Facial expression recognition (FER) is important in vision-related applications. Deep neural networks demonstrate impressive performance for face recognition; however, it should be noted that this method relies heavily on a great deal of manually labeled training data, which is not available for facial expressions in real-world applications. Hence, we propose a powerful facial feature called deep peak–neutral difference (DPND) for FER. DPND is defined as the difference between two deep representations of the fully expressive (peak) and neutral facial expression frames. The difference tends to emphasize the facial parts that are changed in the transition from the neutral to the expressive face and to eliminate the face identity information retained in the fine-tuned deep neural network for facial expression, the network has been trained on large-scale face recognition dataset. Furthermore, unsupervised clustering and semi-supervised classification methods are presented to automatically acquire the neutral and peak frames from the expression sequence. The proposed facial expression feature achieved encouraging results on public databases, which suggests that it has strong potential to recognize facial expressions in real-world applications.
The characteristics of non-linear, low-rank, and feature redundancy often appear in high-dimensional data, which have great trouble for further research. Therefore, a low-rank unsupervised feature selection algorithm based on kernel function is proposed. Firstly, each feature is projected into the high-dimensional kernel space by the kernel function to solve the problem of linear inseparability in the low-dimensional space. At the same time, the self-expression form is introduced into the deviation term and the coefficient matrix is processed with low rank and sparsity. Finally, the sparse regularization factor of the coefficient vector of the kernel matrix is introduced to implement feature selection. In this algorithm, kernel matrix is used to solve linear inseparability, low rank constraints to consider the global information of the data, and self-representation form determines the importance of features. Experiments show that comparing with other algorithms, the classification after feature selection using this algorithm can achieve good results.
This paper addresses flow shop scheduling problem with a batch processor followed by a discrete processor. Incompatible job families and limited buffer size are considered, and the objective is to determine a schedule such that the total completion time is minimised. Flexible buffer service policy is designed, and a greedy heuristic together with the worst-case analysis is developed. We also propose a hybrid method involving a Differential Evolution algorithm. Moreover, two tight lower bounds are provided to measure the performances of the proposed algorithms. Numerical results demonstrate that the proposed algorithms are capable of providing high-quality solutions for large-scale problems within a reasonable computational time. 相似文献
This paper addresses the problem of scheduling on batch and unary machines with incompatible job families such that the total weighted completion time is minimised. A mixed-integer linear programming model is proposed to solve the problem to optimality for small instances. Tight lower bounds and a 4-approximation algorithm are developed. A constraint programming-based method is also proposed. Numerical results demonstrate that the proposed algorithms can obtain high quality solutions and have a competitive performance. Sensitivity analysis indicates that the performance of the proposed algorithms is also robust on different problem structures. 相似文献
The high-end equipment intelligent manufacturing (HEIM) industry is of strategic importance to national and economic security. Engineering management (EM) for HEIM is a complex, innovative process that integrates natural science, technology, management science, social science, and the human spirit. New-generation information technology (IT), including the internet, cloud computing, big data, and artificial intelligence, have made a remarkable influence on HEIM and its engineering management activities, such as product system construction, product life cycle management, manufacturing resources organization, manufacturing model innovation, and reconstruction of the enterprise ecosystem. Engineering management for HEIM is a key topic at the frontier of international academic research. This study systematically reviews the current research on issues pertaining to engineering management for HEIM under the new-generation IT environment. These issues include cross-lifecycle management, network collaboration management, task integration management of innovative development, operation optimization of smart factories, quality and reliability management, information management, and intelligent decision making. The challenges presented by these issues and potential research opportunities are also summarized and discussed. 相似文献
Dynamic programming, branch-and-bound, and constraint programming are the standard solution principles for finding optimal solutions to machine scheduling problems. We propose a new hybrid optimization framework that integrates all three methodologies. The hybrid framework leads to powerful solution procedures. We demonstrate our approach through the optimal solution of the single-machine total weighted completion time scheduling problem subject to release dates, which is known to be strongly NP-hard. Extensive computational experiments indicate that new hybrid algorithms use orders of magnitude less storage than dynamic programming, and yet can still reap the full benefit of the dynamic programming property inherent to the problem. We are able to solve to optimality all 1900 instances with up to 200 jobs. This more than doubles the size of problems that can be solved optimally by the previous best algorithm running on the latest computing hardware. 相似文献
We study the departure times in tandem production lines where the products passing through the lines are either discrete entities or continuous fluid. We call these discrete tandem (DT) or continuous tandem (CT) lines, respectively. We apply sample path analysis techniques to relate the departure time in a DT line to the departure time in a CT line where both lines have equivalent model parameters. We show that the departure time of a quantity q, produced at a machine in a DT line governed by the communication blocking mechanism, converges to the departure time of the same quantity at the same machine in the corresponding CT line as the size of the products in the DT line becomes infinitesimally small. Since continuous fluid models are used in both queueing and control systems to approximate the behavior of discrete systems, this asymptotic result enhances the understanding and the use of such models. Finally, our result also leads to an alternative proof for the convexity of the departure time in CT lines. 相似文献