Search

In this paper, we propose a motion-compensated spatio--temporal Locally adaptive linear minimum mean squared-error (LLMMSE) filter for noisy image sequences with both temporal and spatially adaptive filtering support. Motion compensation and an adaptive temporal filtering support (TFS) structure guarantee the uniformity of the TFS. An intelligent pixel aggregation algorithm is proposed to include homogeneous neighboring pixels and exclude the outlier pixels, resulting in uniform spatial filtering support. By using the proposed spatio--temporal LLMMSE filter with uniform spatio--temporal support, we can reduce noise efficiently without introducing visually disturbing blurring artifacts. Furthermore, we employ an adaptive weighted local mean and variance estimation algorithm to improve the accuracy of estimation. The weights provide an implicit mechanism for deemphasizing the contribution of the outlier pixels which are wrongly kept within the support to avoid blurring. The performance of the proposed filter is quantitatively evaluated and compared with state-of-the-art methods. The results demonstrate that the proposed filter can achieve superior filtering performance, both subjectively and objectively. Index Terms--Locally adaptive linear minimum mean squared-error (LLMMSE), motion-compensated spatio--temporal filter, spatial filtering support (SFS), temporal filtering support (TFS).

Efficient retrieval is one of the crucial services of digital library, and sorting is the core issue of efficient information retrieval. Given a list of candidate book titles, the sorting model is used to generate a sorted list of book titles. When learning-based sorting algorithms are applied in the filed of information retrieval, they often optimize the sorting model by minimizing the value of the pairwise loss function. However, existing analysis shows that minimizing the pairwise loss value does not necessarily lead to the optimal sorting performance of the listwise algorithm. It is also very difficult to combine the online learning sorting algorithm with the listwise algorithm. This paper proposes an online learning sorting algorithm based on listwise, which aims to realize the online learning sorting algorithm under the premise of ensuring the performance advantage of listwise algorithm, thereby reducing the retrieval complexity. Firstly, the problem of combining the online learning sorting algorithm with the listwise algorithm is solved. Secondly, the sorting model is optimized by minimizing the loss function based on the predicted list and the real list. Finally, an adaptive learning rate based on the online-listwise algorithm is proposed. Experimental results show that the proposed algorithm has better retrieval performance and speed. [ABSTRACT FROM AUTHOR]

The paper is about implement tlie " Outstanding Program" for grasper, innovate college-enterprise cooperation base mechanism. It elaborated importance of training base construction in the experimental demonstration center, actively explore and seek coüege-enterprise cooperation ways. Adhere to the industry of training, in order to promote research, to produce a training, to research with production management principles, to build a training base. Making bold and comprehensive innovation and exploration on construction of practice base, teaching models and teaching methods, and finally establishing a new model of professional, production-practice, including project-based practice and innovation practicing environment, realize sharing device resources conditions, university teachers and enterprise engineers. The experiment proves this is the primary problem in the training base construction during to innovation " Outstanding Engineers" as the core training base construction mode, and to find and seize balance of interest point between schools and enterprises. Only in this way, coHege-enterprise cooperation base can play its due role, and is in accordance with development tendency of current national talent training reform. [ABSTRACT FROM AUTHOR]