Because of the broad application of human action recognition technology, action recognition has always been a hot spot in computer vision research. The Long Short-Term Memory (LSTM) network is a classic action recognition algorithm, and many effective hybrid algorithms have been proposed based on basic LSTM infrastructure. Although some progress has been made in accuracy, most of those hybrid algorithms have to have more and more complex structures and deeper network levels. After analyzing the structure of the classic LSTM from the perspective of control theory, we determined that the classic LSTM could strengthen the differential characteristics of human action recognition technology to reflect the change of speed. Thus, an improved LSTM structure with an input differential characteristic module is proposed. Furthermore, in this article, we considered the influence of first-order and second-order differential on the extraction of movement pose information, that is, the influence of movement speed and acceleration on action recognition. We designed four different LSTM units with first-order and second-order differential. Moreover, the experiments were performed for the four units on three common datasets repeatedly. We found that the LSTM network with the input differential feature module proposed in this article can effectively improve action recognition accuracy and stability without deepening the complexity of the network and can be used as a new basic LSTM network architecture.
Abstract.A novel on high-grade CNC machines tools for B Spline curve method of High-speed interpolation arithmetic is introduced. In the high-grade CNC machines tools CNC system existed the type value points is more trouble, the control precision is not strong and so on, In order to solve this problem .Through specific examples in matlab7.0 simulation result showed that that the interpolation error significantly reduced, the control precision is improved markedly, and satisfy the real-time interpolation of high speed, high accuracy requirements.
Gecko climbing robot cable is the core part of the robot system, which plays an important role in the basic life support of the operator. Based on the analysis of the function and performance requirements of the safety cable in the climbing process, the modular design scheme of the retractor is proposed according to the system engineering principle and optimization method, and the system is divided into retractor module, guidance and control module and mechanical structure module. Based on the analysis of the characteristics of the operator’s safety state during the climbing operation, the operator’s safety state set, the operator’s safety state set and the operator’s safety state set are formed. The whole process/full state algorithm is proposed. In the structural design, the basic functions of the retractor are divided into retractor module, guidance and control module and mechanical structure module. Considering the limited space and bearing capacity of the robot vehicle, on the premise of ensuring the safety and reliability, the lightweight optimization design is carried out to realize the function module and the double-pawl-bidirectional backstop-double-optimization retraction and release system with the most concise connection of each function module.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.