. Due to the time-varying, nonlinear and pure hysteresis characteristics of the temperature control system, on the basis of fuzzy PID control, MATLAB is used to design a two-dimension adaptive fuzzy PID controller based on particle swarm optimization. By optimizing each parameter of the system, the water flow of the system can be well controlled and the temperature effect of the control heat pump system can be optimized and improved. Finally, the stability verification is carried out, that is, after the system is stable, the pulse interference signal is suddenly added to cause interference to the system Observe the stabilizing effect of the system The simulation results show that the fuzzy PID based on PSO not only combines the advantages of fuzzy control and PID control, but also optimizes the scale factor and quantization factor to reduce randomness. The resulting optimization curve has almost no overshoot, rise and adjustment time is effectively shortened, and response The speed becomes faster, and the robustness and anti-interference performance of the heat pump system are improved.

There are many problems in the air source heat pump system, such as nonlinear, multi-disturbance, accurate model uncertainty and so on. The traditional PID control scheme has been unable to meet its process requirements. As a kind of model predictive control, Generalized Predictive Control adopts prediction model, feedback correction and rolling optimization, etc, which has the characteristics of strong robustness and low model requirements, and can effectively overcome its shortcomings. In this paper, the mathematical model of air source heat pump is established, and the Generalized Predictive Control strategy is proposed. The Generalized Predictive Controller, PID controller and Fuzzy controller are designed on the MTALAB simulation platform. By comparing the experimental results, we can see that Generalized Predictive Control has better robustness, tracking performance and overall performance.

In order to avoid library intelligent returning books to bookshelf robot during the process of searching for books and bookshelves, the tracking time is long and the path deviation is large. The system uses a navigation system based on a fuzzy PID controller based on particle swarm optimization to avoid problems such as empirical parameter selection and parameter invariance in fuzzy PID control. The particle swarm optimization algorithm is used to optimize the controller parameters, enhance the controller's adaptive ability, and improve its control accuracy. Experiments show that the system improves the accuracy of the path, and has a faster response speed and higher tracking accuracy when moving.

The air source heat pump heat collection control system is a system with strong nonlinearity and large time delay. It is difficult to achieve the expected control effect on the outlet water temperature by using conventional control algorithms. In view of this problem, this paper makes up for the shortcomings of fuzzy control and PID control, and proposes an improved particle swarm optimization (IPSO) fuzzy PID control strategy to control the temperature of the effluent of the air source heat pump heat collection control system. First of all, the particle swarm optimization algorithm is easy to fall into the problem of local optimization. By increasing the vector, the optimal position of individual particles is stored to increase the range of particle search. Then use the improved particle swarm algorithm to optimize the fuzzy PID scale factor and quantization factor to modify the PID parameters in real time to improve the control accuracy of the heat pump system. Finally, a MATLAB / Simulink simulation model for the heat collection temperature control of the heat pump is established and an improved particle swarm optimization algorithm program is written, and compared with the fuzzy PID and conventional PID control algorithms. Simulation results show that the heat-collection control system based on IPSO has strong adaptability, robustness and anti-interference.

Aiming at the current call number extraction technology that only deals with the useful area, which is greatly affected by complex color background factors, a method for locating and extracting call numbers from randomly collected images of books on the shelf under a complex color background is designed. According to the color characteristics at the edge of the call number label, the horizontal projection analysis is used to analyze the image features, filter out other color background interference, and locate the call number area. Combine mathematical morphology and iterative global threshold segmentation algorithm to segment and extract the call number area. The experimental results show that the algorithm's Intersection-over-Union (IOU) index reaches 89.91% and the Precision index reaches 96.11%. Compared with the commonly used algorithms, they are respectively increased by 29.94% and 31.08%, which can effectively extract and segment the call number area without being affected by complex color background, with high efficiency and strong anti-interference.

Multi-focus image fusion technologies compress different focus depth images into an image in which most objects are in focus. However, although existing image fusion techniques, including traditional algorithms and deep learning-based algorithms, can generate high-quality fused images, they need multiple images with different focus depths in the same field of view. This criterion may not be met in some cases where time efficiency is required or the hardware is insufficient. The problem is especially prominent in large-size whole slide images. This paper focused on the multi-focus image fusion of cytopathological digital slide images, and proposed a novel method for generating fused images from single-focus or few-focus images based on conditional generative adversarial network (GAN). Through the adversarial learning of the generator and discriminator, the method is capable of generating fused images with clear textures and large depth of field. Combined with the characteristics of cytopathological images, this paper designs a new generator architecture combining U-Net and DenseBlock, which can effectively improve the network's receptive field and comprehensively encode image features. Meanwhile, this paper develops a semantic segmentation network that identifies the blurred regions in cytopathological images. By integrating the network into the generative model, the quality of the generated fused images is effectively improved. Our method can generate fused images from only single-focus or few-focus images, thereby avoiding the problem of collecting multiple images of different focus depths with increased time and hardware costs. Furthermore, our model is designed to learn the direct mapping of input source images to fused images without the need to manually design complex activity level measurements and fusion rules as in traditional methods.

Aiming at the precise positioning problem of cigarette filter box cartoning machine in the process of pushing the production line, a new type of push mechanism control system is designed. The fuzzy control theory is combined with the PID controller, and the PID parameters are adaptively adjusted by fuzzy control. It can improve system response speed and control accuracy. The results show that the improved push mechanism can realize the flexible filter rod packing, and the positioning accuracy is better than 0.1mm, which improves the packing quality of the filter box cartoning machine.