Entity

Time filter

Source Type


Wu J.,Key Laboratory of Intelligent Perception and Image Understanding | Qi F.,Xidian University | Shi G.,Xidian University
APSIPA ASC 2011 - Asia-Pacific Signal and Information Processing Association Annual Summit and Conference 2011 | Year: 2011

In this paper, we introduce a unified spatial masking function for the estimation of just-noticeable difference (JND). Conventional models estimate several parts independently, and then combine these parts to get the JND. In this work, we treat the spatial masking effect as a nonlinear transformation of the luminance adaptation. To model the transformation, we measure the deviation of image contents from the ideal patterns to establish luminance adaptation rules. Considering both luminance difference and structural regularity, we derive a nonlinear spatial masking function by modulating luminance adaptation with the deviation coefficients. The masking function deduces an accurate estimation of the JND. Experiments demonstrate the validity of the proposed framework.


Pan X.-Y.,Xian University of Science and Technology | Pan X.-Y.,Xidian University | Pan X.-Y.,Key Laboratory of Intelligent Perception and Image Understanding | Liu F.,Xidian University | And 2 more authors.
Ruan Jian Xue Bao/Journal of Software | Year: 2010

By using the density sensitive distance as the similarity measurement, an algorithm of Density Sensitive based Multi-Agent Evolutionary Clustering (DSMAEC), based on multi-agent evolution, is proposed in this paper. DSMAEC designs a new connection based encoding, and the clustering results can be obtained by the process of decoding directly. It does not require the number of clusters to be known beforehand and overcomes the dependence of the domain knowledge. Aim at solving the clustering problem, three effective evolutionary operators are designed for competition, cooperation, and self-learning of an agent. Some experiments about artificial data, UCI data, and synthetic texture images are tested. These results show that DSMAEC can confirm the number of clusters automatically, tackle the data with different structures, and satisfy the diverse clustering request. © by Institute of Software, the Chinese Academy of Sciences.


Wu J.,China Jiliang University | Wu J.,Xidian University | Wu J.,Key Laboratory of Intelligent Perception and Image Understanding | Liu F.,Xidian University | And 2 more authors.
2012 5th International Congress on Image and Signal Processing, CISP 2012 | Year: 2012

The prior information of image plays an important role in compressive sensing (CS) reconstruction. The edge is one of the important information which exists in the image to be recovered. In this work, the edge of images is used as a prior information and extracted from CS measurements in the wavelet-based CS inversion. An edge-based matching pursuit algorithm is developed by merging edge information to guide the pursuit process. Edge information well defines the locations of the significant coefficients to be recovered, thus it contributes so much to the reconstruction quality. Experiment results demonstrate that the proposed algorithm returns superior reconstruction performance for the images with obvious edges and high sparsity compared with other CS algorithms. © 2012 IEEE.


Hou B.,Key Laboratory of Intelligent Perception and Image Understanding | Cheng X.,Key Laboratory of Intelligent Perception and Image Understanding | Jiang H.Q.,Key Laboratory of Intelligent Perception and Image Understanding
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2012

The compressive sensing theory shows that signals and images can be recovered from far fewer samples than that used in Shannon sampling theorem. In practical applications our aim is that an object should be meaningfully reconstructed in considerable detail relative to the full scene elsewhere at low sampling rate so that the target will be found directly from the reconstructed image. To achieve the goal, a new method based on manifold-based compressive sensing is proposed for object specific image reconstruction, in which the whole image is divided into small pieces and reconstructed piece by piece with the probability density function of the target as prior knowledge. Our reconstruction method is very fast since we divided the whole image into small pieces and the small pieces are reconstructed respectively. In our method, modeling a manifold of a target and getting the probability density function of the target is the key issue. And the model, which is used to obtain the probability density function about the target, is a Mixture of Factor Analyzers (MFA). The experiments results show that the target can be reconstructed more clearly than the full scene elsewhere at the low sampling rate with our method. © 2012 Springer-Verlag.


Yang S.,Key Laboratory of Intelligent Perception and Image Understanding | Han Y.,Key Laboratory of Intelligent Perception and Image Understanding | Zhang X.,Key Laboratory of Intelligent Perception and Image Understanding
Proceedings of the International Joint Conference on Neural Networks | Year: 2012

In this paper, we propose a sparse kernel representation classification algorithm (SKRC) for images classification and recognition. The training dictionary is composed by labeled samples directly, and both training dictionary and testing sample are mapped into feature space from original sample space by the sparse kernel which employs the "center" samples matrix constructed by a method similar to k-means clustering. Then in the feature space, the basic sparse representation based classification method is employed. We test our proposed algorithm on some different public database, and the results show that our proposed method can achieve higher classification accuracy without much time consumed. © 2012 IEEE.

Discover hidden collaborations