Chenqiang Gao

Infrared Patch-Image Model for Small Target Detection in a Single Image

The robust detection of small targets is one of the key techniques in infrared search and tracking applications. A novel small target detection method in a single infrared image is proposed in this paper. Initially, the traditional infrared image model is generalized to a new infrared patch-image model using local patch construction. Then, because of the non-local self-correlation property of the infrared background image, based on the new model small target detection is formulated as an optimization problem of recovering low-rank and sparse matrices, which is effectively solved using stable principle component pursuit. Finally, a simple adaptive segmentation method is used to segment the target image and the segmentation result can be refined by post-processing. Extensive synthetic and real data experiments show that under different clutter backgrounds the proposed method not only works more stably for different target sizes and signal-to-clutter ratio values, but also has better detection performance compared with conventional baseline methods.

Decomposable Nonlocal Tensor Dictionary Learning for Multispectral Image Denoising

As compared to the conventional RGB or gray-scale images, multispectral images (MSI) can deliver more faithful representation for real scenes, and enhance the performance of many computer vision tasks. In practice, however, an MSI is always corrupted by various noises. In this paper we propose an effective MSI denoising approach by combinatorially considering two intrinsic characteristics underlying an MSI: the nonlocal similarity over space and the global correlation across spectrum. In specific, by explicitly considering spatial self-similarity of an MSI we construct a nonlocal tensor dictionary learning model with a group-block-sparsity constraint, which makes similar full-band patches (FBP) share the same atoms from the spatial and spectral dictionaries. Furthermore, through exploiting spectral correlation of an MSI and assuming over-redundancy of dictionaries, the constrained nonlocal MSI dictionary learning model can be decomposed into a series of unconstrained low-rank tensor approximation problems, which can be readily solved by off-the-shelf higher order statistics. Experimental results show that our method outperforms all state-of-the-art MSI denoising methods under comprehensive quantitative performance measures.

GLocal tells you more: Coupling GLocal structural for feature selection with sparsity for image and video classification

Abstract The selection of discriminative features is an important and effective technique for many computer vision and multimedia tasks. Using irrelevant features in classification or clustering tasks could deteriorate the performance. Thus, designing efficient feature selection algorithms to remove the irrelevant features is a possible way to improve the classification or clustering performance. With the successful usage of sparse models in image and video classification and understanding, imposing structural sparsity in feature selection has been widely investigated during the past years. Motivated by the merit of sparse models, in this paper we propose a novel feature selection method using a sparse model. Different from the state of the art, our method is built upon l 2 , p -norm and simultaneously considers both the global and local (GLocal) structures of data distribution. Our method is more flexible in selecting the discriminating features as it is able to control the degree of sparseness. Moreover, considering both global and local structures of data distribution makes our feature selection process more effective. An efficient algorithm is proposed to solve the l 2 , p -norm joint sparsity optimization problem in this paper. Experimental results performed on real-world image and video datasets show the effectiveness of our feature selection method compared to several state-of-the-art methods.

Small infrared target detection using sparse ring representation

A new approach, SRR, for small infrared target detection in single image is described herein. SRR is based on local self-similarity descriptor, according to the characteristics of small infrared targets. Using the SRR, the certainty of small targets can be easily measured. Compared to some conventional cutter removal methods, extensive experimental results show that the proposed approach outperforms these methods. Additionally, the proposed approach has a simple structure, which is convenient for implementation. Plans for future research include incorporating efficiently temporal information to further improve detection performance.

Image Classification by Cross-Media Active Learning With Privileged Information

In this paper, we propose a novel cross-media active learning algorithm to reduce the effort on labeling images for training. The Internet images are often associated with rich textual descriptions. Even though such textual information is not available in test images, it is still useful for learning robust classifiers. In light of this, we apply the recently proposed supervised learning paradigm, learning using privileged information, to the active learning task. Specifically, we train classifiers on both visual features and privileged information, and measure the uncertainty of unlabeled data by exploiting the learned classifiers and slacking function. Then, we propose to select unlabeled samples by jointly measuring the cross-media uncertainty and the visual diversity. Our method automatically learns the optimal tradeoff parameter between the two measurements, which in turn makes our algorithms particularly suitable for real-world applications. Extensive experiments demonstrate the effectiveness of our approach.

Interactive Surveillance Event Detection through Mid-level Discriminative Representation

Event detection from real surveillance videos with complicated background environment is always a very hard task. Different from the traditional retrospective and interactive systems designed on this task, which are mainly executed on video fragments located within the event-occurrence time, in this paper we propose a new interactive system constructed on the mid-level discriminative representations (patches/shots) which are closely related to the event (might occur beyond the event-occurrence period) and are easier to be detected than video fragments. By virtue of such easily-distinguished mid-level patterns, our framework realizes an effective labor division between computers and human participants. The task of computers is to train classifiers on a bunch of mid-level discriminative representations, and to sort all the possible mid-level representations in the evaluation sets based on the classifier scores. The task of human participants is then to readily search the events based on the clues offered by these sorted mid-level representations. For computers, such mid-level representations, with more concise and consistent patterns, can be more accurately detected than video fragments utilized in the conventional framework, and on the other hand, a human participant can always much more easily search the events of interest implicated by these location-anchored mid-level representations than conventional video fragments containing entire scenes. Both of these two properties facilitate the availability of our framework in real surveillance event detection applications.

Small infrared target detection based on kernel principal component analysis

Small infrared target is very difficult to detect due to its own characteristics and complex background. In this paper, we present a small target detection method based on kernel principal component analysis (KPCA). First of all, small target samples are generated by using Gaussian intensity functions. Then a linear PCA is performed in feature space after the small target samples are mapped to a high-dimensional feature space via a nonlinear kernel function, and then the target-enhanced image is obtained by computing the distances between the projection vectors of the training samples and the projection vectors of the each block of the detecting images. Finally, the small infrared target is detected by segmenting the target-enhanced image adaptively. We choose some representative infrared images to evaluate the proposed method, and the experiment results show that the algorithm can detect the small infrared targets effectively.

A block coordinate descent approach for sparse principal component analysis

There are mainly two methodologies utilized in current sparse PCA calculation, the greedy approach and the block approach. While the greedy approach tends to be incrementally invalidated in sequentially generating sparse PCs due to the cumulation of computational errors, the block approach is difficult to elaborately rectify individual sparse PCs under certain practical sparsity or nonnegative constraints. In this paper, a simple while effective block coordinate descent (BCD) method is proposed for solving the sparse PCA problem. The main idea is to separate the original sparse PCA problem into a series of simple sub-problems, each having a closed-form solution. By cyclically solving these sub-problems in an analytical way, the BCD algorithm can be easily constructed. Despite its simplicity, the proposed method performs surprisingly well in extensive experiments implemented on a series of synthetic and real data. In specific, as compared to the greedy approach, the proposed method can iteratively ameliorate the deviation errors of all computed sparse PCs and avoid the problem of accumulating errors; as compared to the block approach, the proposed method can easily handle the constraints imposed on each individual sparse PC, such as certain sparsity and/or nonnegativity constraints. Besides, the proposed method converges to a stationary point of the problem, and its computational complexity is approximately linear in both data size and dimensionality, which makes it well suited to handle large-scale problems of sparse PCA.

Large Disparity Motion Layer Extraction via Topological Clustering

In this paper, we present a robust and efficient approach to extract motion layers from a pair of images with large disparity motion. First, motion models are established as: 1) initial SIFT matches are obtained and grouped into a set of clusters using our developed topological clustering algorithm; 2) for each cluster with no less than three matches, an affine transformation is estimated with least-square solution as tentative motion model; and 3) the tentative motion models are refined and the invalid models are pruned. Then, with the obtained motion models, a graph cuts based layer assignment algorithm is employed to segment the scene into several motion layers. Experimental results demonstrate that our method can successfully segment scenes containing objects with large interframe motion or even with significant interframe scale and pose changes. Furthermore, compared with the previous method invented by Wills and its modified version, our method is much faster and more robust.

Fisher-LDA-Based Infrared Small Target Detection in Wavelet Domain

A novel method for fusion detection of infrared small target based on fisher linear discriminant analysis in wavelet domain is presented in this paper. The proposed method consists of two processes. In the first process: a fisher linear discriminant vector is firstly obtained through fisher linear discriminant analysis model based on target and background samples. And then the vector is converted to a linear filter. In the second process: First, every frame of image sequence is decomposed by the discrete wavelet frame. Second, the approximation with level 2 is filtered by the linear filter obtained in the first process. Third, filtered images of three consecutive frames are fused to accumulate the energy of target of interest and greatly reduce false alarms. Finally the segmentation method based on image complexity is utilized to extract the small target. Real infrared image sequences under complex sea and sky background are applied to validate the proposed method. Experimental results show that the proposed approach is efficient and robust.